Google Ad Copy

---
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