Assets, not ads
Google assembles each impression from your assets, showing up to three headlines and two descriptions chosen for the query, so every headline must make sense alone and beside any other. Give each a distinct job: echoing the search, stating a benefit, offering proof, making the offer, naming the brand, asking for the action. Near-duplicates waste slots and teach the system nothing. The two display path fields, 15 characters each, add relevance to the visible URL. Keyword insertion can mirror the query in a headline, but odd or misspelt searches then appear in your ad, so set a sensible default and check the search terms report.
The pinning trade-off
Pinning fixes a headline to position 1, 2 or 3, or a description to position 1 or 2; text pinned to headline position 1 or 2, or description position 1, shows every time. Regulated disclaimers, required product wording and sometimes the brand name justify it. Every pin, though, cuts the combinations Google can try and lowers Ad Strength, which Google presents as a best-practice indicator of asset variety. If something must always show, pin two or three alternative versions to the same position: the message always shows and some testing survives. Google allows up to three enabled responsive search ads per ad group, which leaves room to test a genuinely different message separately.
Trademarks, superlatives and claims
Google does not generally stop advertisers bidding on a competitor's trademark as a keyword, but a trademark owner can file a complaint that restricts use of the mark in ad text in many regions. Superlatives and comparisons, such as best, cheapest or fastest, fall under Google's rules on misleading and unreliable claims and under local codes: in the UK, the CAP Code requires evidence for objective claims and fairness in comparisons. Generated copy invents superlatives easily. The studio's brand hub holds approved claims and banned words and enforces them at generation, so an unsupported cheapest never reaches a headline without legal approval.
Relevance beats cleverness
Search ads answer a stated intent, and Quality Score reflects that through expected click-through rate, ad relevance and landing page experience. Headlines that use the searcher's own words, in ad groups built around one tight theme, with a landing page that delivers what the ad promised, beat clever lines that make the reader work. That structure multiplies copy: many ad groups, each needing fifteen distinct headlines, per market. Generation makes that feasible, but localisation cannot be a translation step. A 30-character English headline rarely survives as German or Finnish; each market's headlines should be written to fit from the start.
Updated 25 September 2026