Skills Library

Ready-made skills that connect your AI tool or agent to OriginalVoices audience research workflows. Each skill guides the AI through a structured research process using Digital Twins. Compatible with Claude, Manus, and other AI tools that support the Agent Skills standard.

Before you start

These skills require the OriginalVoices MCP server to be connected to your AI tool or agent. If you haven't set that up yet, follow the integration guide first.

To install: Download the .md file, then upload it to your AI agent (e.g. in Claude: Customize > Skills > Click + > Upload a skill).

Brand Messaging Validation

Validate your brand messaging with real audience insight before locking it in. This workflow helps you understand how your target audience interprets your brand statements, what creates emotional resonance, what causes confusion, and what differentiates you meaningfully — so you can build brand messaging that actually connects instead of sounding like every other company.

Use when: Requests to test brand positioning, validate taglines or slogans, understand brand perception, refine mission statements, test brand voice, or differentiate from competitors.

Download skill

View full skill

---
name: brand-messaging-validation
version: 1.1.0
description: "Use this skill to validate and refine brand messaging, positioning, and taglines using real audience insight. Triggers include: requests to test brand positioning, validate taglines or slogans, understand brand perception, refine mission statements, test brand voice, or differentiate from competitors. Uses OriginalVoices Digital Twins (ask_twins) to understand how target audiences interpret brand messaging, what resonates emotionally, what creates confusion, and what drives brand affinity — ensuring brand statements connect with real people, not just internal stakeholders."
---

# Brand Messaging Validation Skill

## Overview

Validate your brand messaging with real audience insight before locking it in. This workflow helps you understand how your target audience interprets your brand statements, what creates emotional resonance, what causes confusion, and what differentiates you meaningfully — so you can build brand messaging that actually connects instead of sounding like every other company.

## Workflow Steps

### Step 1: Define What You're Testing

Gather from the user:
- **Target audience**: Who is this brand for? (e.g. "Health-conscious millennials", "Small business owners in creative industries")
- **Current or proposed brand messaging**:
  - Brand positioning statement
  - Tagline/slogan (if applicable)
  - Mission/vision statements (if testing)
  - Key brand values or pillars
  - Tone of voice examples (if testing)
- **Category context**: What space does the brand operate in? Who are main competitors?

### Step 2: Comprehensive Brand Messaging Validation

One comprehensive research call covering category values, positioning interpretation, tagline resonance, brand values authenticity, differentiation, and emotional connection.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Testing stimuli without bias.** When showing concepts, ads, taglines, or messaging options:
- Capture the **unaided gut reaction first** ("what's your first impression?") before asking for reasons.
- **Don't signal the hoped-for winner** — never reveal which option is the client's, and never ask "isn't this great?".
- When comparing options, **counterbalance the order** (don't always show A first) and always allow "neither / none of them".
- **Keep three questions separate:** comprehension ("what message do you take from this? who is it for?") → preference ("which do you prefer, and why?") → persuasion ("does it make you more or less likely to act?"). Never fuse them into one.

```
ask_twins(
  audience: "[detailed target audience]",
  questions: [
    "Tell me about the last time you picked a [product category / industry] brand and felt good about the choice — what won you over?",
    "Thinking about the [category] brands you've dealt with — what have some got right, and what have others got wrong?",
    "Here's how a brand describes itself: '[positioning statement]'. What's your honest first impression?",
    "In your own words, what does this brand actually do, and who is it for? What's fuzzy or unclear, if anything?",
    "Set next to other [category] brands you know, how does this one come across — where does it stand out, and where does it blend in?",
    "[If testing tagline] When you read this tagline: '[tagline]', what does it make you think or feel?",
    "This brand says it stands for: [list 3-4 brand values]. Which of these would actually change how you see a [category] brand — and which are just table stakes?",
    "For a [category] brand, which of those claims would you take at face value — and which would you take with a pinch of salt? What makes the difference?",
    "[If testing tone — show the two examples in a random order, and allow 'neither'] Here are two ways this brand might talk: '[Example 1]' and '[Example 2]'. Which sounds more like a brand you'd trust — or does neither — and what is it about the wording?",
    "Set against [competitor name] or other [category] brands you know, where does this one feel genuinely different, and where does it feel like more of the same?",
    "When you read this brand's messaging, what feeling does it leave you with — and how close does it come to understanding people like you?",
    "Tell me about a brand you've happily recommended to someone — what earned that? Where would this one land against it?",
    "If you described this brand to a friend in one sentence, what would you say — and what would stick with you afterwards?",
    "[If testing alternatives — show the options in a random order, and allow 'neither'] Here are two ways this brand could position itself: '[Option 1]' and '[Option 2]'. Which pulls you in more — or does neither — and what's doing the work?"
  ]
)
```

**Note:** Adjust questions based on what's being tested. If not testing a tagline, remove tagline questions. If not testing tone, remove tone questions. If testing alternatives, include comparison questions. Whenever you show two options (tone or positioning), vary which one appears first across runs and always let respondents choose "neither".

### Step 3: Analyse Brand Messaging Insights

Review responses and assess:

**Clarity:**
- Do people understand what the brand does and who it's for?
- What creates confusion or ambiguity?

**Resonance:**
- What messaging landed emotionally?
- What values or statements people connected with?
- What felt hollow or generic?

**Differentiation:**
- Does the brand feel distinct from competitors?
- What makes it stand out (or fail to stand out)?
- Is differentiation meaningful or superficial?

**Authenticity:**
- Does the messaging feel genuine or corporate?
- Do stated values feel believable?
- Does tone match expectations for the category?

**Memorability:**
- What stuck with people?
- What would they remember or repeat?

### Step 4: Deliver Brand Messaging Report

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers. Where findings have clear quantitative patterns (e.g. "8 out of 12 twins found the positioning unclear"), include a chart to visualize the distribution.

### Report sections:

**1. Executive Summary:**
- Overall verdict on current messaging
- Top 3 strengths
- Top 3 areas for improvement
- Recommended direction (Keep / Refine / Rethink)

**2. Positioning Assessment:**
- How audience interprets the positioning
- What lands clearly vs. what's confusing
- Emotional response and brand personality perception
- Differentiation from competitors
- Supporting quotes

**3. Tagline/Slogan Analysis** (if tested):
- Resonance and memorability
- What it communicates (intended vs. actual)
- Comparison with alternatives (if tested)
- Refinement suggestions

**4. Brand Values Validation:**
- Which values resonate (ranked by importance). Include a chart ranking values by resonance.
- Which feel authentic vs. generic
- Values that drive affinity vs. values that don't matter
- Supporting quotes

**5. Tone & Voice** (if tested):
- Whether tone lands as intended
- Audience preference between options
- Category fit and authenticity

**6. Differentiation Analysis:**
- How brand is perceived vs. competitors
- What makes it stand out (or doesn't)
- Meaningful differentiation vs. superficial claims

**7. Emotional Connection:**
- Primary feelings the messaging evokes. Include a chart of dominant emotional responses.
- Whether brand feels relatable and "gets them"
- Factors that build (or block) affinity

**8. Audience Insights:**
- What this audience cares about in the category
- What builds trust and loyalty
- What turns them off
- Key emotional drivers

**9. Recommendations:**
- Refined positioning statement (if needed)
- Alternative tagline options (if current one didn't land)
- Brand value prioritization (which to lead with)
- Tone adjustments
- Messaging do's and don'ts

**10. Next Steps:**
- If validated: Move forward with confidence
- If needs refinement: Specific changes to make and retest
- If messaging missed: Strategic pivot recommendations

## Brand Messaging Framework

Strong brand messaging answers these questions clearly:

| Question | What It Tests |
|----------|--------------|
| **Who are we?** | Category/space clarity |
| **Who are we for?** | Target audience clarity |
| **What do we do?** | Core offering clarity |
| **Why does it matter?** | Value/outcome clarity |
| **What makes us different?** | Differentiation clarity |
| **Why should you trust us?** | Credibility signals |
| **What should you feel?** | Emotional tone |

If your messaging doesn't clearly answer these, use the research to refine until it does.

## Red Flags (Rethink Messaging)

- Audience can't articulate what the brand does
- Positioning feels generic and interchangeable with competitors
- Values feel like empty corporate speak
- Tagline is forgettable or confusing
- Emotional response is neutral or negative
- People struggle to describe the brand in their own words

## Green Lights (Strong Messaging)

- Audience clearly understands what the brand does and who it's for
- Messaging evokes specific, positive emotions
- Differentiation is clear and meaningful
- Values feel authentic and tied to actions
- Tagline is memorable and captures essence
- People can easily explain the brand to others
- Brand feels relatable and "gets them"

## Tips for Best Results

- **Be willing to hear "no"**: If messaging doesn't resonate, it's better to know now than after a launch
- **Specificity beats aspiration**: "We help X do Y so they can Z" beats "We're transforming the future of [industry]"
- **Watch for comprehension gaps**: If people don't understand what you do, nothing else matters
- **Generic values don't differentiate**: "Innovation, integrity, excellence" — every brand claims these
- **Differentiation must be meaningful**: "We care more" isn't differentiation; "We're the only [category] for [specific need]" is
- **Emotional connection drives loyalty**: People may buy on features but stay loyal to brands they feel connected to
- **Test competitors too**: Understanding how audience perceives competitors reveals positioning opportunities

## Related Skills

- **deep-customer-research**: For deeper audience understanding before testing messaging
- **creative-testing**: To test specific creative executions of your brand
- **icp-discovery**: To find the right audience for your brand
- **landing-page-optimization**: To apply validated messaging to your landing page

Content Generation Optimization

Generate SEO and AEO-optimized content informed by real audience insight. This workflow researches your target audience to understand what they want to know, what tone resonates, and what makes content engaging — then uses those insights to write blog posts, articles, or guides that rank well in search engines, appear in AI answer engines, and actually get read.

Use when: Requests to write blog posts, create SEO content, generate articles, write guides, or create audience-aligned content.

Download skill

View full skill

---
name: content-generation-optimization
version: 1.1.0
description: "Use this skill to generate SEO and AEO-optimized content (blog posts, articles, guides) informed by real audience insight. Triggers include: requests to write blog posts, create SEO content, generate articles, write guides, or create audience-aligned content. Uses OriginalVoices Digital Twins (ask_twins) to understand what the audience wants to know, what tone resonates, what questions they have, and what makes content engaging — then uses those insights to write content that ranks well, gets read, and drives results."
---

# Content Generation Optimization Skill

## Overview

Generate SEO and AEO-optimized content informed by real audience insight. This workflow researches your target audience to understand what they want to know, what tone resonates, and what makes content engaging — then uses those insights to write blog posts, articles, or guides that rank well in search engines, appear in AI answer engines, and actually get read.

## Workflow Steps

### Step 1: Define Content to Generate

Gather from the user:
- **Topic/headline**: What is this content about? (e.g. "How to choose the right CRM for small businesses")
- **Target audience**: Who is this for? (e.g. "Small business owners with 5-20 employees")
- **Content goal**: What should this achieve? (SEO rankings for specific keywords, lead generation, education, thought leadership, AEO visibility)
- **Desired length**: Word count or depth (e.g. "1,500-2,000 words", "comprehensive guide", "quick how-to")
- **Content type**: Blog post, guide, article, how-to, listicle, case study, etc.
- **SEO keywords** (if applicable): Primary and secondary keywords to target
- **Specific requirements**: Tone, structure, sections to include, CTAs, etc.

### Step 2: Research Audience to Inform Content

One comprehensive research call to understand what will make this content valuable, engaging, and optimized for the audience.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Capturing the audience's own words.** This skill generates copy from what real people say, so the research must surface their *own language*, not abstract opinions — generic questions produce generic copy. Include at least two moment-anchored questions built to elicit verbatim phrasing (e.g. "The last time you looked for [X], what did you actually type or say?", "How would you describe [problem] in your own words, to a friend?"). When you later test copy, replace "would you click / stop scrolling?" (demand effect + inflated intent) with "what would you do next, if anything?" — and ask "what do you take this to be offering?" (comprehension) before whether they like it.

```
ask_twins(
  audience: "[detailed target audience]",
  questions: [
    "When you're looking for information about [topic], what are you trying to learn or accomplish? What's your main goal?",
    "What specific questions do you have about [topic] that you want answered?",
    "Think about the last piece you read on [topic] — what did it do well, and where did it fall short or waste your time?",
    "When something's written well on [topic], how would you describe the way it's written? And what kind of writing on this makes you click away?",
    "Think about a piece on [topic] you actually read to the end — how long and how deep was it? When do you want the full deep-dive, and when do you just want the quick version?",
    "The last time you trusted a piece on [topic] enough to act on it, what earned that trust? And what makes you distrust something on this subject?",
    "What would make you actually read an article about [topic] all the way through vs. just skimming?",
    "If you were searching for '[topic]' on Google, what would you type? What would you expect to find?",
    "What kind of examples or real situations, if any, actually help you when you're reading about [topic] — and which feel like filler?",
    "Think of the last article you saved or sent to someone — what was it about [topic] that made it worth passing on?",
    "Are there common mistakes or misconceptions about [topic] you see repeated that you wish someone would address?",
    "What's missing from most content on [topic]? What do you wish someone would explain better?"
  ]
)
```

### Step 3: Analyse Insights & Plan Content Structure

Review audience responses and extract:

**Key Questions to Answer:**
- What specific questions the audience has
- What they're trying to accomplish
- What's missing from existing content

**Tone & Style:**
- Casual vs. professional
- Depth: comprehensive vs. concise
- Structure: how-to, storytelling, data-driven, etc.

**Credibility Signals:**
- What builds trust (data, examples, credentials, case studies)
- What examples or scenarios resonate

**Engagement Hooks:**
- What makes them read vs. skim
- What makes them share or bookmark
- Pain points and frustrations to address

**SEO/AEO Optimization:**
- How they search for this topic (keywords, questions)
- What they expect to find
- How to structure for featured snippets and AI answers

### Step 4: Write the Complete Blog Post/Article

**IMPORTANT**: Using the insights from Step 3, write the complete, publish-ready blog post or article. Output the full written content, not just an outline or summary.

**Structure your written content as follows:**

**Headline:**
- Compelling, click-worthy headline that includes the primary keyword
- Based on what makes the audience interested (from research)

**Introduction (2-3 paragraphs):**
- Open with a hook that addresses their pain point or goal (from research)
- Establish what they'll learn and why it matters to them
- Set expectations for depth and value
- Make them want to keep reading

**Body (Main Content):**
- Use H2 and H3 headers that answer their specific questions (from research)
- Answer the key questions they have in order of importance
- Address frustrations with existing content they mentioned
- Write in their preferred tone (casual/professional based on research)
- Include credibility signals they care about (data, examples, case studies)
- Correct common misconceptions they mentioned
- Use short paragraphs, bullets, and formatting for scannability
- Naturally incorporate SEO keywords
- Structure answers clearly for featured snippets and AI answer engines

**Examples/Evidence (throughout body):**
- Include real-world scenarios or case studies that resonate with them
- Add data or statistics that build credibility (if they value data)
- Use practical examples they can relate to their situation

**Conclusion (2-3 paragraphs):**
- Summarize the key takeaways
- Provide clear, actionable next steps
- Include CTA aligned with content goal (newsletter signup, product trial, related content, etc.)

**Write the complete article now** — from headline through conclusion. Make it ready to publish.

### Step 5: Deliver the Content

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers.

