A Responsive Search Ad has 19 slots. Most law firm agencies pin all of them or none of them. Both are wrong. Here is the exact map of what to pin, what to test, and what to release to Google.
Pin 2 to 3 variants at H1 and H2. Release H3 through H15 entirely. Pin 2 to 3 at D1. Put D2 at the campaign level. Write everything in sentence case. A 20,000-account Optmyzr study found this structure delivers a $14 CPA versus $61 for fully pinned accounts. In Florida, the same structure is also the cleanest path through Bar Rule 4-7.19.
Why RSA pinning for lawyers is nothing like ecommerce
Every RSA guide on the internet treats pinning like an ecommerce problem. It isn’t. In ecommerce, a bad headline costs you a $40 sale. In personal injury, a bad headline in Lafayette, Louisiana costs you $575 per click on traffic that was never going to convert. In Florida, a bad permutation of headlines technically constitutes an unfiled advertisement and puts the attorney in violation of Bar Rule 4-7.19.
“Legal is the only major vertical where both performance optimization and regulatory compliance require the same pinning strategy. That’s not a coincidence. Google built RSAs for retailers. Lawyers have to retrofit.”
The firms winning Google Ads in 2026 are not spending more. They figured out that RSA pinning in a regulated vertical is a three-layer problem: a compliance layer you must control, a performance layer you must test, and an algorithmic layer you must release. Fight Google’s machine learning and you lose. Give it zero guardrails and you violate the Bar.
What the 20,000-account study actually found
Optmyzr ran the largest public study on RSA pinning across 20,000 Google Ads accounts. The question was simple: does pinning help, hurt, or make no difference? The answer turned out to be all three, depending on how you pin.
Google now serves Responsive Search Ads as a single headline in some auctions. That means H1 is the only asset guaranteed to appear in every impression. H2, H3, and beyond can be withheld entirely based on device, intent signals, or auction dynamics.
This is why partial pinning at H1 is no longer optional for law firms. If you leave H1 unpinned, you’re letting Google decide whether a random rotating variant represents your firm in 100% of impressions. Pin 2 to 3 pre-approved H1 variants and every possible single-headline serve is compliant and on-message.
Two other findings from the same dataset that almost no law firm agency has internalized. Sentence case headlines deliver a $7.46 CPA versus $27.47 for title case, a 3.7x gap. And “Average” Ad Strength outperforms “Excellent” Ad Strength on both CPA and conversion rate. The scoring system agencies are chasing is not the scoring system that makes you money.
The decision matrix (the whole strategy on one page)
This is the table to screenshot. Every RSA slot, what you put in it, and what to avoid. Apply this to every new ad group.
Bad RSA vs good RSA
Two Responsive Search Ads for the same Tampa personal injury firm, same budget, same landing page. The difference is entirely in how the assets are structured.
Headline formulas by practice area
Use these as your H1 and H2 pin pool. Load 2 to 3 variants per position. Sentence case. Keep H1 under 25 characters where possible.
Pinning strategy by market tier
CPCs for personal injury keywords range from $45 in secondary markets to $575 in Lafayette, Louisiana. Your pinning strategy should flex with the market.
Florida Bar Rule 4-7.19 and the RSA permutation problem
The Florida Bar requires attorney ads to be submitted for review at least 20 days before first use, with a $150 filing fee per ad. A Responsive Search Ad with 15 headlines and 4 descriptions can technically produce tens of thousands of unique combinations. Each is arguably a separate advertisement.
This is not theoretical. It’s why most Florida law firms either under-use Google Ads or accept compliance risk they do not understand. The same pinning structure that wins on performance also collapses this risk down to a manageable audit path.
A pending amendment would exempt sponsored search results without images, audio, or video from the filing requirement. If it passes, the compliance burden on RSA campaigns drops materially. Until then, the protocol above is the safe path.
Compliance notes for Texas, New York, California
Florida has the strictest RSA compliance regime in the country because of Rule 4-7.19 filing. Other major legal markets don’t require filing, but each has its own traps.
Frequently asked questions
Should I pin all my headlines to stay in control? +
No. The Optmyzr study of 20,000 accounts shows fully pinned ads deliver a $61 CPA versus $14 for partially pinned. You get control at the cost of a 4x increase in your cost per case. The correct approach is partial pinning, 2 to 3 variants at H1, 2 to 3 at H2, unpinned at H3 through H15.
What is partial pinning and why does it beat full pinning? +
Partial pinning means you pin 2 to 3 interchangeable variants to a specific position rather than pinning a single asset or pinning nothing. Google still tests combinations, but only among the variants you approved. You keep message control. Google keeps optimization room. Both win.
Does Ad Strength “Excellent” matter? +
Not the way agencies tell you. “Average” Ad Strength delivers the best CPA at $12.43 and the best conversion rate at 12.65%. “Poor” Ad Strength delivers the best ROAS at 327%. “Excellent” often correlates with over-loaded, diluted ads. Optimize for cost per case, not for the badge.
Is my Florida firm violating Bar rules by running RSAs? +
Possibly, depending on how your ads are structured. Florida Bar Rule 4-7.19 requires attorney ads to be submitted for review with a $150 filing fee per ad. If your RSA generates thousands of unfiled permutations, you have exposure. The fix is the same protocol that improves performance: limit variants per position, use presumptively valid content, document active combinations.
How many RSAs should I run per ad group? +
Two to three. Adding a second RSA to an ad group increases conversions by roughly 6.6%. Adding a third adds another 3.7%. Past three RSAs per ad group, the gains flatten and you start cannibalizing your own data.
Does sentence case really beat title case by that much? +
Yes. The data shows a 3.7x CPA gap, $7.46 for sentence case versus $27.47 for title case. The theory is that sentence case reads more like organic results and less like advertising, which lifts CTR. Easiest change in your account today. Takes 10 minutes.
Want me to look at how your firm’s RSAs are pinned?
Send me read-only access to your Google Ads account. I’ll audit your pinning structure, tell you which positions are costing you cases, and for Florida firms I’ll flag language that creates Bar filing exposure. If your account is already structured well I’ll tell you that too and you can move on to something else.
