Most people writing a responsive search ad are still doing the old job. They compose one line they are proud of, then produce fourteen near-copies of it so the interface stops complaining. Ad Strength goes green. Nothing improves.
That happens because responsive search ads changed what a copywriter is for. You are no longer choosing the words a searcher sees, Google does that, per auction, from the parts you supplied. What you still control is the range of arguments in the cupboard. If all fifteen headlines make the same argument in fifteen registers, every combination Google can build says the same thing.
What follows is the decision layer, not a swipe file: how to work out what your fifteen slots should cover, when pinning is worth the cost it carries, and how to read the asset report without gutting a test halfway through. The video below walks through the mechanics in the interface.
Responsive Search Ads Explained, Best Practices and Common Mistakes
Source video: Google Ads Responsive Search Ads Explained – Best Practices, Tips & Do’s and Don’ts
PART 1 · DIAGNOSE
You Are Stocking a Cupboard, Not Writing a Sentence
IN BRIEFA responsive search ad takes up to 15 headlines and 4 descriptions and builds a different ad for each search. Your unit of work is therefore the set, not the line. Judge the set by how many distinct arguments it holds, the same portfolio thinking that shapes a properly managed Google Ads account.
Google’s own guidance is blunt about what it wants: as many unique assets as you can supply, up to 15 headlines and 4 descriptions, each highlighting different selling points and calls to action, with repetition of the same content across assets explicitly discouraged.
Read that as a brief rather than a rule and the work changes shape. Before writing anything, list the reasons a Malaysian buyer would choose you, price, speed, proximity, guarantee, credentials, range, risk removal. Those reasons are your slots. Phrasing comes after.
- An argument is not a synonym. “Fast delivery” and “quick shipping” are one slot, not two.
- Objections count as arguments. “No deposit required” answers the doubt that stops the click.
- Some slots exist to be lost. You will not know which claim carries the account until it competes for a month.
Not sure which of your selling points deserve a headline slot?
Most accounts discover they have three real arguments and twelve rewordings of one. See how a managed account decides what the ads should claim
BENCHMARK BRIEFING 1 OF 4
What Google Actually Grades in an RSA
IN BRIEFAd Strength is not a quality opinion and does not decide whether your ad can serve. It grades four mechanical things, keyword relevance, asset quantity, asset diversity, and sitelink supply. Knowing which lever a low rating points at saves the guesswork, much as reading Quality Score by component does.
| Factor | What Google says it measures | Published threshold | What a weak score points at |
|---|---|---|---|
| Asset-keyword relevance | Whether headlines carry your relevant keywords | Keyword must sit whole in one headline | Ad group is too broad, or copy is too clever |
| Asset quantity | How many combinations you allow Google to build | Up to 15 headlines, 4 descriptions | Empty slots, or over-pinning |
| Asset diversity | Whether assets say genuinely different things | No published number, repetition penalised | One argument wearing fifteen outfits |
| Sitelink supply | Sitelinks available at any account level | Six or more, uploaded or dynamic | An account setting, not a copy problem |
Aggregated by IZI Digital Marketing from Google’s documentation on Ad Strength for responsive search ads, its reference page on responsive search ads, and its best practices for creating effective responsive search ads, 2026. The final column is editorial interpretation, not a Google metric.
The last row is the one that catches people out. A perfectly written ad can sit below “Good” because the account has fewer than six sitelinks, nothing to do with the copy. Check that before rewriting a single headline.
PART 2 · DIAGNOSE
Should You Pin Anything at All?
IN BRIEFPinning locks a headline to one position and blocks every other headline from it. Google does not recommend it for most advertisers because it shrinks the combination pool. Pin only when something must appear, and pin several alternatives to the same slot, a control decision that pairs with choosing keyword match types deliberately.
Pinning is the one place where responsive search ads still let you overrule the machine, which is exactly why it gets overused. The cost is real: fewer combinations, fewer queries matched, and a lower Ad Strength rating as a side effect. The benefit is also real when a claim is legally or commercially compulsory.
DECISION BOX · PIN, OR LEAVE IT TO GOOGLE
| Your situation | Decision | Cost you accept |
|---|---|---|
| A regulator or franchise requires exact wording | Pin it, position 1 | Fewer combinations, lower rating |
| Brand name must never be dropped | Pin two or three variants to one slot | Modest, variety survives |
| You simply prefer your own favourite line | Do not pin | You lose an opinion, not performance |
| Ads keep showing an offer that has expired | Delete the asset, do not pin around it | None, pinning is the wrong tool |
Verdict: Pin for obligations, never for preferences. When you must pin, pin a small group of alternatives to the same position so the system still has something to choose between.