**PRIMARY DELIVERABLE — The Complete Written Blog Post/Article:**

Output the full, publish-ready blog post or article you wrote in Step 4, including:
- Headline
- Complete introduction
- Full body with all sections and headers
- Examples and evidence throughout
- Complete conclusion with CTA
- Ready to copy and publish

**THEN provide supporting materials:**

**1. Content Brief:**
- **Target audience**: Who this is for
- **Primary goal**: SEO, lead gen, education, etc.
- **Word count**: Actual length
- **Tone**: Casual/professional based on research
- **Primary keyword**: Main SEO target
- **Secondary keywords**: Supporting terms

**2. Key Insights Used:**
- Top 3-5 audience insights that shaped the content
- Questions answered (from research)
- Pain points addressed
- Credibility signals included
- Engagement hooks used

**3. SEO/AEO Optimization Summary:**
- **Target keywords**: Primary and secondary
- **Search intent**: What audience is looking for
- **Featured snippet opportunity**: How content is structured to capture it
- **AEO optimization**: How it's structured for AI answer engines
- **Internal linking opportunities**: Related content to link to
- **Meta description suggestion**: Based on audience search behavior

**4. Performance Recommendations:**
- How to measure success (traffic, rankings, engagement, conversions)
- Distribution channels (where to promote based on audience research)
- Potential updates or follow-ups based on content gaps identified

## Content Generation Framework

| Content Element | Audience Insight Source | Optimization |
|----------------|------------------------|--------------|
| **Headline** | What makes them click | Include primary keyword, promise value |
| **Introduction** | Pain points and goals | Hook with their frustration/goal, set expectations |
| **Structure** | Questions they have | H2/H3 headers that answer specific questions |
| **Tone** | Casual vs. professional preference | Match their preferred style |
| **Examples** | What makes it relatable | Use scenarios they mentioned or similar |
| **Credibility** | What builds trust | Include data/examples/credentials they value |
| **Depth** | How comprehensive they want | Match preferred length and detail level |
| **SEO** | How they search | Natural keyword usage, semantic relevance |
| **AEO** | Questions they ask | Clear, direct answers; structured data |
| **CTA** | What makes them take action | Aligned with content goal and their needs |

## Tips for Best Results

- **Write for the audience, optimize for search**: Use insights to inform content, then layer in SEO naturally
- **Answer the real questions**: Don't assume — use the specific questions from research
- **Match their depth preference**: Don't write 3,000 words if they want quick, actionable advice
- **Use their language**: Incorporate words and phrases they used in responses
- **Address frustrations**: Call out what existing content gets wrong and do it better
- **Build credibility their way**: If they value data, include data; if they value stories, include stories
- **Structure for skimming**: Use headers, bullets, short paragraphs — they said what makes them read vs. skim
- **Optimize for featured snippets**: Use clear question-answer format for key questions
- **Make it actionable**: Include practical next steps they can take

## Related Skills

- **deep-customer-research**: For deeper audience understanding before content creation
- **landing-page-optimization**: To optimise the pages content links to
- **brand-messaging-validation**: To ensure content aligns with validated brand voice
- **creative-testing**: To test content angles and headlines before writing

Creative Brief

Generate a creative brief grounded in real audience insight using OriginalVoices Digital Twins. Instead of writing briefs based on assumptions or internal opinions, this workflow builds the brief from validated audience understanding — so creative teams start with what actually matters to real people.

Use when: Requests to write a creative brief, brief a creative team or agency, define a campaign brief, establish creative direction, or prepare a brief for designers, copywriters, or content creators.

Download skill

View full skill

---
name: creative-brief
version: 1.1.0
description: "Use this skill to generate a creative brief grounded in real audience insight. Triggers include: requests to write a creative brief, brief a creative team or agency, define a campaign brief, establish creative direction, or prepare a brief for designers, copywriters, or content creators. Uses OriginalVoices Digital Twins (ask_twins) to deeply understand the target audience's language, emotions, motivations, and preferences — then synthesises findings into a structured creative brief that gives creative teams a validated human foundation to build from, not assumptions."
---

# Creative Brief

## Overview

Generate a creative brief grounded in real audience insight using OriginalVoices Digital Twins. Instead of writing briefs based on assumptions or internal opinions, this workflow builds the brief from validated audience understanding — so creative teams start with what actually matters to real people.

## Workflow Steps

### Step 1: Gather Inputs

Collect from the user:

- **Product or service**: What is being promoted or communicated?
- **Campaign objective**: What should this creative achieve? (e.g. awareness, trial, conversion, repositioning)
- **Target audience**: Who is this for? Be as specific as possible — demographics, interests, lifestyle, values
- **Deliverables**: What will the creative team produce? (e.g. social ads, video, landing page, packaging, brand campaign)
- **Brand guidelines** (if any): Existing tone, visual identity, or messaging guardrails
- **Competitors or context** (optional): Who else is in the space? What's the competitive landscape?
- **Budget/timeline context** (optional): Any constraints that shape the brief

### Step 2: Audience Research via Digital Twins

Conduct deep audience research to build the brief on validated insight, not guesswork. Ask 10-12 questions covering emotional drivers, language, pain points, aspirations, preferences, and what makes the audience act.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

```
ask_twins(
  audience: "[detailed target audience description]",
  questions: [
    "Walk me through the last time you shopped for or used something in [product category] — what were you trying to get right, and what did you actually care about?",
    "Thinking about your recent experiences with [product category], what has worked well for you and what has been more of a struggle?",
    "Tell me about a time [the situation this product deals with] stirred up a real reaction for you — what was going on, and how did you feel?",
    "Tell me about a recent ad or piece of content in [category] that actually made you stop — what was it, and what pulled you in?",
    "What kind of tone or voice feels right for a [category] brand talking to someone like you — and what feels off?",
    "Tell me about the last time you went looking for a [product/solution] in this space — what set it off, and how did it play out?",
    "Has a brand in this space ever won or lost your trust? What did they do — and what would it take for a new one to earn a try?",
    "When something in this category has actually worked well for you, what did it change about your day — and where does it still fall short?",
    "What words or phrases do you use when talking about [problem/category] with friends or family?",
    "When [category] advertising has caught your eye — or made you roll your eyes — what made the difference?",
    "Tell me about a brand that made you feel genuinely understood — what did they say or show? And one that clearly didn't get you?",
    "Tell me about the last time you took a recommendation on something in [category] — who or what pointed you to it?"
  ]
)
```

**Note:** Tailor questions to the specific product and campaign. If the brief is for a visual campaign, add questions about imagery and aesthetics. If it's for a specific channel, ask about that channel's consumption habits.

### Step 3: Identify the Key Insight

From the research, identify the single most important human truth — the insight that should anchor the entire brief. A strong insight:

- Reveals a tension, desire, or unmet need the audience feels deeply
- Is specific enough to inspire creative work (not generic)
- Creates an "of course" reaction — it feels true and recognisable
- Connects the audience's world to what the product can deliver

**Test your insight:** If you could say it about any brand in any category, it's too generic. If it only makes sense for this audience and this product, you've found it.

### Step 4: Deliver the Creative Brief

Format the output using proper markdown headers: use `##` for the brief title, `###` for major sections, `####` for subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers.

Structure the brief as follows:

**1. Background & Context**
- What's the situation? Why is this creative needed now?
- Product/service overview
- Competitive landscape (informed by what the audience actually said about alternatives)

**2. Objective**
- What must this creative achieve?
- Primary KPI or success metric

**3. Target Audience Profile**
- Demographics and psychographics
- How they think and feel about the category (grounded in Twin responses)
- Their language — actual words and phrases they use
- What they care about most, in their own words
- Key emotional drivers

**4. Key Insight**
- The single human truth that anchors this brief
- Supporting evidence from audience responses

**5. Single-Minded Message**
- The one thing the audience should think, feel, or do after seeing this creative
- Must flow directly from the insight

**6. Support Points**
- 3-4 reasons to believe
- Product truths that validate the message
- Proof points or credibility signals the audience said matter to them

**7. Tone & Manner**
- How the creative should feel (grounded in what the audience said resonates)
- What to avoid (grounded in what the audience said turns them off)
- Audience quotes that capture the right tone

**8. Audience Do's and Don'ts**

| Do (resonates with this audience) | Don't (falls flat or alienates) |
|---|---|
| [Specific thing that works, with quote] | [Specific thing to avoid, with quote] |
| [Another thing that works] | [Another thing to avoid] |
| [Continue based on findings] | [Continue based on findings] |

**9. Mandatories & Guardrails**
- Brand guidelines to follow
- Legal or compliance requirements
- Channel-specific constraints

**10. Deliverables & Format**
- What the creative team needs to produce
- Dimensions, formats, or specifications

## Tips for Best Results

- **Lead with audience language**: Use the actual words Twins used — they reveal how real people think and talk about this category
- **One insight, not five**: The best briefs are ruthlessly focused. Pick the most powerful insight and build everything around it
- **Tensions are gold**: If the audience revealed a tension (e.g. "I want X but I'm afraid of Y"), that's often the strongest creative territory
- **Specificity inspires creativity**: "Busy mums who feel guilty about screen time" inspires better work than "parents aged 25-45"
- **Include raw quotes**: Creative teams connect with real voices more than summarised findings
- **Don't prescribe executions**: The brief should inspire creative solutions, not dictate them. Define the "what" and "why", let the creative team own the "how"

## Common Pitfalls to Avoid

- **Brief by committee**: Cramming every stakeholder's opinion into the brief dilutes focus
- **Generic insights**: "People want quality at a good price" isn't an insight — it's a truism
- **Multiple messages**: If the brief has more than one key message, it has no key message
- **Skipping the audience**: Writing the brief first and then looking for audience data to support it defeats the purpose
- **Internal language in the brief**: If the audience wouldn't use a word, it shouldn't be in the insight or message
- **Confusing features with benefits**: The audience cares about outcomes, not specifications

## Creative Brief Quality Checklist

Before delivering, verify:

- [ ] Insight is specific to this audience and product (not interchangeable)
- [ ] Single-minded message is truly single-minded (one idea, not three)
- [ ] Audience profile uses their actual language, not marketing speak
- [ ] Tone guidance is grounded in what the audience said, not internal preference
- [ ] Do's and Don'ts are backed by real audience quotes
- [ ] Brief would inspire creative work (not just inform it)
- [ ] A creative team could start working from this brief alone

Creative Testing

Test and validate creative concepts — ads, marketing copy, taglines, scripts, packaging, or any creative output — using real audience feedback from OriginalVoices Digital Twins before committing budget. This workflow gathers honest gut reactions, identifies which concepts resonate and why, and provides actionable recommendations for optimisation.

Use when: Requests to test ad concepts, validate marketing copy or taglines, compare creative directions, get audience reactions to campaign ideas, or pre-test scripts or email copy. Supports testing up to 10-12 concepts in a single session.

Download skill

View full skill

---
name: creative-testing
version: 1.1.0
description: "Use this skill to test and validate creative concepts — ads, copy, taglines, scripts, packaging — with real audience feedback before spending budget. Triggers include: requests to test ad concepts, validate marketing copy or taglines, compare creative directions, get audience reactions to campaign ideas, or pre-test scripts or email copy. Supports testing up to 10-12 concepts in a single session. Uses OriginalVoices Digital Twins (ask_twins) to gather honest gut reactions from target audiences."
---

# Creative Testing Skill

## Overview

Test and validate creative concepts — ads, marketing copy, taglines, scripts, packaging, or any creative output — using real audience feedback from OriginalVoices Digital Twins before committing budget. This workflow gathers honest gut reactions, identifies which concepts resonate and why, and provides actionable recommendations for optimisation.

## Workflow Steps

### Step 1: Gather Creative Concepts & Audience

Collect from the user:
- **Creative concepts**: The ads, copy, taglines, scripts, or other creative to test (up to 10-12)
- **Target audience**: Who are these for? Be specific — age, location, interests, lifestyle
- **Context** (optional): Where will these run? What's the campaign goal?

### Step 2: Design the Testing Approach

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Testing stimuli without bias.** When showing concepts, ads, taglines, or messaging options:
- Capture the **unaided gut reaction first** ("what's your first impression?") before asking for reasons.
- **Don't signal the hoped-for winner** — never reveal which option is the client's, and never ask "isn't this great?".
- When comparing options, **counterbalance the order** (don't always show A first) and always allow "neither / none of them".
- **Keep three questions separate:** comprehension ("what message do you take from this? who is it for?") → preference ("which do you prefer, and why?") → persuasion ("does it make you more or less likely to act?"). Never fuse them into one.

Choose the right approach based on the number of concepts:

**Small batch (2-4 concepts)** — Direct comparison:
Present all concepts together and ask for direct comparison.

```
ask_twins(
  audience: "[detailed target audience]",
  questions: [
    "Here are [number] [concepts]. Reading each one cold, give me your gut first impression of each in a few words — before you think about why.",
    "Concept A: '[full text of concept A]'. What's the main message you take from this, and who do you think it's meant for?",
    "Concept B: '[full text of concept B]'. What's the main message you take from this, and who do you think it's meant for?",
    "Concept C: '[full text of concept C]'. What's the main message you take from this, and who do you think it's meant for?",
    "Which of these pulls you in most, and which leaves you cold? Walk me through why — and if none of them grab you, say so.",
    "Thinking about the last time an ad actually got you to do something, does any of these come close — and does any make you less likely to bother? Which, and why?",
    "Is there anything in any of these that feels off, confusing, or that you'd scroll past — and what exactly triggers that?",
    "Take the one that landed best for you: what would you change to make it hit harder, and what should stay exactly as it is?"
  ]
)
```

**Large batch (5-12 concepts)** — Split into groups:
Break concepts into groups of 3-4, test each group with individual reaction prompts, then run a final ranking question across all.

```
# Group 1
ask_twins(
  audience: "[detailed target audience]",
  questions: [
    "Concept A: '[full text]'. Reading it cold, what message do you take from it, and how does it land with you?",
    "Concept B: '[full text]'. Reading it cold, what message do you take from it, and how does it land with you?",
    "Concept C: '[full text]'. Reading it cold, what message do you take from it, and how does it land with you?",
    "Of A, B, and C, which pulls you in most and which leaves you flat? Walk me through why — and if none land, say so."
  ]
)

# Group 2
ask_twins(
  audience: "[same audience]",
  questions: [
    "Concept D: '[full text]'. Reading it cold, what message do you take from it, and how does it land with you?",
    "Concept E: '[full text]'. Reading it cold, what message do you take from it, and how does it land with you?",
    "Concept F: '[full text]'. Reading it cold, what message do you take from it, and how does it land with you?",
    "Of D, E, and F, which pulls you in most and which leaves you flat? Walk me through why — and if none land, say so."
  ]
)

# Final ranking
ask_twins(
  audience: "[same audience]",
  questions: [
    "Here are the top concepts from our testing: [list finalists]. Thinking about how you've actually acted on ads like this before, which comes closest to getting you to [desired action], and which wouldn't move you at all? Why?",
    "Which of these feels like it actually gets someone like you — and which feels off or like it's trying too hard? What's the tell?",
    "Take whichever one landed best for you: what would you change to make it stronger, and what should stay untouched?"
  ]
)
```

### Step 3: Analyse Results

For each concept, evaluate:

| Dimension | What to Assess |
|-----------|---------------|
| **Overall preference** | How it ranked relative to others |
| **Strengths** | What specifically resonated — words, feelings, associations |
| **Weaknesses** | What fell flat, confused, or turned people off |
| **Emotional impact** | What feelings it triggered (excitement, curiosity, trust, indifference) |
| **Relevance** | How connected it felt to the audience's real life |
| **Improvement suggestions** | What the audience would change |

**Look for patterns across respondents:**
- Consistent praise or criticism across multiple Twins
- Emotional language (strong signals) vs. polite but flat responses (weak signals)
- Specific use cases or scenarios people mention unprompted

### Step 4: Deliver the Creative Testing Report

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers. Where findings have clear quantitative patterns, include a chart to visualize the distribution.

### Report sections:

**1. Winner & Rationale**
- The top-performing concept and why
- Key audience quote that captures why it wins

**2. Concept-by-Concept Breakdown**

Include a chart ranking concepts by overall audience preference.

For each concept tested:
- **Verdict**: Winner / Strong / Moderate / Weak
- **Strengths**: What worked (with audience quotes)
- **Weaknesses**: What didn't land
- **Emotional response**: What feelings it triggered
- **Audience quote**: One representative reaction

**3. Key Insight**
- The primary insight about what this audience responds to
- What this reveals about their preferences, values, or communication style

**4. Recommendations**
- Which concept(s) to move forward with
- Specific refinements based on audience feedback
- What to test next (if applicable)

**5. Improvement Suggestions**
- For the winning concept: audience-suggested improvements
- For runner-up concepts: what would need to change to make them competitive

## Tips for Best Results

- **Present concepts at equivalent polish levels**: Don't compare a rough draft against a polished version
- **Prioritise initial gut reactions**: First impressions are the most honest signals
- **Don't reveal which concept you prefer**: Let the audience react without bias
- **Include direct Digital Twin quotes**: They carry more weight than summaries
- **Test genuinely different angles**: 10 variations of the same idea isn't useful — test distinct creative directions
- **Watch for "polite" feedback**: "That's nice" isn't a win. Look for strong emotional language and specific reasons

## Related Skills

- **facebook-ad-copy**: Generate audience-informed Facebook/Instagram ad variations
- **google-ad-copy**: Generate audience-informed Google RSA assets
- **brand-messaging-validation**: Test brand positioning and taglines specifically
- **landing-page-optimization**: Test landing page messaging and structure

Deep Customer Research

Deeply understand what your customers really think and feel about any topic. This workflow uses OriginalVoices Digital Twins to conduct comprehensive qualitative research in a single pass — asking 10-12 carefully designed questions per audience that cover the full landscape of motivations, emotions, behaviours, and unmet needs.

Use when: Validating product-market fit, exploring how an audience feels about a trend, or gathering insights to inform product strategy, positioning, or roadmap decisions.

