Free PPC

Toolkit

What this framework solves

Most ad accounts ship 15 headlines that say the same thing in different words, then wonder why the algorithm cannot find a winning combination. The gap between a headline that earns the click and one that gets skipped is not creativity. It is specificity, alignment, and proof. This framework replaces “variation” with structured angle distribution and replaces clever copy with claims that trace back to the landing page.

The core idea

Every headline has one of six jobs: confirm relevance (keyword), communicate a benefit (outcome), establish credibility (proof), drive action (CTA), differentiate (why you), or create urgency (time pressure). If a headline cannot be assigned to one of these jobs, it is filler. If two headlines do the same job with different words, one is wasted. The specificity test is ruthless: could a competitor use this exact headline unchanged? If yes, it contains zero information. “4.9 Stars, 2,100 Reviews” passes. “Highly Rated” fails. Every claim must trace back to the landing page, because message match is the strongest predictor of both Quality Score and conversion rate.

How to apply it

1. Audit angle distribution across 15 headlines. Standard split: 2-3 keyword, 3-4 benefit, 2-3 proof, 2-3 CTA, 2-3 differentiator, 0-2 urgency (only with a real deadline). RSAs with 5+ distinct angles outperform single-angle RSAs by 15 to 25 percent on CTR.
2. Run the specificity test on every headline. Replace every generic (“Quality Service,” “Best in Class,” “Professional Team”) with a specific data point from the landing page: a review count, a year founded, a response time, a price point.
3. Front-load descriptions before the 45-character mobile cutoff. The CTA, the offer, and the proof point must appear before character 45. A description whose payoff is at character 80 is invisible on mobile, which is most of search.
4. Build copy from the landing page, not from imagination. Scrape or screenshot the page first. Build a validated claims list. Promising “Free Trial” when the page offers a demo is the most expensive copy mistake because it breaks the alignment chain on every click.
5. Default to no pinning. Unpinned RSAs give Google 43,680 combinations. Pinning collapses that to hundreds. Pin only with compliance justification, demonstrated keyword-presence quality issues, or 1,000+ impressions of clear performance data, and always pin 2-3 variants to the same slot.
6. One RSA per ad group is usually correct. Three is Google’s max, and only justified when each targets a clearly different angle strategy and the ad group has 1,000+ monthly impressions to feed each one.
7. Diagnose CTR vs. CVR problems separately. Low CTR on transactional keywords is a copy specificity gap or a differentiation gap. Good CTR with low CVR is a downstream alignment break: the ad earned the click by promising what the page does not deliver. Fix the page before rewriting the ad.
8. Mine asset performance labels after 1,000+ impressions. Build on what earns “Best” labels, retire what stays at “Low” past 30 days. The labels are the feedback loop that prevents the RSA from going stale.

When to use it

Use this when CTR is below industry floor (3-6 percent on non-brand search depending on vertical), when Ad Strength stays at “Poor” or “Average,” when an account is rolling out new RSAs, or as a hygiene pass on any account that has not refreshed copy in 6+ months. It calibrates by vertical: ecommerce leads with price, shipping, reviews; lead gen pre-qualifies in the headline (“Enterprise Teams of 50+”) to filter unqualified clicks; B2B SaaS targets the user and the approver simultaneously; local services compresses availability + credibility + proximity into the headline; regulated verticals trade aggressive claims for proof-by-credential. It does not apply when the underlying landing page lacks the raw material to fuel proof or urgency angles. In that case, fix the page first; no amount of headline rotation manufactures evidence the page does not contain.