BENCHMARK BRIEFING 2 OF 4
What Google’s Own Data Says Each Move Is Worth
IN BRIEFGoogle publishes uplift figures for three specific responsive search ad moves, and they are not equal. Lifting a poor ad to excellent is worth far more than adding a third ad to a group that already has two, which tells you what to do first when time is short, the same sequencing logic behind a first campaign built in the right order.
| Move | Reported average gain | Effort required |
|---|---|---|
| Ad Strength from poor to excellent |
15% more conversions |
A working session on assets |
| First ad group RSA becomes two |
6.6% more conversions |
One additional ad build |
| Second RSA becomes a third |
3.7% more conversions |
Another full ad build |
Figures reported by Google in its Ad Strength documentation for responsive search ads and its guide to responsive search ads, attributed to Google internal data for August 2025 and stated as averages at similar cost per conversion. Compiled and ranked by IZI Digital Marketing; the effort column is an editorial estimate.
Note what the shape of that chart implies. Returns fall off quickly once a second ad exists, so an afternoon spent fixing the assets in your existing ad usually beats a week spent building more ads around them.
PART 3 · DESIGN
Fifteen Headlines, Not One Headline Fifteen Times
IN BRIEFFill the slots from a list of argument types rather than from inspiration. Keyword-match, offer, proof, objection, urgency and action each earn separate headlines, and a keyword only helps Ad Strength when it sits whole inside one, a discipline as unglamorous and as decisive as maintaining a negative keyword list.
Work through six families and stop when each has coverage. Filling responsive search ads this way is deliberately mechanical, because inspiration produces clusters and clusters produce identical combinations.
- Keyword-match headlines. Two or three carrying the ad group’s main search phrase intact, since a keyword split across two headlines does not count.
- Offer headlines. What the buyer gets, free quote, same-day site visit, fixed monthly fee.
- Proof headlines. Years operating, certifications held, service coverage. Only what you can defend.
- Objection headlines. The doubt that stops the click: no lock-in, no deposit, no obligation.
- Urgency or timing headlines. Response times, availability, seasonal windows.
- Action headlines. The plain instruction, call, WhatsApp, book, request.
One caution on tone. Headlines are capped at 30 characters, which tempts writers into telegraphic fragments that read like a classified advert. If a line would embarrass you on a shopfront, it will not do better in an auction.
BENCHMARK BRIEFING 3 OF 4
A Coverage Map for the Fifteen Slots
IN BRIEFA workable default spreads fifteen headlines and four descriptions across the six families rather than distributing them evenly. Keyword and offer lines take the largest share because they do the matching work, an allocation worth auditing the way a structured SEO audit checks coverage before tactics.
| Family | Headlines | Descriptions | Job it does |
|---|---|---|---|
| Keyword match | 4 | 1 | Signals relevance to the query |
| Offer | 3 | 1 | Gives a reason to click now |
| Proof | 3 | 1 | Makes the claim believable |
| Objection | 2 | 1 | Removes the reason to hesitate |
| Timing | 2 | 0 | Sets an expectation of speed |
| Action | 1 | 0 | Tells the reader what to do |
Illustrative planning model by IZI Digital Marketing, built on Google’s best practices for creating effective responsive search ads and its documentation on asset diversity and keyword relevance in Ad Strength. Allocations are a starting point for planning, not measured client results.
Treat the numbers as a starting split, not a rule. A service with one dominant search phrase needs fewer keyword-match lines; a category with many phrases needs more, and Google says as much when it advises maximising headline count for keyword-heavy ad groups.
Want the coverage map applied to your own ad groups?
The gaps usually appear in the first hour and cost nothing to close. Compare what a managed Google Ads scope covers each month
PART 4 · DEPLOY
Building the Ad Without Undoing Your Own Work
IN BRIEFBuild in a fixed order, keywords, then assets, then sitelinks, then a second ad, and change one thing at a time afterwards. The sequence matters because Ad Strength recalculates whenever keywords change, which is the diagnose-before-deploy habit applied to ad copy.
Build every responsive search ad in these five steps. Each one either supplies an input Ad Strength grades or removes a suspect from the next review.
- Settle the ad group first. Ad Strength is scored against the keywords in the group, so a group still being reshaped cannot produce a stable rating.