Download skill

View full skill

---
name: deep-customer-research
version: 1.1.0
description: "Use this skill to conduct deep qualitative customer research on any topic. Triggers include: requests to understand customer pain points, unmet needs, motivations, or emotions around a product, category, brand, or market. Also use when validating product-market fit, exploring how an audience feels about a trend, or gathering insights to inform product strategy, positioning, or roadmap decisions. Uses OriginalVoices Digital Twins (ask_twins) to ask 10-12 comprehensive questions per audience in a single pass covering behaviour, pain points, emotions, priorities, decision-making, unmet needs, trust signals, and willingness to pay."
---

# Deep Customer Research Skill

## Overview

Deeply understand what your customers really think and feel about any topic. This workflow uses OriginalVoices Digital Twins to conduct comprehensive qualitative research in a single pass — asking 10-12 carefully designed questions per audience that cover the full landscape of motivations, emotions, behaviours, and unmet needs.

## Workflow Steps

### Step 1: Define the Research Topic & Audience

Gather from the user:
- **Research topic or question**: What do they want to understand? (e.g. "How do parents feel about screen time management apps?")
- **Target audience**: Who should we hear from? Be as specific as possible — age, location, interests, lifestyle, values. (e.g. "Parents aged 28-45 in the US with children under 12 who are concerned about their kids' screen time")

### Step 2: Design the Question Set

Craft 10-12 open-ended questions that cover the full research landscape in a single round. This count is PER AUDIENCE: when comparing segments (e.g. Gen Z vs Gen X), each cohort gets its own full set of 10-12 comparable questions — never split one question budget across cohorts. The questions should progress from broad context through to specific opinions and forward-looking needs.

**The #1 quality goal: maximise the *range* of answers.** A good question surfaces spread, tension, and surprise across respondents. A weak question is *narrow* or *convergent* — it points everyone at the same obvious answer, so the twins echo each other and you learn nothing. Before finalising any question, ask: *"Could I predict most of the answers before asking? Would 15 different people answer this 15 different ways?"* If the answers would cluster, rewrite the question.

**Rules — apply every one of these to every question you write. These are requirements, not suggestions. The examples below are illustrations of the rules, not templates to copy verbatim — adapt the wording to the specific topic and audience.**

1. **Ask about a specific moment or behaviour, not an abstract definition.** Never ask "How do you define / think about [concept]?" — it collapses to textbook answers. Ask "Walk me through the last time…" / "Tell me about a time…".
2. **Keep the valence open.** Never presuppose the experience was good or bad. Banned phrasings: "the time [X] annoyed/frustrated/disappointed you", "where does it fall short", "the last time [brand] let you down". Instead ask "how did it go?" or make it two-sided: "what worked well and what didn't?".
3. **One idea per question.** If a question contains "and" joining two different topics (cost *and* quality), split it into two questions.
4. **No socially-correct or yes/no answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe until there's no obvious "right" reply — usually by adding a trade-off ("when is quality *not* worth it?").
5. **Stay neutral — never lead.** The question must not contain the answer or the sentiment. No "Don't you find…", no "How frustrating is…".
6. **Build in room to diverge.** Prefer questions with a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over flat single-answer prompts.
7. **Open-ended only.** Every question starts with or invites "How / What / Why / Walk me through / Tell me about". No closed questions.

Run the two final tests in Step 3 (Range + Neutrality) on each question before sending. If a question fails any rule above, rewrite it.

**How the rules play out (illustrative — do not copy verbatim):**

- **Anchor in lived experience, not abstract definitions.** Questions that ask people to *define* a concept ("How do you define long-term savings?") collapse to near-identical textbook answers. Instead ask for a specific moment, story, or behaviour ("Tell me about the last time you thought about your long-term savings — what prompted it and how did it feel?"). Concrete beats conceptual, every time.
- **Ask for the specific over the general.** "Walk me through the last time…", "Describe a moment when…", "Tell me about your most recent experience with…". Recalled episodes produce vivid, divergent, quotable answers; general opinions produce bland consensus.
- **Keep the valence open — this is what separates a strong open question from a leading one.** Don't presuppose the experience was good or bad. "The last time [topic] annoyed you…" assumes it did; "Tell me about your last experience with [topic] — how did it go?" lets the respondent supply the emotion, and range comes *from* that freedom. When you want to probe frustration or delight, make it two-sided: "…what worked well, and what didn't?" rather than only asking for the negative.
- **Create room for divergence.** Build in a fork ("What would make you do X — and what would hold you back?"), a trade-off ("What would you give up to get Y?"), or a spectrum ("Where do you sit between A and B, and why?"). Questions with a built-in tension pull answers apart without loading the answer.
- **Surface the exceptions, neutrally.** "What's something about [topic] you think most people get wrong?" or "When does [category] work well for you, and when does it fall short?" pull out the tail of the distribution — but phrase them so a happy respondent and an unhappy one can both answer honestly.
- **One idea per question.** Don't bundle ("How do you feel about cost, quality, and convenience?") — split them. Bundled questions get partial, uneven answers.
- **Avoid questions with a socially-correct answer.** "Is saving money important to you?" → everyone says yes. Reframe so there's no obvious "right" reply ("When is saving *not* worth it for you?").
- **Stay neutral and non-leading.** Don't smuggle in the answer or the sentiment ("Don't you find X frustrating?", "where does it fall short?"). Ask "What's your experience with X?" and let them tell you whether it's good, bad, or mixed.

**Question design framework — cover all of these areas:**

Each example below is framed to pull answers *apart* — anchored in a specific moment, choice, or tension rather than an abstract definition.

| Area | Purpose | Example (open, experience-anchored) |
|------|---------|---------|
| Current behaviour | How they deal with the problem today | "Walk me through the last time you dealt with [problem] — what did you actually do, step by step?" |
| Pain points | What works and what doesn't | "Thinking about your recent experiences with [topic], what has gone well and what has been more of a struggle?" |
| Emotional drivers | What feelings are at play | "Tell me about a time [topic] stirred up a strong reaction for you — what was going on, and how did you feel?" |
| Priorities | What matters most (with trade-off) | "If you could only get one thing right about [category], what would it be — and what would you let slide?" |
| Decision-making | How they choose | "Tell me about the last time you chose a [product/solution] in this space — what tipped the decision?" |
| Current solutions | What they use now and why | "What are you using for [problem] right now, and how well is it working for you?" |
| Unmet needs | Gaps in what's available | "Is there anything you wish [products] in this space could do that none of them seem to?" |
| Ideal outcome | What great looks like | "Imagine you've just had the perfect experience with [solution] — describe what happened." |
| Triggers | What would make them act | "What would make you switch from your current approach — and what's kept you with it so far?" |
| Trust & credibility | What builds confidence | "Has a brand in this space ever earned or lost your trust? What did they do, and what would it take to earn it?" |
| Social influence | Role of others | "When you make decisions about [category], what part do other people — friends, reviews, social media — play, if any?" |
| Willingness to pay / trade-offs | Value perception | "What's the most you'd realistically pay for something that solved [problem] — and what would make it feel worth it or not?" |

### Step 3: Ask All Questions in a Single Call

Send all 10-12 questions to the Digital Twins in one `ask_twins` call per audience (one call per cohort when comparing segments, each with the full question set).

```
ask_twins(
  audience: "[detailed audience description]",
  questions: [
    "Walk me through the last time you dealt with [problem] — what did you actually do, step by step?",
    "Thinking about your recent experiences with [topic], what has gone well and what has been more of a struggle?",
    "Tell me about a time [topic] stirred up a strong reaction for you — what was going on, and how did you feel?",
    "If you could only get one thing right about [category], what would it be — and what would you let slide?",
    "Tell me about the last time you chose a [product/solution] in this space — what tipped the decision?",
    "What are you using for [problem] right now, and how well is it working for you?",
    "Is there anything you wish [products] in this space could do that none of them seem to?",
    "Imagine you've just had the perfect experience with [solution] — describe what happened.",
    "What would make you switch from your current approach — and what's kept you with it so far?",
    "Has a brand in this space ever earned or lost your trust? What did they do, and what would it take to earn it?",
    "When you make decisions about [category], what part do other people — friends, reviews, social media — play, if any?",
    "What's the most you'd realistically pay for something that solved [problem] — and what would make it feel worth it or not?"
  ]
)
```

**Before sending, run each question through this quick check:**

| Weak question | Why it fails | Strong (open + wide range) |
|---|---|---|
| "How do you define long-term savings?" | Definitional → everyone gives the same textbook answer | "Tell me about the last time you thought about your long-term savings — what set it off and how did it feel?" |
| "Is quality important when you choose [product]?" | Socially-correct answer → everyone says yes | "When is quality worth paying up for in [category], and when isn't it?" |
| "Do you like [category]?" | Yes/no, no spread | "What do you like about [category], and what would you change if you could?" |
| "The last time [product] let you down, what happened?" | **Leading** — presupposes it let them down; forces a negative | "Tell me about your most recent experience with [product] — how did it go?" |
| "How do you feel about cost, quality and service?" | Bundled → partial, uneven answers | Split into three separate questions, one idea each |

Two tests before you send: (1) **Range** — could most respondents answer this the same way, or is there an obvious "right" answer? If so, anchor it in a specific moment or add a trade-off. (2) **Neutrality** — does the question assume how they feel (annoyed, disappointed, delighted)? If so, open the valence so a happy, unhappy, or indifferent respondent can each answer it honestly. A question should pass *both* — wide range, no leading.

### Step 4: Analyse & Synthesise

Review all Digital Twin responses and compile findings:

- **Key themes**: Patterns across respondents and across questions
- **Emotional landscape**: Dominant feelings (frustration, anxiety, hope, guilt, indifference)
- **Current behaviour map**: How the audience navigates this space today
- **Pain point ranking**: Most intense and most common frustrations
- **Unmet needs**: Clearest gaps between what's available and what's wanted
- **Decision drivers**: What influences choice and action
- **Trust signals**: What builds or breaks confidence
- **Surprising insights**: Anything unexpected or counterintuitive

### Step 5: Deliver the Research Report

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for individual themes/subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers. Where findings have clear quantitative patterns (e.g. "7 out of 15 twins mentioned cost"), include a chart to visualize the distribution.

### Report sections:

1. **Executive Summary** — Top-line findings in 2-3 sentences
2. **Key Themes** — 3-5 most significant patterns, with supporting quotes from Digital Twins. Include a chart showing theme prevalence.
3. **Emotional Landscape** — What emotions drive behaviour in this space. Include a chart of dominant emotions.
4. **Current Behaviour & Pain Points** — How the audience deals with the problem today and what's broken
5. **Unmet Needs & Ideal Outcomes** — Gaps and what "great" looks like to the audience
6. **Decision Drivers & Trust Signals** — What influences choice and earns confidence. Include a chart ranking decision factors.
7. **Opportunities** — Actionable insights for product, marketing, or strategy
8. **Recommendations** — Specific next steps based on the research

## Tips for Best Results

- **Be specific with audiences**: "Women aged 25-35 in the UK interested in fitness who have tried meal planning apps" yields far better results than "women interested in health"
- **Optimise for range, not coverage**: The best questions produce *spread* — 15 people answering 15 different ways. If you can predict the answers in advance, the question is too narrow. Rewrite it (see the weak→strong check in Step 3).
- **Prefer stories to definitions**: "Tell me about the last time…" and "Walk me through…" beat "How do you define…" or "What do you think about…" — recalled episodes are vivid, specific, and divergent; abstractions converge.
- **Build in a tension or trade-off**: A fork ("what would make you — and what would stop you?") or a trade-off ("what would you give up for…?") pulls answers apart and surfaces the interesting minority views.
- **Ask open-ended questions**: Start with "How", "What", "Why", "Walk me through..."
- **Adapt the question template**: The 12 questions above are a framework — tailor them to the specific topic
- **Let the Twins speak**: Include direct quotes from Digital Twins in findings — they carry authenticity and emotional weight
- **Don't lead**: Ask "What matters to you about X?" not "Don't you think X is important?"
- **Look for tension**: The most valuable insights often sit in contradictions

## Related Skills

- **icp-discovery**: If you need to find the right audience before researching them
- **feature-concept-testing**: If you want to validate a specific feature idea
- **brand-messaging-validation**: If you want to test how your brand resonates
- **creative-testing**: If you want to test specific creative concepts

Email Campaign Copy

Generate email marketing campaigns informed by real subscriber insight. This workflow researches how your target audience engages with email — what captures their attention, what they value, and what drives action — then creates emails that speak to what real people actually care about.

Use when: Requests to write email campaigns, create newsletter content, improve email open rates, write welcome sequences, launch announcement emails, or re-engagement campaigns.

Download skill

View full skill

---
name: email-campaign-copy
version: 1.1.0
description: "Use this skill to generate email marketing copy grounded in real audience insight. Triggers include: requests to write email campaigns, create newsletter content, improve email open rates, write welcome sequences, launch announcement emails, or re-engagement campaigns. Uses OriginalVoices Digital Twins (ask_twins) to understand what makes the audience open emails, what messaging resonates, what triggers action, and what causes unsubscribes — then generates audience-informed email variations rooted in real subscriber perspectives."
---

# Email Campaign Copy Skill

## Overview

Generate email marketing campaigns informed by real subscriber insight. This workflow researches how your target audience engages with email — what captures their attention, what they value, and what drives action — then creates emails that speak to what real people actually care about.

## Workflow Steps

### Step 1: Gather Campaign Context

Collect from the user:
- **Target audience**: Who is this email for? (e.g. "SaaS customers in their first 30 days")
- **Email type/goal**: Welcome series, product launch, newsletter, re-engagement, promotion, etc.
- **Key message or offer**: What's the main point?
- **Number of variations**: How many? (Default: 5-8)

Optional:
- **Current performance**: Open rates, click rates, unsubscribe rates
- **Tone/brand voice**: Any specific voice to match?

### Step 2: Research Audience Email Behavior & Preferences

One comprehensive research call covering email behaviour, what drives opens and unsubscribes, value perception, and messaging preferences. Keep the research phase open-ended — save subject-line and opening-line A/B picks for the copy-validation step (Step 5), where forced choice belongs.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Capturing the audience's own words.** This skill generates copy from what real people say, so the research must surface their *own language*, not abstract opinions — generic questions produce generic copy. Include at least two moment-anchored questions built to elicit verbatim phrasing (e.g. "The last time you looked for [X], what did you actually type or say?", "How would you describe [problem] in your own words, to a friend?"). When you later test copy, replace "would you click / stop scrolling?" (demand effect + inflated intent) with "what would you do next, if anything?" — and ask "what do you take this to be offering?" (comprehension) before whether they like it.

```
ask_twins(
  audience: "[detailed target audience]",
  questions: [
    "Tell me about the last marketing email you actually opened and read — what was it, and what made you open it?",
    "Tell me about the last marketing email you deleted straight away or unsubscribed from — what was it about it that put you off?",
    "How would you describe [topic/offer] in your own words — the way you'd explain it to a friend?",
    "Think about a marketing email you were glad you read — what did it give you that made it feel worth your time rather than just noise?",
    "When you open a marketing email, what usually makes you keep reading past the first line — and what makes you close it?",
    "Tell me about the last time an email actually got you to click through to something — what was it, and what made you act?",
    "When you're weighing up whether something like [offer/product/concept] is worth it, what pulls you in — and what makes you hold back?",
    "Think about a brand whose emails you've stayed subscribed to — how often do they email you, and what makes that feel about right rather than too much?",
    "When you read an email that felt like it really got you, how was it written — the tone, the length, the way it talked to you?",
    "What's a subject line or phrase from a marketing email that actually stuck with you — good or bad — and what was it about it?"
  ]
)
```

### Step 3: Extract Key Insights

Analyze responses and identify:

**Subject Line Drivers:**
- What triggers opens (benefit, curiosity, urgency, personalization)
- What feels spammy or manipulative
- Optimal tone and length

**Email Body Preferences:**
- Optimal opening hook
- Tone that resonates (casual, direct, story-driven)
- Preferred length and structure

**CTA & Action:**
- What drives clicks
- How to frame the ask without being pushy

**Friction Points:**
- What causes unsubscribes
- Frequency tolerance
- What feels like noise

### Step 4: Generate Email Variations

Create 5-8 email variations, each built on different insights from research.

