Sign in Start free
ADS & LANDING PAGES

Full RSA asset set with pinning plan

Use when building a complete responsive search ad and you need a pinning strategy, not just lines.

responsive-search-ad-asset-set.md
Download .md 
Act as a Google Ads specialist building one responsive search ad.

Ad group: {{AD_GROUP_NAME}}
Top 5 keywords in the ad group: {{KEYWORDS}}
Offer and price or starting price: {{OFFER}}
Differentiators: {{DIFFERENTIATORS}}
Compliance limits (things I am not allowed to claim): {{COMPLIANCE_LIMITS}}

Produce:
1. 15 headlines (max 30 chars) grouped into four buckets: keyword match, offer, differentiator, CTA.
2. 4 descriptions (max 90 chars).
3. A pinning plan: which headlines to pin to position 1, which to position 2, and which to leave unpinned, with one sentence of reasoning per pin.
4. A duplication check: flag any two assets that carry the same idea, since duplicate meaning weakens the asset mix.

Constraints:
- Show the character count after every asset.
- Never breach the compliance limits listed above.
- Do not fabricate figures. Where a number would help but I have not supplied one, write {{NUMBER_NEEDED}} in the line.
- Avoid em dashes, "in today's fast-paced world" style openers, and "it is not just X, it is Y" phrasing.

Finish with three assets you would swap in first if the ad underperforms, and what each one is testing.

Fill in before running

Replace each placeholder with your own detail. The more specific you are, the less the model invents.

  • {{AD_GROUP_NAME}}
  • {{KEYWORDS}}
  • {{OFFER}}
  • {{DIFFERENTIATORS}}
  • {{COMPLIANCE_LIMITS}}
  • {{NUMBER_NEEDED}}

Getting a better result

  1. Give the real keyword list from the search terms report rather than what you think you bid on.
  2. Over-pinning kills asset variety - accept the plan only if at least eight assets stay unpinned.
  3. Ask it to redo the buckets if one bucket has fewer than three headlines.

Questions about this prompt

When do I want this rather than the character limits prompt?

When you are building the whole ad rather than sourcing lines. The character limits prompt gives you copy; this adds the four buckets, the pinning plan and a duplication check, which together decide how Google is allowed to combine your assets. If you only need a few more headlines for an ad that already runs, the other prompt is faster.

What do I need to hand before running it?

The real top five keywords from your search terms report rather than the ones you believe you bid on, the offer with its price, your differentiators, and {{COMPLIANCE_LIMITS}}. That last field is the one people leave blank. Empty, the model has nothing to check itself against, and you get assets a regulated account cannot run.

What does the output contain and what is the useful bit?

Fifteen bucketed headlines, four descriptions, a pinning plan with a sentence of reasoning per pin, a duplication flag, and three swap-in assets with what each tests. The duplication check earns its place: two assets carrying the same idea look like variety in the interface and are not, so the mix is thinner than the asset count suggests.

Where does this go wrong?

Accepting a plan that pins too much. Every pin removes combinations, and pinning ten assets turns a responsive ad into an expanded text ad with extra admin. Count the unpinned assets and send back anything leaving fewer than eight free. How aggressively to pin is argued both ways among practitioners, so treat the plan as a position, not a rule.