- Write to the six families. Fill fifteen headlines and four descriptions, keeping each main keyword whole inside at least one headline.
- Check sitelinks before blaming copy. Six or more across account, campaign or ad group level is the published threshold.
- Pin only what is compulsory. If you must pin, pin two or three alternatives to the same position.
- Add a second ad, then stop. Two ads at good or excellent strength per group is Google’s stated recommendation; a third adds noticeably less.
Step one is where most rework comes from. Reshuffling keywords after the copy is written resets the score and makes the previous fortnight’s reading meaningless, the same reason conversion tracking gets verified before anything is judged.
BENCHMARK BRIEFING 4 OF 4
A Six-Week Rhythm for Testing Responsive Search Ad Copy
IN BRIEFAsset performance labels need weeks of impressions before they mean anything, and swapping copy weekly guarantees you never learn which line worked. Plan the reading dates in advance, as you would when diagnosing an account that spends without converting.
| Period | Work done | Leave alone | Conclusion available |
|---|---|---|---|
| Week 1 | Build to the coverage map, add sitelinks | Keywords, bids | Whether inputs are complete |
| Weeks 2–3 | Nothing, collecting impressions | Everything | None yet, and that is fine |
| Week 4 | First read of the asset report | Assets rated low but unproven | Which families are being shown |
| Week 5 | Replace only the clearly weak assets | Pins, keywords, the second ad | Which arguments underperform |
| Week 6 | Cost per enquiry compared to week 1 | Nothing further | Whether the copy work paid |
Illustrative model by IZI Digital Marketing, built on Google’s documentation on the ad-level asset report for responsive search ads and its statement that Ad Strength recalculates whenever assets or keywords change. A planning rhythm, not measured client results.
Weeks two and three are the ones clients want to skip, and skipping them is how accounts accumulate copy nobody can defend. Impressions have to accumulate before any label describes reality.
PART 5 · DRIVE
Reading the Asset Report Without Overreacting
IN BRIEFAsset labels tell you what was shown and how it performed relative to your other assets, not whether the argument is wrong. Replace the weak line with a different argument, never with a reworded version of itself, and judge the result against enquiries, which is the standard IZI applies before recommending any channel.
Three habits separate accounts whose responsive search ads improve from accounts that simply churn copy.
- Replace arguments, not words. A low-rated objection headline usually means the wrong objection, not the wrong phrasing.
- Ignore assets with almost no impressions. They have not lost a test; they have not entered one.
- Judge on enquiries, not on labels. The report ranks assets against each other, and an ad group can be full of relatively strong assets while the offer behind them still fails.
When the copy is genuinely good and enquiries still stall, the constraint has usually moved downstream. That is a conversation about the destination, a website built to convert rather than to describe, not another round of headline writing.
COMMON QUESTIONS
Responsive Search Ads: Common Questions
1. How many headlines should a responsive search ad have?
Supply as many distinct ones as you honestly can, up to the maximum of fifteen. It depends on your ad group, Google recommends maximising headline count when a group carries many keywords, and fewer are needed when it carries one dominant phrase. The limit that matters is how many genuinely different arguments you have, not the slot count.
2. Does pinning hurt performance in responsive search ads?
It usually costs you reach, which is why Google does not recommend it for most advertisers. It depends on why you are pinning, a compulsory legal or brand line justifies it, a personal preference does not. If you must pin, pin two or three alternatives to the same position so the system still has something to test.
3. Does Ad Strength affect whether my ad can show?
No. Google states plainly that the Ad Strength rating does not directly influence serving eligibility. It depends on how you use the rating, treat it as a checklist of missing inputs during creation and editing, not as a score that gates your ad or measures your landing page.
4. How many responsive search ads should each ad group have?
Two, at good or excellent strength, is Google’s stated recommendation. It depends on your capacity to maintain them, Google reports a 6.6% average conversion increase when a group goes from one ad to two, and a smaller 3.7% average increase going from two to three. Two well-built ads beat three neglected ones.
5. Why is my Ad Strength stuck below “Good” when the copy looks fine?
Most often because of an input outside the ad itself. It depends which factor is short, sitelinks are the usual culprit, since Google’s threshold is six or more across account, campaign or ad group level. Check that, and check whether a keyword sits whole inside at least one headline, before rewriting anything.
Fifteen headlines written, and still not sure what your ads are arguing?
A Blueprint session maps your ad groups against the six argument families, shows you which slots are duplicates, and tells you plainly which claim to test next. You leave with the decision made, not another list of ideas.