**For each email, deliver:**
- **Subject line** (6-10 words, grounded in what drives opens)
- **Preview text** (40-100 chars, complements subject line)
- **Email body**:
  - Opening hook (2-3 sentences)
  - Core message (3-5 short paragraphs or bullets)
  - CTA (clear, specific, action-oriented)
  - Sign-off
- **Insight label** — Which research finding drives this email
- **Best use case** — When/why to use this variant

**Spread variations across different angles:**
- Value-first (lead with clear benefit)
- Problem-aware (open with pain point)
- Direct/no-fluff (straight to the point)
- Social proof (credibility and trust)
- Story-driven (context or scenario)
- Curiosity-driven (only if research shows it resonates)

### Step 5: Validate Top Variants (Optional)

Test 3-5 strongest emails back with the audience. This is where forced-choice A/B picks belong — the research phase stayed open-ended, so use this step to compare specific subject lines, openings, and emails head to head.

```
ask_twins(
  audience: "[same audience]",
  questions: [
    "You see these subject lines in your inbox: A: '[Subject A]' B: '[Subject B]' C: '[Subject C]'. Which would you open first, and what makes the other two easy to skip?",
    "Here are two ways this email could open. A: '[Opening line A]' B: '[Opening line B]'. Which one keeps you reading, and where does the other lose you?",
    "You open this email and read the first few lines: '[Opening]'. In your own words, what do you take it to be offering — and what would you do next, if anything?",
    "Between these two emails, which feels more worth your time? A: '[Email A summary]' B: '[Email B summary]'. What makes the other one miss?",
    "Does this email feel worth your time or like noise? What makes you say that?",
    "Picture this CTA at the end of the email: '[CTA copy]'. What would you do when you reach it — and what, if anything, would make you more likely to act?"
  ]
)
```

Update emails based on feedback before finalizing.

### Step 6: Deliver Email Campaign Set

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for individual emails, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers.

**1. Campaign Overview:**
- Audience insights summary
- Key findings (what resonates, what to avoid)
- Recommended approach

**2. Email Variations (5-8 complete emails):**
- Subject line + preview text
- Full body copy + CTA
- Insight driving each email
- Best use case

**3. Subject Line Library:** 10-15 tested subject lines ranked by predicted performance

**4. A/B Testing Plan:**
- Which emails to test first
- What elements to test (subject, opening, CTA, length)
- Success metrics to track

**5. Do's and Don'ts (based on research):**
- What resonates with this audience
- What triggers unsubscribes

## Email Best Practices

| Element | Guideline |
|---------|-----------|
| **Subject line** | 6-10 words; clear benefit or curiosity without clickbait |
| **Preview text** | Complement subject line; don't repeat; add context |
| **Opening hook** | First 2 sentences determine if they keep reading |
| **Email body** | Short paragraphs (2-3 sentences); scannable; whitespace |
| **CTA** | One primary CTA; clear action; repeat if email is long |
| **Length** | Match audience preference (research will reveal) |
| **Frequency** | Align with tolerance from research |

## Email Sequence Considerations

For multi-email sequences (welcome series, onboarding, launch):

- **Map the journey**: What do they need to know first, second, third?
- **Vary the approach**: Don't repeat the same structure in every email
- **Build on previous emails**: Reference what they've learned or done
- **Respect timing**: Space emails based on frequency tolerance
- **Clear exit**: Make it easy to unsubscribe; forcing people to stay backfires

## Tips for Best Results

- **Subject lines make or break opens**: Test multiple angles; what works for one audience won't work for another
- **The first sentence is everything**: If the opening doesn't hook them, they're gone
- **Use their language**: Pull exact phrases from Digital Twin responses
- **Clarity over cleverness**: If they need to re-read to understand, simplify
- **One clear goal per email**: One CTA, one purpose
- **Respect their inbox**: If research shows frequency concerns, acknowledge it
- **Test drastically different approaches**: Not just subject variations; test fundamentally different structures and tones

## Related Skills

- **deep-customer-research**: For deeper audience understanding before writing campaigns
- **creative-testing**: To test email concepts before full production
- **landing-page-optimization**: To optimise the page email CTAs link to
- **brand-messaging-validation**: If brand voice needs validation first

Facebook Ad Copy

Generate Facebook and Instagram ad copy informed by real audience insight. Instead of relying on AI guesswork to write ads, this workflow first talks to your target audience to understand their language, preferences, concerns, and experiences — then uses those insights to create ads that speak to what real people actually care about.

Use when: Requests to create Facebook or Meta ad campaigns, write primary text and headlines for Facebook ads, refresh stale social ad creative, build ad sets for different audience segments, or generate audience-tested ad variations.

Download skill

View full skill

---
name: facebook-ad-copy
version: 1.1.0
description: "Use this skill to generate Facebook and Instagram ad copy grounded in real audience insight. Triggers include: requests to create Facebook or Meta ad campaigns, write primary text and headlines for Facebook ads, refresh stale social ad creative, build ad sets for different audience segments, or generate audience-tested ad variations. Uses OriginalVoices Digital Twins (ask_twins) to deeply research the target audience's language, preferences, concerns, and experiences, then generates 10-15 audience-informed ad variations rooted in what real people actually said — not AI guesswork."
---

# Facebook Ad Copy Skill

## Overview

Generate Facebook and Instagram ad copy informed by real audience insight. Instead of relying on AI guesswork to write ads, this workflow first talks to your target audience to understand their language, preferences, concerns, and experiences — then uses those insights to create ads that speak to what real people actually care about.

## Workflow Steps

### Step 1: Gather Inputs

Only two things are required:

- **Target audience**: Who are these ads for? (e.g. "Women aged 25-40 in the US interested in skincare and wellness")
- **Product/concept**: Either a description in their own words, or a link to a product page / landing page / pitch deck

Optional:
- **Number of ad variations**: How many do they want? Default to 10-15 if not specified
- **Tone or brand voice preferences**: Any specific tone to match?

### Step 2: Audience Research

This is the most important step. Use `ask_twins` to understand the audience's world — their language, what they care about, what gets in their way, what grabs their attention, and how they relate to the problem space. Ask broad questions about the general topic area, not just about the specific product.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Capturing the audience's own words.** This skill generates copy from what real people say, so the research must surface their *own language*, not abstract opinions — generic questions produce generic copy. Include at least two moment-anchored questions built to elicit verbatim phrasing (e.g. "The last time you looked for [X], what did you actually type or say?", "How would you describe [problem] in your own words, to a friend?"). When you later test copy, replace "would you click / stop scrolling?" (demand effect + inflated intent) with "what would you do next, if anything?" — and ask "what do you take this to be offering?" (comprehension) before whether they like it.

```
ask_twins(
  audience: "[detailed target audience description]",
  questions: [
    "Walk me through the last time [general topic area, e.g. skincare, fitness, cooking, managing finances] came up for you — what did you actually do, and how did it go?",
    "How would you describe [the problem/topic] in your own words — the way you'd explain it to a friend who was dealing with it too?",
    "Tell me about your most recent experience with a [product/solution in this space] — what worked well, and what didn't?",
    "When you think about trying a new [product in this category], what pulls you toward it — and what holds you back?",
    "Has a brand in this space ever earned or lost your trust? What did they do, and what would it take to earn it?",
    "Think about the last time you discovered a [product in this category] you decided to try — how did you come across it, and what made you give it a go?"
  ]
)
```

**Why this matters:** These questions surface the real language people use, the emotions they feel, the objections they hold, and the experiences that shape their decisions. This is what makes the ads resonate — not generic marketing copy.

### Step 3: Analyse Audience Insights

Review all responses and extract:

- **Pain points in their own words**: The exact language they use to describe frustrations
- **Preferences and priorities**: What they look for and value in this category
- **Experiences and context**: What they've tried before and how it went
- **Concerns and objections**: What would hold them back from trying something new
- **Trust signals**: What builds or breaks confidence
- **Emotional drivers**: The feelings that motivate action (relief, excitement, validation, fear of missing out)

### Step 4: Generate Ad Variations

Using the audience insights, generate 10-15 ad variations (or the number requested). Every ad should be directly traceable to something the audience actually said — a pain point, a preference, a phrase, an emotion.

**Facebook Ad Copy Structure:**
- **Primary text**: Main body copy (125 chars visible before "See more"; full text up to 1,000+)
- **Headline**: Appears below the image/video (up to 40 chars recommended)
- **Description**: Below the headline (up to 30 chars recommended)
- **CTA button**: Shop Now, Learn More, Sign Up, Get Offer, etc.

**Spread variations across different angles, drawing from the research:**
- **Pain point ads**: Open with a frustration the audience described, in their words
- **Benefit-first ads**: Lead with the outcome or feeling they said they want most
- **Social proof / trust ads**: Built around the trust signals and credibility markers they mentioned
- **Experience-based ads**: Reference the common experiences they shared (what they've tried, what failed)
- **Objection-handling ads**: Directly address a concern or hesitation from the research

For each ad, note which audience insight it's built on.

### Step 5: Validate with Audience (Optional but Recommended)

Test 3-5 of the strongest ads back with the audience.

```
ask_twins(
  audience: "[same target audience]",
  questions: [
    "You're scrolling Facebook and this ad goes past: '[primary text + headline]'. In your own words, what do you take it to be offering — and what would you do next, if anything?",
    "Here are three ways an ad could open. A: '[opening A]' B: '[opening B]' C: '[opening C]'. Which one sounds most like it's meant for you, and what makes the other two miss?",
    "Is there anything in these ads that doesn't ring true or feels off to you? What is it, and why?"
  ]
)
```

### Step 6: Deliver Final Ad Set

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections (e.g. by angle: Pain Point Ads, Benefit-First Ads), `####` for individual ad variations, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers.

Present the full set of ad variations with:
- Each ad in full (primary text, headline, description, CTA)
- The audience insight driving each ad clearly labelled
- Ads grouped by angle (pain point, benefit, trust, experience, objection)
- Validation results (if done)
- Top 3-5 recommended ads to test first
- A/B testing suggestions (which ads to test against each other)

## Facebook Ad Copy Guidelines

| Element | Recommended | Max |
|---------|-------------|-----|
| Primary text | 125 chars (before truncation) | ~1,000+ chars |
| Headline | 27-40 chars | 255 chars |
| Description | 27-30 chars | 255 chars |

## Tips for Best Results

- **The research makes the ads**: The audience research step is what separates these ads from generic AI output. Don't rush it
- **Write how people talk**: If the audience says "I just want something that actually works", put that in the ad — not "Experience seamless efficacy"
- **Front-load the hook**: First 125 characters are all that shows before "See more"
- **One insight per ad**: Each ad should be built on one clear audience insight, not five crammed together
- **Include the "why" for each ad**: Labelling which insight drives each ad helps the user understand why it should work and makes it easier to iterate
- **Variety matters**: 15 ads that all say the same thing in different words aren't useful. Spread across genuinely different angles and motivations from the research

## Related Skills

- **google-ad-copy**: For Google Search RSA assets
- **creative-testing**: To test ad concepts before writing full copy
- **icp-discovery**: To find the right audience before writing ads
- **landing-page-optimization**: To optimise the page ads drive traffic to

Feature Concept Testing

Validate feature concepts with real user insight before committing engineering resources. This workflow tests whether a feature idea solves a real problem, how users would actually use it, and what makes it valuable — so you build features people want instead of features that seem good in theory.

Use when: Requests to validate feature ideas, test product concepts, understand if users would use a feature, gauge interest in new functionality, or reduce risk of building unwanted features.

Download skill

View full skill

---
name: feature-concept-testing
version: 1.1.0
description: "Use this skill to test and validate new feature concepts before building them. Triggers include: requests to validate feature ideas, test product concepts, understand if users would use a feature, gauge interest in new functionality, or reduce risk of building unwanted features. Uses OriginalVoices Digital Twins (ask_twins) to understand whether a feature solves a real problem, how users would actually use it, what concerns they have, and whether they'd pay for it — preventing wasted engineering effort on features nobody wants."
---

# Feature Concept Testing Skill

## Overview

Validate feature concepts with real user insight before committing engineering resources. This workflow tests whether a feature idea solves a real problem, how users would actually use it, and what makes it valuable — so you build features people want instead of features that seem good in theory.

## Workflow Steps

### Step 1: Define the Feature Concept

Gather from the user:
- **Target audience**: Who is this feature for? (e.g. "Power users of project management software")
- **Feature description**: What does it do? (functionality, not technical implementation)
- **Problem it solves**: What user problem or need does this address?
- **Context**: Where does it fit? When would users encounter it?

### Step 2: Validate Problem & Feature Concept

One comprehensive research call covering problem validation, feature appeal, usage scenarios, value perception, and concerns.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Testing stimuli without bias.** When showing concepts, ads, taglines, or messaging options:
- Capture the **unaided gut reaction first** ("what's your first impression?") before asking for reasons.
- **Don't signal the hoped-for winner** — never reveal which option is the client's, and never ask "isn't this great?".
- When comparing options, **counterbalance the order** (don't always show A first) and always allow "neither / none of them".
- **Keep three questions separate:** comprehension ("what message do you take from this? who is it for?") → preference ("which do you prefer, and why?") → persuasion ("does it make you more or less likely to act?"). Never fuse them into one.

```
ask_twins(
  audience: "[detailed target audience]",
  questions: [
    "Walk me through the last time [problem area this feature addresses] tripped you up — what did you do about it?",
    "Tell me about the last time [problem] happened — what did it cost you in time or hassle, and how often does it come around?",
    "Here's an idea: [product] could let you [feature description]. In your own words, what would that actually do for someone like you?",
    "If that existed, what — if anything — would it change about how you handle [problem] today?",
    "Picture the next time [problem] comes up — where, if anywhere, would something like this fit into what you're already doing?",
    "Realistically, how often do you see yourself reaching for something like this — and what would push it up or down?",
    "How does this compare with the way you deal with [problem] now — where might it win, and where might your current way still be better?",
    "What would it take for something like this to feel like a natural part of your routine rather than an extra step?",
    "What would give you pause about relying on something like this — what could go wrong?",
    "How much would you actually miss something like this if it never existed? What makes you say that?",
    "Tell me about a time you switched to a different [product type] because of one specific feature — what made it worth it, and how does this compare?",
    "What's an add-on or upgrade you currently pay extra for in tools like this — and what made it worth paying for? Where would something like this fall?"
  ]
)
```

**Optional: Test Alternative Approaches**
```
ask_twins(
  audience: "[same audience]",
  questions: [
    "Here's a different way to tackle [problem]: [alternative approach]. What do you make of it next to the first idea — and when might one beat the other?",
    "Some people would want [Feature as described], others a stripped-back version, others neither — where do you land, and why?",
    "If you could design the perfect fix for [problem] from scratch, what would it look like?"
  ]
)
```

### Step 3: Analyse Feature Validation Insights

Review responses and assess:

**Problem-Solution Fit:**
- Is the problem real and painful?
- Does the feature actually solve it?
- Is it meaningfully better than alternatives?

**Usage Likelihood:**
- Would users actually use it?
- How often? Does it fit their workflow?

**Value Perception:**
- Must-have vs. nice-to-have?
- Would they pay for it?
- Would they switch products for it?

**Concerns & Risks:**
- What objections exist?
- What would prevent adoption?

### Step 4: Deliver Feature Validation Report

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers. Where findings have clear quantitative patterns, include a chart to visualize the distribution.

### Report sections:

**1. Executive Recommendation:**
- Build / Iterate / Don't Build
- Confidence level (high, medium, low)
- Key rationale (2-3 sentences)

**2. Problem Validation:**
- Is the problem real? (supporting quotes)
- How painful and frequent?
- Current workarounds and costs

**3. Feature Assessment:**
- Does it solve the problem? (user interpretation)
- Would users use it? (frequency, scenarios)
- Differentiation vs. existing solutions

**4. Value Analysis:**
- What value does it create?
- Must-have vs. nice-to-have sentiment. Include a chart showing the distribution of must-have vs. nice-to-have vs. wouldn't-use responses.
- Willingness to pay (if tested)

**5. Concerns & Objections:**
- Top concerns (with quotes)
- Adoption risks
- Dealbreakers vs. acceptable limitations

**6. Recommendations:**

If **Build**:
- Key requirements to meet user needs
- Potential MVP scope
- Critical success factors

If **Iterate**:
- What to change about the concept
- Alternative approaches to test

If **Don't Build**:
- Why it failed validation
- What to focus on instead

**7. Next Steps:**
- If validated: Prototyping, design, MVP scope
- If needs iteration: What to test next
- If not validated: Where to focus effort instead

## Feature Validation Framework

| Criterion | Strong Signal | Weak Signal |
|-----------|---------------|-------------|
| **Problem intensity** | "This is incredibly frustrating, happens constantly" | "Minor annoyance occasionally" |
| **Solution fit** | "This would solve my problem perfectly" | "Not sure this would help" |
| **Usage frequency** | "I'd use this daily/multiple times per week" | "Might use it once in a while" |
| **Workflow fit** | "This fits naturally into how I work" | "I'd have to remember to use it" |
| **Value perception** | "Game-changer" / "I'd pay for this" | "That's nice I guess" |
| **Adoption likelihood** | "I'd start using this immediately" | "I'd stick with current approach" |

**Strong validation**: Multiple strong signals — Build with confidence
**Moderate validation**: Mixed signals — Iterate, test again
**Weak validation**: Mostly weak signals — Don't build

## Red Flags (Don't Build)

- Users struggle to describe when/how they'd use the feature
- The problem isn't painful or frequent
- Users say "that's cool" but can't articulate why it's valuable
- Current workarounds are simple and acceptable
- Users say "I guess I might use it" instead of "I need this"
- Concerns and objections outweigh enthusiasm

## Green Lights (Build)

- Users describe specific, frequent scenarios where they'd use it
- The problem is painful and current workarounds are costly
- Users say "this would change how I work"
- Multiple users independently describe similar use cases
- Users ask when it will be available
- Feature fits naturally into existing workflows
- Users would pay for it or switch products for it

## Tips for Best Results

- **Test the problem first**: If the problem isn't real, the feature doesn't matter
- **Watch for polite enthusiasm**: "That sounds cool" does not equal "I would use this" does not equal "I need this"
- **Look for specificity**: Vague positivity means nothing; specific use cases indicate real intent
- **Frequency matters**: Features used daily can be simple; features used monthly must be instantly intuitive
- **Follow the energy**: Strong emotional reactions are more valuable than lukewarm interest
- **Test the simplest version**: Test the core concept first, not a feature with 10 capabilities

## Related Skills

- **deep-customer-research**: For broader understanding of user needs before testing specific features
- **icp-discovery**: To find the right audience for your feature
- **creative-testing**: To test how you communicate the feature
- **brand-messaging-validation**: To validate how the feature fits your brand story

Google Ad Copy

Generate Google Responsive Search Ad (RSA) assets informed by real audience insight. This workflow researches your target audience first, then creates RSA-ready headlines and descriptions that speak to what real people actually care about — ensuring higher relevance scores and better performance.

