Definition
Google's help centre describes a responsive search ad as an ad built from multiple headlines and descriptions, which Google Ads tests in different combinations to learn which perform best. When the ad shows, a subset of the headlines and descriptions appears, in an order the system chooses, with how many appear depending on the device and the space. In search campaigns it is the standard text ad type.
Three properties follow from that design. Each headline must stand on its own, since its neighbours are not known in advance. The ad's relevance to a search is a property of the whole pool: the more lines that reflect the ad group's theme, the more relevant combinations are available. And the ad changes over its life, because the combinations it favours narrow as results accumulate. The responsive search ads guide covers how many lines to write and what roles they should play.
In practice
Pinning. A headline or description can be pinned to a position so that it always appears there. Pinning is justified for mandatory wording, such as a regulated disclaimer, or for a brand name that must lead. Every pin removes combinations the system could have tried, so pinning several alternative lines to the same position is the lighter way to guarantee a message while keeping some variety.
Ad strength. As the ad is built, Google rates its assets for relevance, quality and diversity, with suggestions such as adding headlines or making them more distinct. It is a completeness check on the pool, and Google states that it does not enter the auction. A pool rated well can still carry a weak message.
Writing distinct lines. Headlines that say the same thing in slightly different words waste slots. A pool that covers different questions a searcher might have, such as what, why this provider, what it costs and what to do next, gives the system real alternatives. The creative production playbook treats the pool as a brief to be filled, and the naming conventions guide shows how to name ad groups and assets so that a change of angle can be traced in reports.
Reading asset and combination reports
Because there is no single ad, reporting happens at two levels. The asset report rates each headline and description once it has enough data, and the combinations report shows which groupings served most often. Both are useful for spotting a weak line to replace. Neither is a controlled test: lines that serve more were chosen more because the system predicted they would do well, so their exposure and their results are entangled. A line with few impressions has not lost a fair contest; it was rarely picked.
Where a controlled comparison matters, Google's ad variations and experiments split traffic deliberately. The search ad asset test lays out such a design, with a hypothesis, a metric and rules for reading the result, and judges it on conversions rather than on CTRShare of impressions that result in a click. alone.
Responsive search ads and Ad Rank
The pool feeds the auction. Ad relevance and expected click-through, two of the components behind Quality ScoreGoogle's 1–10 keyword diagnostic., depend on how well the available lines match the search, and the auction-time version of those signals feeds Ad RankThe value Google Ads calculates at each auction to decide if and where an ad shows.. The details around the ad, such as sitelinks and callouts, are a separate system: Ad AssetsExtra information such as sitelinks, callouts and phone numbers that Google Ads can add to an ad. extend the ad but are not part of its headline pool, as the ad assets guide explains.
A bike hire shop unpins its brand
Sallowspoke Bike Hire runs one responsive search ad with twelve headlines, pinning "Sallowspoke Bike Hire" to position one and "Book Online Today" to position two. Ad strength reads Poor and impressions fall 18% over two months. The shop unpins the call to action, pins three brand variations to position one instead of one, and replaces four near-identical headlines with lines answering different questions: e-bike availability, family bikes, trail maps included and free helmet hire. Impressions recover to above their earlier level within four weeks. The combinations report shows the brand line most often paired with "E-Bikes Available Daily", and bookings from the ad rise from 46 to 58 a month.
Related terms
Ad AssetsExtra information such as sitelinks, callouts and phone numbers that Google Ads can add to an ad. sit around the ad and are managed separately. Ad RankThe value Google Ads calculates at each auction to decide if and where an ad shows. is the auction value the ad's quality feeds into, and Quality ScoreGoogle's 1–10 keyword diagnostic. is the keyword-level diagnostic that reflects it. CTRShare of impressions that result in a click. is the first result the ad is judged on, and the CTR calculator compares ads when only raw clicks and impressions are to hand. The Google Ads search hub shows where ads sit among keywords, bidding and assets.
Common confusions
- The preview is the ad. It is one combination of many, and no single person sees them all.
- Excellent ad strength means a good ad. It means a complete, varied pool; persuasion is judged by results.
- The asset report is an A/B test. Exposure is decided by the system, so lines are not compared fairly.
- Pinning makes the ad perform better. It guarantees a message and reduces the combinations tested.
- Sitelinks are part of the ad. They are separate assets with their own rules for showing.