Use when: Requests to create Google Search or Display ad campaigns, write Responsive Search Ad (RSA) headlines and descriptions, refresh underperforming Google ad copy, or generate audience-informed ad variations for A/B testing.

Download skill

View full skill

---
name: google-ad-copy
version: 1.1.0
description: "Use this skill to generate Google ad copy grounded in real audience insight. Triggers include: requests to create Google Search or Display ad campaigns, write Responsive Search Ad (RSA) headlines and descriptions, refresh underperforming Google ad copy, or generate audience-informed ad variations for A/B testing. Uses OriginalVoices Digital Twins (ask_twins) to deeply research the target audience's language, preferences, concerns, and experiences, then generates audience-informed RSA assets rooted in what real people actually said — not AI guesswork."
---

# Google Ad Copy Skill

## Overview

Generate Google Responsive Search Ad (RSA) assets informed by real audience insight. This workflow researches your target audience first, then creates RSA-ready headlines and descriptions that speak to what real people actually care about — ensuring higher relevance scores and better performance.

## Workflow Steps

### Step 1: Gather Inputs

Required:
- **Target audience**: Who are these ads for? (e.g. "Small business owners aged 30-50 in the US looking for accounting software")
- **Product/offer**: Description or link to product page / landing page
- **Primary keywords**: 2-3 main search terms you're targeting (for relevance and Quality Score)

Optional:
- **Landing page URL**: To ensure message match
- **Unique value prop**: What makes this different from competitors

### Step 2: Audience Research

Use `ask_twins` to understand the audience's world — their language, pain points, and search behavior. This informs ad copy that resonates.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Capturing the audience's own words.** This skill generates copy from what real people say, so the research must surface their *own language*, not abstract opinions — generic questions produce generic copy. Include at least two moment-anchored questions built to elicit verbatim phrasing (e.g. "The last time you looked for [X], what did you actually type or say?", "How would you describe [problem] in your own words, to a friend?"). When you later test copy, replace "would you click / stop scrolling?" (demand effect + inflated intent) with "what would you do next, if anything?" — and ask "what do you take this to be offering?" (comprehension) before whether they like it.

The search-language question below (what did you actually type into Google) is the model verbatim question for this skill — it gives you the exact keywords to drop into headlines. Keep it grounded in a real, recalled search rather than a hypothetical.

```
ask_twins(
  audience: "[detailed target audience description]",
  questions: [
    "Walk me through the last time [problem/category] came up for you — what did you actually do about it, and how did it go?",
    "Tell me about the last time you chose a [product category] — what tipped the decision, and what did you rule out?",
    "The last time you went looking for [solution] online, what exact words did you type into Google? Give me the actual phrases, however rough.",
    "Think about the last Google ad you actually clicked on — or one you deliberately skipped — what was it about, and what made you act?",
    "When you think about trying a new [product/solution], what pulls you toward it — and what makes you hesitate?",
    "How would you describe [problem/solution] in your own words — the way you'd say it to a friend?"
  ]
)
```

### Step 3: Extract Key Insights

Analyze responses and identify:
- **Search language**: Exact phrases they'd type into Google (use these in headlines)
- **Pain points**: Their top frustrations (lead with these)
- **Value drivers**: What matters most (benefits to emphasize)
- **Trust factors**: What builds confidence (use in descriptions)
- **Attention triggers**: What makes them click vs scroll past

### Step 4: Generate RSA Assets

Create **15 headlines** and **4 descriptions** using the audience insights. Follow Google RSA format strictly.

**Headlines (30 characters max each):**
- Include primary keyword in at least 3 headlines (for relevance)
- Vary approaches: pain point, benefit, question, social proof, urgency
- Mix short (15-20 chars) and long (25-30 chars) for better combinations
- Use audience's exact language where possible
- Every headline must be unique and work independently

**Descriptions (90 characters max each):**
- Description 1: Lead benefit + CTA
- Description 2: Address objection + unique value
- Description 3: Social proof + CTA
- Description 4: Alternative angle (pain point or different benefit)

**Format each asset with:**
- Character count verification (must be exact)
- Audience insight it's based on
- Strategic note (e.g., "keyword headline", "pain point hook")

### Step 5: Structure for Google Ads

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for individual ad variations, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers.

Deliver assets in this format:

**HEADLINES (15 required)**

| # | Headline | Chars | Insight | Strategy |
|---|----------|-------|---------|----------|
| H1 | [headline text] | 28 | [insight] | Keyword + benefit |
| H2 | [headline text] | 22 | [insight] | Pain point |
| ... | ... | ... | ... | ... |

**DESCRIPTIONS (4 required)**

| # | Description | Chars | Insight | Strategy |
|---|-------------|-------|---------|----------|
| D1 | [description text] | 87 | [insight] | Benefit + CTA |
| D2 | [description text] | 89 | [insight] | Objection handling |
| D3 | [description text] | 84 | [insight] | Social proof |
| D4 | [description text] | 90 | [insight] | Alternative angle |

**PINNING RECOMMENDATIONS:**
- Pin H1 to position 1 only if it must always show (e.g., brand name + keyword)
- Otherwise let Google optimize combinations
- Never pin more than 2-3 assets total

### Step 6: Validate Top Performers (Optional)

Test 3-5 strongest headline + description combinations with the audience:

```
ask_twins(
  audience: "[same target audience]",
  questions: [
    "You're searching Google for '[search term]' and this ad comes up: Headline: '[H1]' Description: '[D1]'. In your own words, what do you take it to be offering — and what would you do next, if anything?",
    "Between these ads, which feels most like a match for what you searched? A: '[H2 + D2]' B: '[H5 + D3]' C: '[H8 + D1]'. What makes the other two miss?"
  ]
)
```

Update assets based on feedback before finalizing.

### Step 7: Quality Score Checklist

Before delivering, verify:
- Primary keyword appears in at least 3 headlines
- All headlines ≤30 characters (Google rejects otherwise)
- All descriptions ≤90 characters
- Headlines are diverse (not 15 variations of the same thing)
- Descriptions work with any headline combination
- Copy matches landing page messaging
- Clear CTA in at least 2 descriptions
- Assets use audience's actual language from research

## RSA Best Practices

**Headline Strategy:**
- 3-5 headlines with primary keyword (for relevance)
- 3-5 headlines addressing different pain points
- 2-3 benefit-focused headlines
- 2-3 social proof or trust-building headlines
- 1-2 question or curiosity headlines

**Description Strategy:**
- Lead with the strongest benefit (based on research)
- Address the top objection from audience research
- Include social proof or trust signals
- Provide alternative angle for variety

**Character Optimization:**
- Shorter headlines (15-22 chars) pair well with longer ones
- Max out description length (85-90 chars) to provide context
- Front-load important words (first 15 chars of headlines are critical)

**Keyword Integration:**
- Include keywords naturally, not stuffed
- Use exact phrases from the "what did you actually type into Google" responses
- Match search intent (don't promise what landing page doesn't deliver)

## Common Pitfalls to Avoid

- **Keyword stuffing**: Google's algorithm detects this and lowers Quality Score
- **Generic headlines**: "Best Software" performs worse than specific benefits from research
- **Over-pinning**: Pinning too many assets limits Google's optimization
- **Identical variations**: "Get Started Today" and "Start Today" waste asset slots
- **Ignoring character limits**: Google truncates or rejects — verify every asset
- **Feature dumping**: Lead with benefits and outcomes the audience cares about
- **No CTA**: Every description should guide the user to act

## Output Checklist

Final deliverable includes:
1. 15 unique headlines (all ≤30 chars)
2. 4 unique descriptions (all ≤90 chars)
3. Character count verified for every asset
4. Table format ready to copy into Google Ads
5. Insight source documented for each asset
6. Pinning recommendations (if any)
7. Quality Score checklist completed
8. Top 3-5 predicted best-performing combinations highlighted

## Related Skills

- **facebook-ad-copy**: For Facebook/Instagram ad copy
- **creative-testing**: To test ad concepts before writing full copy
- **icp-discovery**: To find the right audience before writing ads
- **landing-page-optimization**: To optimise the page ads drive traffic to

ICP & Target Audience Discovery

Identify your ideal target audience by evaluating your product, concept, or idea across multiple demographic segments simultaneously. Rather than making assumptions about who to reach, this workflow presents your offering to 5+ distinct audiences and reveals who resonates most — and precisely why. The result is a ranked segment list, targeted messaging insights for your primary ICP, and comprehensive feedback across all audiences.

Use when: A user doesn't know who to target, is entering a new market, launching a new product, or wants to compare audience receptivity across segments before committing budget.

Download skill

View full skill

---
name: icp-discovery
version: 1.1.0
description: "Use this skill to discover the ideal customer profile (ICP) and target audience for a product, concept, or idea by testing it across multiple demographic segments. Triggers include: requests to find the best audience for a product, identify an ICP, discover who a product resonates with most, explore which demographic segments respond best to a concept, validate target market assumptions, find product-market fit, or uncover unexpected audience opportunities. Also use when a user doesn't know who to target, is entering a new market, launching a new product, or wants to compare audience receptivity across segments before committing budget. Uses OriginalVoices Digital Twins (ask_twins) to test the same product/concept across 5+ distinct demographic segments simultaneously, then ranks segments by fit and extracts messaging guidance for the top ICP."
---

# ICP & Target Audience Discovery Skill

## Overview

Identify your ideal target audience by evaluating your product, concept, or idea across multiple demographic segments simultaneously. Rather than making assumptions about who to reach, this workflow presents your offering to 5+ distinct audiences and reveals who resonates most — and precisely why. The result is a ranked segment list, targeted messaging insights for your primary ICP, and comprehensive feedback across all audiences.

## Workflow Steps

### Step 1: Understand the Product/Concept

Request that the user either:
- **Describe** their product, concept, or idea in a few sentences, or
- **Share a link** to a landing page, product page, or pitch deck

Extract a clear 1-2 sentence summary describing what it is and what problem it addresses. This summary becomes the foundation for questions presented to each segment.

### Step 2: Select Candidate Segments

Based on the product/concept, propose 5-7 distinct demographic segments representing a diverse cross-section of potential audiences. Segments should vary across multiple dimensions — not just age.

**Segment design principles:**
- Vary by age, gender, life stage, interests, income, and lifestyle
- Include at least one "unexpected" segment the user might not have considered
- Ensure segments don't overlap heavily

**Example segment set for a meal planning app:**
1. Busy working parents aged 30-45 with young children
2. Health-conscious young adults aged 20-30 living in cities
3. Men aged 35-45 interested in health and fitness
4. University students aged 18-24
5. Women aged 45-60

### Step 3: Build the Question Set

Design 4-6 questions that test receptivity, relevance, emotional response, and purchase intent. These same questions will be asked to every segment for direct comparison.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Testing stimuli without bias.** When showing concepts, ads, taglines, or messaging options:
- Capture the **unaided gut reaction first** ("what's your first impression?") before asking for reasons.
- **Don't signal the hoped-for winner** — never reveal which option is the client's, and never ask "isn't this great?".
- When comparing options, **counterbalance the order** (don't always show A first) and always allow "neither / none of them".
- **Keep three questions separate:** comprehension ("what message do you take from this? who is it for?") → preference ("which do you prefer, and why?") → persuasion ("does it make you more or less likely to act?"). Never fuse them into one.

**Core questions (adapt to the specific product/concept):**

```
questions: [
  "I'd like to tell you about [product/concept: 1-2 sentence description]. Setting aside whether it's for you, what's your honest first impression?",
  "Tell me about the last time [the problem this solves] came up in your life — what happened, and how much did it matter to you?",
  "The last time you tried something new to deal with [the problem this solves], what made you give it a go — and what would hold you back with something like this?",
  "What would something like this have to get right to earn a place in your life — and what would make it not worth it?",
  "If you were describing this to a friend, what would you say it is and who it's for?",
  "What's the first thing you'd want to know before deciding whether something like this is for you?"
]
```

### Step 4: Run Segment-by-Segment Research

Make a separate `ask_twins` call for each segment, using the same questions every time. This ensures clean, comparable data across all segments.

```
# Segment 1
ask_twins(
  audience: "Busy working parents aged 30-45 with young children in the US",
  questions: [
    "I'd like to tell you about [product description]. Setting aside whether it's for you, what's your honest first impression?",
    "Tell me about the last time [problem] came up in your life — what happened, and how much did it matter to you?",
    "The last time you tried something new to deal with [problem], what made you give it a go — and what would hold you back with something like this?",
    "What would something like this have to get right to earn a place in your life — and what would make it not worth it?",
    "If you were describing this to a friend, what would you say it is and who it's for?",
    "What's the first thing you'd want to know before deciding whether something like this is for you?"
  ]
)

# Segment 2
ask_twins(
  audience: "Health-conscious Gen Z adults aged 20-28 living in cities",
  questions: [
    # Same questions as above
  ]
)

# ... Repeat for all segments
```

### Step 5: Score & Rank Each Segment

For each segment, evaluate responses across these dimensions:

| Dimension | What to Look For |
|-----------|-----------------|
| **Interest level** | Enthusiasm vs. indifference. Did they light up or shrug? |
| **Problem relevance** | Is the problem real and active in their life, or theoretical? |
| **Purchase intent** | Would they actually try/buy it, or is it a polite "maybe"? |
| **Emotional resonance** | Did it trigger a strong emotional response (excitement, relief, hope)? |
| **Objection severity** | Are their hesitations minor (price, timing) or fundamental (don't need it)? |
| **Natural language fit** | Did they describe the product in a way that could become marketing copy? |
| **Word-of-mouth potential** | Could they easily explain it to a friend? Did their description sound compelling? |

**Rank all segments from strongest to weakest fit** based on a holistic assessment of these dimensions.

### Step 6: Deep-Dive on Top ICP

For the #1 ranked segment, extract detailed messaging guidance:

- **Why they care**: The core motivation driving their interest
- **Their language**: Exact words and phrases they used to describe the product and problem
- **Key benefit**: The single benefit that resonated most
- **Primary objection**: The main thing holding them back — and how to address it
- **Emotional hook**: The feeling that drives them (relief, excitement, validation, fear)
- **How they'd describe it**: Their "friend description" becomes the basis for positioning
- **Trust signals needed**: What they said they'd want to know or see before trying it
- **Recommended messaging angle**: A 1-2 sentence positioning statement built from their actual words

### Step 7: Deliver the Full Report

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for individual segments/subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers. Where findings have clear quantitative patterns, include a chart to visualize the distribution.

### Report sections:

**1. Executive Summary**
- The #1 target audience and why, in 2-3 sentences
- The biggest surprise from the research

**2. Segment Rankings**

| Rank | Segment | Interest | Relevance | Intent | Key Insight |
|------|---------|----------|-----------|--------|-------------|
| 1 | [Top segment] | High | High | High | [One-line insight] |
| 2 | [Second segment] | ... | ... | ... | ... |
| ... | ... | ... | ... | ... | ... |

**3. Top ICP Deep-Dive**
- Detailed messaging guidance (from Step 6)
- 3-5 direct quotes from Digital Twins that capture why this segment is the best fit
- Recommended positioning statement
- Suggested ad/marketing angles
- Include a chart comparing segment fit scores across all segments.

**4. Segment-by-Segment Breakdown**
For each segment (including lower-ranked ones):
- **Overall fit**: High / Medium / Low
- **What resonated**: What they liked or responded to
- **What didn't land**: Where the product/concept fell flat
- **Key quote**: One representative quote
- **Verdict**: Should you target this segment? (Yes / Secondary / No — with reasoning)

**5. Messaging Matrix**

| Segment | Lead Message | Emotional Hook | Key Objection | Recommended? |
|---------|-------------|----------------|---------------|-------------|
| [Segment 1] | ... | ... | ... | Primary |
| [Segment 2] | ... | ... | ... | Secondary |
| [Segment 3] | ... | ... | ... | Deprioritise |
| ... | ... | ... | ... | ... |

**6. Strategic Recommendations**
- Primary ICP to target and the messaging angle to lead with
- Secondary audiences worth testing with smaller budget
- Segments to avoid and why
- Suggested next steps (e.g. "Run the Google Ad Copy workflow targeting your primary ICP")

## Tips for Best Results

- **Test at least 5 segments**: Fewer than 5 doesn't give enough contrast. 6-7 is ideal
- **Include a wildcard segment**: Always test one audience you wouldn't obviously target — surprises happen
- **Keep questions identical across segments**: Consistency is what makes comparison valid
- **Don't oversell the product**: Describe it factually in 1-2 sentences. Let the audience's genuine reaction tell the story
- **Watch for "polite interest" vs. real excitement**: Some audiences will say "sounds nice" without any real intent. Look for strong emotional language, specific use cases they imagine, and unprompted enthusiasm
- **The "friend description" question is gold**: How someone would explain your product to a friend is often better positioning than anything a marketer would write
- **Low-ranking segments are still valuable**: Understanding who doesn't care — and why — is just as strategic as finding who does

## Related Skills

- **deep-customer-research**: Once you've found your ICP, go deeper on their needs
- **facebook-ad-copy**: Generate ad copy targeting your top ICP
- **google-ad-copy**: Generate search ads targeting your top ICP
- **creative-testing**: Test creative concepts with your discovered audience

Idea Validation

Validate a business idea, product concept, or value proposition with real audience feedback before you invest time or money building it. This workflow uses OriginalVoices Digital Twins to test whether the problem is real, how people deal with it today, and whether your proposed solution would genuinely change their behaviour — so you get a clear signal before you commit.

Use when: Requests to validate a startup idea, test a product concept, check if an idea has legs, assess market demand, validate a value proposition, or get honest feedback on a new business concept.

Download skill

View full skill

---
name: idea-validation
version: 1.1.0
description: "Use this skill to validate a business idea, product concept, or value proposition with real audience feedback before investing time or money. Triggers include: requests to validate a startup idea, test a product concept, check if an idea has legs, assess market demand, validate a value proposition, or get honest feedback on a new business concept. Uses OriginalVoices Digital Twins (ask_twins) to understand whether the problem is real, how painful it is, how people solve it today, and whether the proposed idea would genuinely change their behaviour — delivering a clear Go / Pivot / Kill verdict with evidence."
---

# Idea Validation

## Overview

Validate a business idea, product concept, or value proposition with real audience feedback before you invest time or money building it. This workflow uses OriginalVoices Digital Twins to test whether the problem is real, how people deal with it today, and whether your proposed solution would genuinely change their behaviour — so you get a clear signal before you commit.

## Workflow Steps

### Step 1: Gather the Idea

Collect from the user:

- **The idea**: What is the product, service, or concept? (A short description is fine — even a one-liner)
- **The problem it solves**: What pain point or need does it address?
- **Target audience**: Who is this for? Be as specific as possible — demographics, interests, lifestyle, situation
- **How it works** (optional): Any details on the proposed solution, pricing model, or delivery method
- **What exists today** (optional): Known competitors or alternatives the audience might already use

### Step 2: Validate with Digital Twins

Test the idea with the target audience using 10-12 questions that progress from problem validation through to purchase intent. This sequence is deliberate — it establishes whether the problem is real before introducing the solution.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Testing stimuli without bias.** When showing concepts, ads, taglines, or messaging options:
- Capture the **unaided gut reaction first** ("what's your first impression?") before asking for reasons.
- **Don't signal the hoped-for winner** — never reveal which option is the client's, and never ask "isn't this great?".
- When comparing options, **counterbalance the order** (don't always show A first) and always allow "neither / none of them".
- **Keep three questions separate:** comprehension ("what message do you take from this? who is it for?") → preference ("which do you prefer, and why?") → persuasion ("does it make you more or less likely to act?"). Never fuse them into one.

```
ask_twins(
  audience: "[detailed target audience description]",
  questions: [
    "When it comes to [problem area], walk me through the last time it got in your way — what happened, and how often does that come up?",
    "How do you currently handle [problem]? Walk me through what you do today.",
    "Thinking about how you handle it now — what works fine, and what do you wish was different?",
    "Tell me about a time you went looking for a better way to deal with [problem] — what did you find, and why did you stick with it or drop it?",
    "Compared with the other hassles in your week, where does [problem] rank — and what makes it that big or small a deal?",
    "I'd like your honest first reaction to an idea: [describe the idea clearly and concisely, in plain language]. What goes through your mind?",
    "If this existed, what — if anything — would it change in your day-to-day? And where might it fall short for you?",
    "What questions or concerns come to mind when you hear about this idea? What would hold you back from trying it?",
    "Tell me about the last time you switched from a way of doing something you were used to — what finally tipped you? What would it take here?",
    "Think about what you already spend on [problem] today — in money, time, or workarounds. Against that, what would something like this be worth to you?",
    "When did you last tell someone about a product you'd found — what made it worth mentioning? Who, if anyone, comes to mind for this one?",
    "If you could change one thing about this idea to make it perfect for you, what would it be?"
  ]
)
```

**Important:** Present the idea in plain language, the way you'd explain it to a friend. Avoid marketing speak or inflated claims — you want honest reactions, not polite nods.

### Step 3: Assess the Signals

Analyse responses across five validation dimensions:

**Problem Validation**
- Is the problem real and frequent, or theoretical?
- How painful is it (mild annoyance vs. genuine frustration)?
- Are people actively looking for solutions?

**Solution Fit**
- Did the idea spark genuine interest or polite indifference?
- Does it solve the problem as the audience experiences it?
- Did anyone say "I need this" or "where can I get this"?

**Switching Willingness**
- Would they actually change their current behaviour?
- What barriers to adoption came up?
- How entrenched are current habits or solutions?

**Value Perception**
- Do price expectations align with a viable business model?
- Is it seen as a "must-have" or "nice-to-have"?
- Would they pay for it, or only use it if free?

**Word of Mouth Potential**
- Would they tell others about it?
- Can they articulate the value clearly?
- Who would they recommend it to?

### Step 4: Deliver the Validation Verdict

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers. Where findings have clear quantitative patterns, include a chart to visualize the distribution.

### Report sections:

**1. Verdict: Go / Pivot / Kill**

| Verdict | When to use |
|---------|------------|
| **Go** | Problem is real and painful, solution sparked genuine excitement, audience would pay and switch |
| **Pivot** | Problem is real but the proposed solution missed the mark — the audience pointed to a better direction |
| **Kill** | Problem isn't painful enough, audience is indifferent to the solution, or insurmountable barriers exist |

**2. Evidence Summary**

Include a chart showing signal strength across validation dimensions.

| Dimension | Signal | Strength (Strong / Mixed / Weak) |
|-----------|--------|----------------------------------|
| Problem exists | [What the audience said] | [Rating] |
| Problem is painful | [Pain level and frequency] | [Rating] |
| Solution resonates | [Gut reactions] | [Rating] |
| Would switch | [Switching willingness] | [Rating] |
| Would pay | [Price expectations] | [Rating] |
| Would recommend | [Word of mouth signals] | [Rating] |

**3. Strongest Signals**
- What resonated most (with supporting quotes)
- The clearest signs of demand or excitement

**4. Biggest Risks**
- What concerned the audience most
- Barriers to adoption they raised
- Gaps between the idea and what they actually need

**5. Audience-Suggested Improvements**
- Changes the audience said would make the idea better
- Features or aspects they specifically asked for
- What would turn it from "interesting" to "essential"

**6. Recommended Next Steps**

For **Go**:
- Priority features to build first (based on what the audience valued most)
- Messaging angles that resonated
- Audience segments that showed strongest interest

For **Pivot**:
- What direction the audience pointed toward
- Which elements to keep vs. rethink
- Suggested reframing of the idea

For **Kill**:
- Why the idea doesn't have sufficient demand
- Whether a different audience might respond better
- Salvageable elements worth exploring elsewhere

## Tips for Best Results

- **Test the problem before the solution**: The question sequence is deliberate — if the problem isn't real, the solution doesn't matter
- **Present the idea simply**: If you can't explain it in 2-3 sentences, the audience won't get it either. That's a signal in itself
- **Listen for energy, not politeness**: "That's interesting" is lukewarm. "Where can I get this?" is validation
- **Pain level matters**: A real problem that's mildly annoying won't drive behaviour change. Look for genuine frustration
- **Switching cost is the hidden killer**: Even great ideas fail if the cost of changing behaviour is too high
- **Price expectations reveal truth**: If the audience expects it to be free, they don't value it enough to sustain a business

## Validation Signal Guide

### Strong Positive Signals
- Audience describes the problem with emotion and specific examples
- Gut reaction to the idea is enthusiastic, not just polite
- They ask "when can I get this?" or "does this exist?"
- They immediately think of someone they'd tell about it
- Price expectations align with a viable model
- They describe current workarounds that are painful or expensive

### Warning Signs (Consider Pivoting)
- Problem is real but the solution doesn't quite fit how they experience it
- Interest is conditional — "I'd use it if..." with significant conditions
- They like the idea but wouldn't pay for it
- They can't articulate who they'd recommend it to
- Current solutions are "good enough" even if imperfect

### Red Flags (Consider Killing)
- Audience struggles to relate to the problem
- Gut reaction is indifferent or confused
- "That's interesting" without follow-up enthusiasm
- They wouldn't switch from their current approach
- Price expectation is zero or far below viability
- No one they'd recommend it to
- The problem exists but isn't painful enough to drive action

## Common Pitfalls to Avoid

- **Confirmation bias**: Don't cherry-pick positive responses and ignore concerns. The warnings are the most valuable part
- **Solution-first thinking**: If you skip problem validation and jump straight to "do you like my idea?", you'll get polite answers, not honest ones
- **Ignoring switching costs**: People may love the idea in theory but never change their actual behaviour
- **Confusing interest with intent**: "That sounds cool" is not the same as "I would pay for that"
- **Testing with the wrong audience**: A great idea tested with the wrong people will look like a bad idea. Be specific about who you're testing with
- **Over-describing the idea**: The more you explain and sell, the less honest the reaction. Keep it simple and let the audience respond naturally

Landing Page Optimization

Optimize your landing page for conversion using real audience insight. This workflow researches how your target audience evaluates landing pages — what grabs attention, what creates trust, what triggers doubt, and what drives action — then uses those insights to craft landing page elements that convert.

Use when: Requests to improve conversion rates, test landing page copy, validate value propositions, optimize hero sections, test different messaging angles, or reduce bounce rates.

Download skill

View full skill

---
name: landing-page-optimization
version: 1.1.0
description: "Use this skill to optimize landing page messaging, structure, and conversion elements using real audience insight. Triggers include: requests to improve conversion rates, test landing page copy, validate value propositions, optimize hero sections, test different messaging angles, or reduce bounce rates. Uses OriginalVoices Digital Twins (ask_twins) to understand what resonates with visitors, what creates confusion or doubt, and what drives action — then generates optimized landing page variants grounded in real user responses."
---

# Landing Page Optimization Skill

## Overview

Optimize your landing page for conversion using real audience insight. This workflow researches how your target audience evaluates landing pages — what grabs attention, what creates trust, what triggers doubt, and what drives action — then uses those insights to craft landing page elements that convert.

## Workflow Steps

### Step 1: Gather Context

Collect from the user:
- **Current landing page**: URL or description of what the page currently says/shows
- **Target audience**: Who is this page for? (e.g. "Startup founders looking for project management tools")
- **Primary goal**: What action should visitors take? (sign up, purchase, book demo, etc.)

### Step 2: Research Landing Page Preferences & Conversion Drivers

One comprehensive research call covering current messaging evaluation, problem awareness, headline testing, trust signals, and CTA preferences.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Capturing the audience's own words.** This skill generates copy from what real people say, so the research must surface their *own language*, not abstract opinions — generic questions produce generic copy. Include at least two moment-anchored questions built to elicit verbatim phrasing (e.g. "The last time you looked for [X], what did you actually type or say?", "How would you describe [problem] in your own words, to a friend?"). When you later test copy, replace "would you click / stop scrolling?" (demand effect + inflated intent) with "what would you do next, if anything?" — and ask "what do you take this to be offering?" (comprehension) before whether they like it.

**Testing stimuli without bias.** When showing headlines, value props, or page variants:
- Capture the **unaided gut reaction first** ("what's your first impression?") before asking for reasons.
- **Don't signal the hoped-for winner** — don't ask "does this grab your attention?"; ask "what, if anything, would you do after landing here?".
- When comparing options, **counterbalance the order** and always allow "neither".
- **Keep three questions separate:** comprehension ("what do you think this page is offering, and who for?") → preference ("which version works better for you, and why?") → persuasion ("what would make you act, or hesitate?"). Never fuse them.

```
ask_twins(
  audience: "[detailed target audience]",
  questions: [
    "When you land on a website for a [product/service type], what do you look for first, and what makes you stay or leave?",
    "Think about the last time you went looking for a [product/service type] — what set you off looking, and what were you hoping to find?",
    "Here are two openings for a page like this — A: '[Test Headline A]' and B: '[Test Headline B]'. What's your first impression of each, and which — if either — makes you want to read on? Why?",
    "A page opens with: '[Value prop angle 1]'. What's your first impression, and what would you do next, if anything?",
    "Here are two ways of putting it — A: '[Benefit-focused message]' and B: '[Problem-focused message]'. What does each one say to you, and which lands better — or neither? Why?",
    "Think about the last time you signed up for or bought a new [product category] — what made you hesitate, if anything, and what tipped you over the line?",
    "Think of the last time a website earned enough of your trust to [take desired action] — what did it do, and where do most fall short?",
    "When you're deciding whether to trust a [product/service] you've just found, what actually makes a difference to you — and what do you ignore or distrust?",
    "Here are two buttons — A: '[CTA option 1]' and B: '[CTA option 2]'. What does each make you expect will happen if you click, and which — if either — feels more like something you'd do? Why?",
    "Think about the last [product category] you were seriously considering — did you want to jump straight in and try it, talk to someone first, or something else? What drove that?",
    "When you reach a '[primary CTA]' button, what's going through your mind in the moment before you click — or decide not to?",
    "Think about the last sign-up form that made you give up on it — what did it ask for that was a step too far? How does being asked for [list of fields] compare?"
  ]
)
```

**If there's an existing landing page to audit, include these additional questions:**
```
    "Here's the headline the page opens with: '[current headline]'. What's your first impression, and what would you do next, if anything?",
    "The page says '[current value proposition]'. In your own words, what do you take this to be offering, and who for? Where did you have to guess?",
    "What questions would you have after reading this page? What would you want to know more about?"
```

### Step 3: Extract Key Insights

Analyze responses and identify:

**Hero Section Insights:**
- What grabs attention vs. what's ignored
- Most compelling headline angle
- Clearest value proposition framing
- What needs to be communicated immediately

**Trust & Credibility:**
- Which trust elements matter most
- What testimonial styles resonate
- Key objections that need addressing

**CTA & Conversion:**
- Optimal CTA copy and positioning
- Form friction points to reduce
- Optimal offer structure (trial, demo, immediate purchase)

**Content Hierarchy:**
- What information they need first
- What must be above the fold
- What can wait for below the fold

### Step 4: Generate Optimized Landing Page Variants

Create 3-5 landing page variants, each grounded in different insights from research.

**For each variant, deliver:**
- **Headline** — Hook that grabs attention (based on what resonated)
- **Subheadline** — Clarifies value prop (addresses top questions/concerns)
- **Value propositions** — 3-5 key benefits (in their language)
- **Social proof** — Type and positioning (testimonials, logos, stats)
- **Primary CTA** — Button copy and offer (based on what drives action)
- **Trust elements** — What to include above fold (security, guarantees, etc.)
- **Key sections** — Recommended page structure

**Label each variant with:**
- **Target insight** — Which research finding drives this variant
- **Best for** — What type of visitor or use case
- **Key hypothesis** — What makes this variant different and why it should work

**Spread variants across different angles:**
- Problem-focused (lead with pain point)
- Benefit-focused (lead with outcome)
- Trust-first (emphasize credibility and social proof)
- Direct/minimal (for audiences who prefer no-fluff)
- Story-driven (if research shows they value context)

### Step 5: Validate Top Variants (Optional)

Test 2-3 strongest variants back with the audience:

```
ask_twins(
  audience: "[same audience]",
  questions: [
    "You land on a page opening with '[Variant A headline]' — '[Variant A subheadline]'. What's your first impression, and what would you do next, if anything?",
    "Here are two openings — A: '[Variant A hero]' and B: '[Variant B hero]'. What does each say to you, and which — if either — pulls you in more? Why?",
    "Based on '[Variant summary]', what do you take this product to do? Where, if anywhere, did it leave you guessing?",
    "After reading this, what would you do next — and if you're not ready to [take desired action], what's holding you back?"
  ]
)
```

Update variants based on feedback before finalizing.

### Step 6: Deliver Optimization Report

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers. Where findings have clear quantitative patterns, include a chart to visualize the distribution.

### Report sections:

**1. Executive Summary:**
- Top recommended variant and expected impact
- Key finding (most important insight from research)

**2. Key Insights from Research:**
- What messaging resonates vs. what creates confusion
- Top objections and concerns to address. Include a chart ranking objections by frequency.
- Trust signals that matter most
- Optimal CTA and offer structure

**3. Landing Page Variants (3-5 full variants):**
- Complete copy for each (headline, subheadline, value props, CTA, etc.)
- Insight driving each variant
- When to use each (audience segment, traffic source, etc.)

**4. A/B Testing Plan:**
- Which variants to test first
- What elements to test (headline, CTA, social proof, etc.)
- Success metrics to track

**5. Quick Wins:**
- Simple changes that can be implemented immediately

**6. Common Pitfalls to Avoid:**
- Based on what the audience reacted negatively to

## Landing Page Element Guidelines

| Element | Best Practice |
|---------|---------------|
| **Headline** | Lead with strongest hook from testing; 6-12 words; avoid jargon |
| **Subheadline** | Clarify value prop; address top question; 10-20 words |
| **Value props** | Use audience's language; focus on outcomes; 3-5 bullets |
| **CTA** | Action-oriented; address friction points; repeat 2-3 times |
| **Social proof** | Above fold; match format that resonates |
| **Trust elements** | Address top concerns; position near CTA |

## Tips for Best Results

- **Clarity beats cleverness**: If they struggle to understand what the product does, clever copy won't convert
- **Address objections explicitly**: If research surfaced concerns, address them on the page
- **Use their language**: Use exact phrases from Digital Twin responses
- **Hierarchy matters**: Put the most important information (based on research) above the fold
- **Don't bury the CTA**: If they're ready to act, make it easy
- **Test drastically different variants**: Test genuinely different messaging angles, not just headline variations
- **Optimize for your audience**: Let research guide you, not generic "best practices"

## Related Skills

- **deep-customer-research**: For deeper audience understanding before optimization
- **creative-testing**: To test specific creative concepts or variants
- **brand-messaging-validation**: If brand positioning needs validation first
- **facebook-ad-copy**: To write ads that drive traffic to the optimized page

Product Description Optimization

Optimize existing product descriptions for ecommerce using real audience insight. This workflow researches your target audience to understand what matters when they're buying, what information they need to make a decision, what language resonates, and what builds trust — then uses those insights to refine and improve your existing product descriptions for better conversion.

Use when: Requests to improve product descriptions, optimize ecommerce copy, increase conversion rates, refine product pages, make product descriptions resonate with target audiences, or test product messaging.

Download skill

View full skill

---
name: product-description-optimization
version: 1.1.0
description: "Use this skill to optimize existing product descriptions for ecommerce using real audience insight. Triggers include: requests to improve product descriptions, optimize ecommerce copy, increase conversion rates, refine product pages, make product descriptions resonate with target audiences, or test product messaging. Uses OriginalVoices Digital Twins (ask_twins) to understand what matters to buyers, what information they need, what language resonates, and what drives purchase decisions — then optimizes existing descriptions to convert better."
---

# Product Description Optimization Skill

## Overview

Optimize existing product descriptions for ecommerce using real audience insight. This workflow researches your target audience to understand what matters when they're buying, what information they need to make a decision, what language resonates, and what builds trust — then uses those insights to refine and improve your existing product descriptions for better conversion.

## Workflow Steps

### Step 1: Gather Product & Description

Collect from the user:
- **Product name**: What is being sold
- **Product category**: Type of product (e.g. "skincare", "home decor", "tech accessories", "fitness equipment")
- **Target audience**: Who is this product for? (e.g. "Women 25-40 interested in clean beauty", "Remote workers setting up home offices")
- **Current product description**: The existing description to optimize (copy the full text)
- **Product price point** (optional): Budget, mid-range, premium
- **Key product features**: Main features or benefits (if not clear from description)
- **Optimization goal**: What should improve? (conversion rate, clarity, trust, differentiation, SEO, etc.)

### Step 2: Test Existing Description with Audience

One comprehensive research call to get specific feedback on the existing product description and understand how to optimize it.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

**Capturing the audience's own words.** This skill generates copy from what real people say, so the research must surface their *own language*, not abstract opinions — generic questions produce generic copy. Include at least two moment-anchored questions built to elicit verbatim phrasing (e.g. "The last time you looked for [X], what did you actually type or say?", "How would you describe [problem] in your own words, to a friend?"). When you later test copy, replace "would you click / stop scrolling?" (demand effect + inflated intent) with "what would you do next, if anything?" — and ask "what do you take this to be offering?" (comprehension) before whether they like it.

```
ask_twins(
  audience: "[detailed target audience]",
  questions: [
    "Here's a product description for [product name]: '[paste full current description]'. Read it once — what's your first, gut reaction?",
    "In your own words, what do you take this product to be, and who do you think it's for? Where did you have to guess or reread?",
    "What in that description landed for you, and what fell flat or left you cold?",
    "Think about the last time you bought a [product category] product — what did you need to know before you were comfortable buying, and how much of that is here?",
    "After reading this, what would you still be wondering about [product name]?",
    "What, if anything, would give you pause about this product based on this description?",
    "How would you describe the tone of this writing, and how does that sit with you for a [product category] product?",
    "If you were skimming this on your phone, what would you want more of, and what would you skip past?",
    "The last time a product description made you trust it enough to buy, what did it include? How does this one compare?",
    "If you saw this next to two other [product category] options, what — if anything — would pull you toward this one or push you to the others?",
    "Think back to the last time you added something like this to your cart — what was the final nudge, and is that nudge anywhere in this description?",
    "How would you describe this product in your own words to a friend who was thinking about buying it?"
  ]
)
```

### Step 3: Analyse Description Feedback

Review audience responses and extract specific insights about the existing description:

**What's Working:**
- What they liked or found compelling
- What stood out as good
- What's clear and effective

**What's Missing:**
- Information they need that isn't there
- Questions that aren't answered
- Details that would build confidence

**What's Unclear:**
- Confusing or ambiguous parts
- What they didn't understand
- Where clarity is needed

**Concerns & Objections:**
- Hesitations they have after reading it
- What makes them unsure about buying
- Red flags or turnoffs in the current description

**Tone & Style Feedback:**
- Whether current tone feels right or off
- If it's too long, too short, or about right
- If it feels too casual, formal, or salesy

**Differentiation:**
- Whether it stands out or blends in with competitors
- What would make it more distinctive

**Optimization Priorities:**
- What needs to change to make them more likely to buy
- What would make them click "Add to Cart"
- What would build trust and credibility

### Step 4: Optimize the Product Description

**IMPORTANT**: Using the insights from Step 3, now optimize the existing product description with light tweaks and additions. Keep what's working, add what's missing, adjust what needs improvement.

**Optimization approach - LIGHT TOUCH:**

**Keep what's working:**
- Preserve elements they liked or found compelling
- Maintain the original structure if it's effective
- Keep tone/style if it resonated

**Add what's missing:**
- Insert key information they said they need (1-2 sentences or bullets)
- Add answers to questions they have
- Include trust signals or details they want (materials, care instructions, sizing, etc.)

**Tweak what's off:**
- Adjust language that felt too technical, salesy, or unclear
- Rephrase confusing parts for clarity
- Soften or strengthen tone if needed

**Address concerns:**
- Add 1-2 lines that proactively handle objections or hesitations
- Include reassurances about quality, durability, or fit if those were concerns

**Optimization guidelines:**
- Make minimal changes to preserve what's working
- Add missing information naturally (don't force it)
- Keep the original format unless feedback suggests changing it
- Focus on filling gaps, not complete rewrites
- Maintain the brand voice while improving clarity

**Output the optimized description now** - with changes clearly integrated into the original structure.

### Step 5: Deliver Optimization Package

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers.

**PRIMARY DELIVERABLE - Optimized Product Description:**

Output the complete, optimized product description ready to use, including:
- Product title (if optimization improves it)
- Full optimized description
- Formatted and ready to publish

**THEN provide supporting materials:**

**1. Optimization Summary:**
- **What changed**: Key changes made and why
- **Key insights applied**: Top 3-5 audience insights that shaped the optimization
- **Expected impact**: What should improve (clarity, trust, conversion, etc.)

**2. Before vs. After Comparison:**
- **Original**: [excerpt of key changes from original]
- **Optimized**: [how it was improved]
- **Why it's better**: Based on audience insight

**3. Description Feedback Summary:**
- **What worked**: What they liked about the original
- **What was missing**: Information gaps they identified
- **What was unclear**: Confusing or ambiguous parts
- **Concerns raised**: Hesitations or objections from the description
- **Tone feedback**: Whether style felt right or needed adjustment
- **Purchase drivers**: What would make them click "Add to Cart"

**4. Additional Recommendations:**
- Other elements to optimize (product images, reviews display, guarantees, FAQs)
- A/B test suggestions (if applicable)
- Related products or upsell opportunities based on what matters to audience

## Product Description Optimization Framework

| Element | Audience Insight Source | Optimization Approach |
|---------|------------------------|----------------------|
| **Opening hook** | What drives purchase decision | Lead with most compelling benefit/value |
| **Information hierarchy** | What they need to see to buy | Prioritize essential info first |
| **Tone** | Casual vs. professional preference | Match their preferred style |
| **Features vs. benefits** | What they care about most | Balance based on preference |
| **Trust signals** | What builds credibility | Include certifications, materials, guarantees they value |
| **Length** | Short/bullets vs. storytelling | Match preferred format |
| **Language** | Words that attract or repel | Use resonant language, avoid turnoffs |
| **Objection handling** | Concerns and hesitations | Address red flags proactively |
| **CTA** | What makes them click "Add to Cart" | Create urgency or clear next step |

## Tips for Best Results

- **Light touch wins**: Keep what's working, add what's missing - don't rewrite everything
- **Add, don't replace**: Insert missing information naturally into the existing structure
- **Preserve the voice**: Maintain the brand's tone while improving clarity and completeness
- **Fill specific gaps**: Add the exact information they said was missing (care instructions, sizing, durability)
- **Be specific**: "Machine washable" beats "easy care"; "True to size" beats "great fit"
- **Answer unasked questions**: Include info they said descriptions often miss
- **Tweak, don't transform**: Adjust unclear language rather than rewriting entire sections
- **Test minimal changes**: Sometimes adding one bullet point converts better than rewriting paragraphs

## Common Product Description Pitfalls

- **Feature dumping**: Listing features without explaining why they matter
- **Generic language**: "Premium quality", "best in class" without proof
- **Missing key information**: Not answering questions buyers have
- **Wrong tone**: Too casual for premium products, too formal for lifestyle brands
- **No trust signals**: Missing certifications, materials, guarantees buyers need
- **Ignoring objections**: Not addressing common concerns or hesitations
- **Too long or too short**: Not matching audience preference for description length
- **Overly salesy**: Hype and superlatives that reduce credibility

## Red Flags (Original Description Issues)

- Missing information buyers said they need
- Tone doesn't match audience preference
- Leads with features when audience cares about benefits (or vice versa)
- No trust signals included (materials, certifications, guarantees)
- Doesn't address common objections or concerns
- Uses language that turns off the target audience
- Too vague or generic

## Green Lights (Strong Optimization)

- Leads with what matters most to the audience
- Includes all information they need to buy confidently
- Tone and style match their preferences
- Addresses their objections and concerns proactively
- Includes trust signals they care about
- Answers questions descriptions typically miss
- Feels authentic and credible, not oversold
- Clear next step (CTA) aligned with purchase drivers

Roundtable Twin Interviews

Run an adaptive moderated interview with Digital Twins. One question is asked to a group of twins, all twins answer independently, the answers are analysed for themes, tensions, and surprises, then the next question is designed based on what emerged. This repeats for as many rounds as you want, each one sharper than the last. It is iterative research, not a single-pass survey — like a skilled moderator running a focus group, but faster, cheaper, and without groupthink.

Use when: Requests to run a focus group, conduct qualitative interviews, explore a topic in depth with real people, run a roundtable discussion, or do iterative audience research.

Download skill

View full skill

---
name: roundtable-twin-interviews
version: 1.1.0
description: "Use this skill to run an adaptive moderated interview with Digital Twins. One question is asked to a group of twins, their answers are analysed for themes and tensions, then the next question is designed based on what emerged — repeated across 8 rounds of questions, each one going deeper. Produces a comprehensive research report. Triggers include: requests to run a focus group, conduct qualitative interviews, explore a topic in depth with real people, run a roundtable discussion, or do iterative audience research. Uses OriginalVoices Digital Twins (ask_twins) across multiple rounds of questions."
---

# Roundtable Twin Interviews Skill

## Overview

Run an adaptive moderated interview with Digital Twins. One question is asked to a group of twins, all twins answer independently, the answers are analysed for themes, tensions, and surprises, then the next question is designed based on what emerged. This repeats for as many rounds as you want, each one sharper than the last. It is iterative research, not a single-pass survey — like a skilled moderator running a focus group, but faster, cheaper, and without groupthink.

## How It Works

1. One question is asked to a group of Digital Twins
2. All twins answer independently — their full responses are recorded
3. The answers are analysed for themes, tensions, and surprises
4. The next question is designed based on what emerged — probing deeper into the most interesting threads
5. Repeat for 8 rounds of questions, each one sharper than the last

## Workflow Steps

### Step 1: Gather Inputs from the User

Ask the user the following questions one at a time, waiting for each answer:

1. **"What's the topic for the roundtable?"** — The subject to explore (e.g. "How people feel about AI-generated content in news media")
2. **"Who is the audience?"** — (e.g. "UK adults 25-45 who read news daily")
3. **"How many rounds of questions?"** — Default: 8. Minimum: 3. Each round is one question asked to all twins, then analysed before the next. More rounds = deeper exploration.
4. **"How many twins per round?"** — Default: 20. Minimum: 15. More twins = broader perspective.

Once all four inputs are collected, confirm back with a one-line summary (e.g. *"Got it — 8 rounds of questions with 20 twins exploring how Gen Z feels about homeownership. Each round I'll ask one question, analyse the responses, then design the next question based on what emerged. Running now."*) and then execute the full session autonomously without stopping.

### Step 2: Run the Roundtable (The Loop)

Each round of questions follows the same cycle: **ask one question → collect all twin responses → analyse → design the next question**. Repeat for all rounds.

#### 2a. Design the Moderator Question

The question strategy adapts based on position in the session:

| Question | Strategy | Example approach |
|----------|----------|-----------------|
| Question 1 | **Open & exploratory** — get people talking, no leading | "When you think about [topic], what comes to mind first?" |
| Questions 2-7 | **Probe emerging themes** — pick a tension, contradiction, or strong theme from the previous round's answers and dig into it | "Several of you mentioned X but disagreed on Y — tell me more about where you stand on this." |
| Final question | **Synthesis** — force prioritisation and final takeaways | "If you could tell a brand/product team one thing about [topic] that they're probably getting wrong, what would it be?" |

**Critical:** Each question must be shaped by what twins actually said in the previous round. Do not use generic follow-ups. Read the responses, identify the most interesting thread, and pull on it.

**Question quality gate.** Because each round's question is written live from the previous round, every question you generate must pass two gates before you send it — **Range:** would 15 people answer it differently, or does it have an obvious answer? **Neutrality:** is it open-ended and non-leading (no yes/no, no baked-in sentiment)? When you probe a tension, keep it two-sided — don't adopt the sentiment of the loudest voices or ask "why is [X] so frustrating?"; ask "where do you each land on [X], and why?".

#### 2b. Query the Digital Twins

```
ask_twins(
  audience: "[audience description from Step 1]",
  questions: ["[moderator question for this round]"]
)
```

#### 2c. Record Responses

Store every twin's ID and full response. Never truncate or summarise individual responses — you'll need the full detail for the final report.

#### 2d. Between-Round Analysis

After each round of questions, output a brief analysis to the conversation:

**Question [n] Analysis:**
- **Dominant themes** — What most twins agreed on
- **Tensions** — Where twins disagreed or contradicted each other
- **Surprises** — Anything unexpected or counterintuitive
- **Next question direction** — What thread you're pulling on next and why

This keeps the user informed of progress and shows the adaptive logic at work. Do not stop or wait for input — proceed immediately to the next round.

#### 2e. Repeat

Move to the next round of questions. Continue until all rounds are complete. Do not pause between rounds or ask the user for direction — run the full session autonomously.

#### 2f. Handling Interruptions

If the session is interrupted (e.g. response length limits), output what you have so far and tell the user to say **"continue"** to resume from the next round. Track which round you're on so you can pick up where you left off.

### Step 3: Deliver the Research Report

Once all rounds are complete, output a comprehensive research report directly in the conversation.

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for subsections, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers. Where findings have clear quantitative patterns (e.g. "14 out of 20 twins expressed concern about X"), include a chart to visualise the distribution.

### Report sections:

**1. Executive Summary**
Top-line findings in 2-3 sentences. What is the single most important thing this research revealed?

**2. Key Themes**
3-5 most significant patterns that emerged across the session. For each theme:
- What it is and why it matters
- How it evolved across questions (did it deepen, shift, or get complicated?)
- How prevalent it was (e.g. "mentioned by 15 out of 20 twins")
- Supporting quotes from Digital Twins
- Include a chart showing theme prevalence

**3. Emotional Landscape**
What emotions drive behaviour and opinion in this space? Map the dominant feelings that came through — frustration, anxiety, hope, excitement, indifference, guilt, pride. Include a chart of dominant emotions and supporting quotes that show the emotional texture behind the data.

**4. Tensions & Disagreements**
What divided the group. For each tension:
- The two (or more) sides
- Who held which view and why
- Why it matters strategically
- Whether the tension resolved, deepened, or remained unresolved across questions

**5. The Arc of the Conversation**
How the discussion evolved from Question 1 to the final question. What shifted? What deepened? What emerged late that wasn't visible early on? This section is unique to the Roundtable format — it captures insight that single-pass research misses.

**6. Standout Quotes**
8-12 of the most powerful, surprising, or representative quotes from across the session. Group by theme or tension where it helps tell the story.

> "[quote]"
> — Twin [ID], Question [n]

**7. Opportunities**
Actionable insights for product, marketing, or strategy. What gaps exist? What do people want that they're not getting? What would change their behaviour?

**8. Strategic Implications & Recommendations**
5-7 specific recommendations based on the research. Each should include:
- What to do
- Why the research supports it
- What risk or opportunity it addresses

**9. Methodology**
This research was conducted using OriginalVoices Digital Twins — AI representations of real people, trained on data owned by the individuals they represent. [n] twins from the target audience participated across [n] rounds of questions. Each question was designed in response to themes emerging from the previous round's answers, creating an adaptive research arc that mirrors a moderated focus group.

## Tips for Best Results

- **Default to 8 rounds of questions**: This gives enough depth for themes to emerge, evolve, and resolve. Fewer than 5 rounds rarely surfaces the tensions that make research valuable
- **20 twins is the sweet spot**: Enough for diversity of perspective without becoming noise. Go higher (25-30) for topics where you expect wide variation
- **Trust the arc**: The best insights often come in questions 5-7, once surface-level responses have been exhausted and twins start revealing deeper motivations
- **Look for what changed**: If twins say one thing in Question 2 but something different in Question 6, that shift is often the most valuable finding
- **Don't lead in Question 1**: The opening question should be genuinely open. If you anchor too early, you'll get confirmation, not discovery

## Common Pitfalls to Avoid

- **Generic middle-round questions**: "Tell me more about that" is not a good probe. Reference specific things twins said.
- **Ignoring minority views**: If 3 out of 20 twins said something different, that's often where the insight lives
- **Summarising instead of analysing**: The report should be analytical. "Some people liked X and some didn't" is not analysis. "The group split along generational lines, with under-30s seeing X as essential and over-40s seeing it as intrusive — suggesting a segmented approach" is.

## Related Skills

- **deep-customer-research**: For single-pass comprehensive research when you don't need iterative depth
- **icp-discovery**: If you need to find the right audience before running a Roundtable
- **feature-concept-testing**: If you want to validate a specific feature rather than explore a topic
- **creative-testing**: If you want to test specific creative assets rather than explore a theme

Video Script Generation

Generate video scripts for creators or AI, informed by real audience insight. This workflow researches your target audience to understand what hooks them in the first seconds, what tone and pacing they prefer, what keeps them watching, and what drives engagement — then uses those insights to write complete video scripts that resonate and perform.

Use when: Requests to write video scripts, create content for YouTube/TikTok/Instagram, generate video narration, write video copy, or create scripts for video ads.

Download skill

View full skill

---
name: video-script-generation
version: 1.1.0
description: "Use this skill to generate video scripts for creators or AI, informed by real audience insight. Triggers include: requests to write video scripts, create content for YouTube/TikTok/Instagram, generate video narration, write video copy, or create scripts for video ads. Uses OriginalVoices Digital Twins (ask_twins) to understand what hooks the audience, what tone resonates, what keeps them watching, and what drives engagement — then writes complete video scripts optimized for that audience."
---

# Video Script Generation Skill

## Overview

Generate video scripts for creators or AI, informed by real audience insight. This workflow researches your target audience to understand what hooks them in the first seconds, what tone and pacing they prefer, what keeps them watching, and what drives engagement — then uses those insights to write complete video scripts that resonate and perform.

## Workflow Steps

### Step 1: Define Video Context

Gather from the user:
- **Video topic/concept**: What is the video about? (e.g. "How to choose the right CRM", "Product demo for productivity app", "Brand story video")
- **Target audience**: Who is this video for? (e.g. "Small business owners", "Gen Z interested in personal finance", "Marketing managers")
- **Video goal**: What should it achieve? (Education, product awareness, entertainment, conversion, brand building)
- **Platform**: Where will this video be published? (YouTube, TikTok, Instagram Reels, LinkedIn, website, video ad)
- **Video length**: Target duration (e.g. "30 seconds", "60 seconds", "3-5 minutes", "10+ minutes")
- **Product/brand info** (if applicable): What needs to be mentioned or featured?
- **Tone preference** (optional): Any specific tone requirements? (Educational, entertaining, inspirational, etc.)

### Step 2: Research Audience Video Preferences

One comprehensive research call to understand what will make this video effective for this audience.

**Question quality — apply every rule to every question before sending.** The goal is the widest possible range of honest answers. A weak question is narrow, leading, or closed, so respondents converge on the same obvious reply and you learn nothing. Run each question through these rules:

1. **Anchor in a specific moment or behaviour, not an abstract definition.** Ask "Walk me through the last time…" or "Tell me about a time…", never "How do you define / think about [concept]?".
2. **Keep the valence open — never lead.** Don't presuppose the experience was good or bad. Banned phrasings: "what frustrates you most about…", "the time [X] annoyed you", "what do you love about this". Ask "how did it go?" or make it two-sided: "what worked well, and what didn't?".
3. **One idea per question.** Split anything double-barrelled into separate questions.
4. **Open-ended only — no yes/no, no multiple choice.** Every question invites How / What / Why / Walk me through / Tell me about. Never hand respondents a menu of options to pick from — let them supply their own.
5. **No socially-correct answers.** If almost everyone would answer the same way ("Is quality important?" → yes), reframe with a trade-off ("when is quality *not* worth it?").
6. **Ground intent in real behaviour.** Don't ask hypotheticals like "would you buy / click / switch?" — they inflate. Anchor in what they last actually did ("the last time you bought something like this, what tipped the decision?").
7. **Build in room to diverge.** Prefer a fork, trade-off, or spectrum ("what would make you — and what would hold you back?") over a flat single-answer prompt.

Before sending, check each question passes both gates — **Range:** could most people answer it the same way, or is there an obvious "right" answer? **Neutrality:** does it assume how they feel, or hand them the answer? If it fails either, rewrite it.

```
ask_twins(
  audience: "[detailed target audience]",
  questions: [
    "Tell me about the last video on [topic] that made you stop scrolling — what was happening in those first few seconds?",
    "Tell me about a video on [topic] you watched all the way through — what was the pace like, and did that keep you or start to lose you?",
    "Think of a [topic] video whose tone just clicked for you — how would you describe the way it talked to you? And one whose tone put you off?",
    "Tell me about the last [topic] video you watched to the end, and the last one you clicked away from early — how long was each, and what made the difference?",
    "Walk me through the last time you were watching a [topic] video and something made you keep going instead of clicking away — what held you there?",
    "Tell me about a [topic] video that actually taught you something useful — what did you walk away with?",
    "Think of a [topic] video that really worked for you — how was it put together, and why did that click?",
    "When you picture the [topic] videos you actually finish, what's usually on screen — a person, visuals, text? Tell me about one that stuck with you.",
    "Tell me about the last [topic] video you shared, saved, or commented on — what made you do it?",
    "Tell me about the last [topic] video you bailed on within seconds — what made you nope out?",
    "Think of a [topic] video that hooked you — did it get straight to the point or build up first, and how did that work for you?",
    "Tell me about a time a video asked you to do something at the end — subscribe, buy, visit — that felt fine, and a time it felt pushy. What was the difference?"
  ]
)
```

### Step 3: Analyse Video Content Insights

Review audience responses and extract:

**Hook Strategy:**
- What captures attention in the first 3-5 seconds
- What makes them stop scrolling
- Hook styles that work vs. don't work

**Tone & Style:**
- Preferred tone (casual, professional, entertaining)
- Pacing preference (fast vs. slow)
- Energy level

**Content Structure:**
- Preferred format (story, how-to, tips, explanation)
- Optimal length for this topic and audience
- Information depth they want

**Engagement Drivers:**
- What keeps them watching
- What makes them engage (comment, share, save)
- What creates value for them

**Turnoffs:**
- What makes them stop watching
- What feels pushy or off-putting
- What to avoid

**CTA Approach:**
- Whether CTAs feel natural or pushy
- How to integrate them effectively

### Step 4: Write the Complete Video Script

**IMPORTANT**: Using the insights from Step 3, now write the complete video script ready to record. Output the full script with hooks, narration, and timing notes.

**Script structure:**

**[0:00-0:05] HOOK (First 3-5 seconds)**
- Open with the hook that stops them scrolling (based on research)
- Address their problem, curiosity, or interest immediately
- Make it impossible to scroll past

**[0:05-0:XX] INTRO (If appropriate for length)**
- Quick context or credibility (if needed)
- Set expectations for what they'll learn/get
- Keep it brief - only if research showed they want build-up

**[0:XX-X:XX] MAIN CONTENT**
- Deliver on the hook's promise
- Use their preferred tone and pacing (from research)
- Structure based on format preference (story, how-to, tips, explanation)
- Include the information they said they expect
- Keep attention with variety, value, or entertainment
- Match the depth they want (quick tips vs. detailed explanation)

**[X:XX-X:XX] PAYOFF/CONCLUSION**
- Summarize key takeaway or deliver final value
- Reinforce the benefit they got from watching
- Create satisfaction or curiosity

**[X:XX-END] CALL-TO-ACTION (If appropriate)**
- Natural CTA based on video goal (subscribe, visit website, try product, etc.)
- Integrate smoothly based on research (don't be pushy if they said that's a turnoff)
- Make it feel like a natural next step

**Writing guidelines:**
- Write in spoken language (how people talk, not formal writing)
- Use their preferred tone (casual/professional/entertaining from research)
- Match pacing preference (short punchy sentences for fast-paced, longer for detailed)
- Include [timing markers] so creators know pacing
- Include [VISUAL NOTES] in brackets if helpful for context, but focus on the script
- Write exactly what should be said/shown on screen
- Make it ready to record or input into AI video tools

**Output the complete script now** - from hook through CTA, ready to use.

### Step 5: Deliver Video Script Package

Format the output using proper markdown headers: use `##` for the report title, `###` for major sections, `####` for individual scripts, `>` blockquotes for Digital Twin quotes, and `---` horizontal rules between major sections. Never use plain text or bold-only text as section headers — always use proper markdown headers.

**PRIMARY DELIVERABLE - Complete Video Script:**

Output the full, production-ready video script including:
- Timing markers [0:00-0:05]
- Hook
- Intro (if appropriate)
- Main content narration/dialogue
- Payoff/conclusion
- Call-to-action (if appropriate)
- [Visual notes] where helpful
- Ready to record or use in AI video generation tools

**THEN provide supporting materials:**

**1. Script Overview:**
- **Target length**: Estimated final video duration
- **Tone**: Casual/professional/entertaining based on research
- **Format**: Story/how-to/tips/explanation
- **Platform optimization**: Specific to YouTube/TikTok/Instagram/etc.

**2. Key Insights Used:**
- **Hook strategy**: What captures their attention (with research quote)
- **Tone & pacing**: What style resonates with them
- **Content structure**: What format they prefer
- **Engagement drivers**: What keeps them watching
- **CTA approach**: How to ask for action without being pushy

**3. Audience Video Profile:**
- What hooks them in the first seconds
- Preferred tone and pacing
- Optimal video length for this topic
- What keeps them watching vs. clicking away
- What drives engagement (shares, comments, saves)

**4. Production Notes:**
- Delivery tips based on tone preference
- Pacing guidance (fast vs. slow)
- Visual suggestions (if audience expressed preference)
- Platform-specific considerations

**5. Optimization Recommendations:**
- Alternative hooks to test
- Engagement tactics based on what drives shares/comments
- CTA variations to try
- A/B test suggestions

## Video Script Framework

| Script Element | Audience Insight Source | Optimization Approach |
|----------------|------------------------|----------------------|
| **Hook (0-5 sec)** | What stops them scrolling | Lead with their problem/curiosity/interest |
| **Intro** | Whether they want build-up | Skip if they want to "get to the point"; brief if they want context |
| **Tone** | Casual/professional/entertaining preference | Match their preferred style throughout |
| **Pacing** | Fast vs. slow preference | Short sentences for fast; longer, detailed for slow |
| **Content structure** | Story/how-to/tips/explanation | Organize based on format preference |
| **Depth** | How much information they want | Match detail level to their expectations |
| **Length** | Optimal duration for topic | Stay within their attention span sweet spot |
| **Engagement hooks** | What keeps them watching | Use variety, value, or entertainment tactics |
| **CTA** | Whether CTAs feel natural or pushy | Integrate smoothly or skip if it's a turnoff |

## Tips for Best Results

- **First 5 seconds are everything**: The hook must stop the scroll based on research
- **Write how people talk**: Spoken language, not written language
- **Match their energy**: Fast-paced and punchy, or calm and detailed
- **Deliver on the hook**: Whatever you promise in the first 5 seconds, deliver quickly
- **Cut ruthlessly**: If research says they want it short, every word must earn its place
- **Use pattern interrupts**: Variety keeps attention (questions, facts, stories, etc.)
- **Natural CTAs**: If they said pushy feels bad, make it a soft suggestion
- **Platform matters**: TikTok scripts ≠ YouTube scripts ≠ LinkedIn scripts

## Common Video Script Pitfalls

- **Slow hook**: Taking too long to get to the point when they want it fast
- **Wrong tone**: Being too casual for professional topics or too formal for entertainment
- **Too long**: Ignoring research that says they prefer 60 seconds when you write 5 minutes
- **Buried value**: Saving the best for last when they'll click away before they get there
- **Generic opening**: "Hey guys, today we're talking about..." when they need a strong hook
- **Pushy CTA**: Hard-selling when research says they hate that
- **Written language**: Sounding like an essay instead of natural speech
- **No payoff**: Building up without delivering satisfaction at the end

## Red Flags (Revise Script)

- Hook doesn't address what they said stops them scrolling
- Tone doesn't match their preference (too casual, formal, or salesy)
- Length exceeds their attention span for this topic
- Pacing is off (too fast when they want detail, too slow when they want quick tips)
- CTA feels forced when they said pushy is a turnoff
- Content doesn't deliver what they said they expect from videos on this topic

## Green Lights (Strong Script)

- Hook immediately addresses what captures their attention
- Tone and pacing match their preferences exactly
- Length fits their attention span for this topic
- Delivers the information or value they expect
- Structure matches format preference (story, how-to, tips)
- Keeps momentum with variety and engagement tactics
- CTA feels natural and appropriate for their expectations
- Written in spoken, natural language
- Platform-optimized for where it'll be published