Which fashion queries trigger AI answers, and what that means for content
AI answers appear far more often on informational searches than on buying searches. For fashion publishers and retailers, that shapes which content is at risk and which is not.
KEY TAKEAWAYS Summary by the editors
- An Ahrefs analysis of 146 million desktop results from September 2025 found AI Overviews on about 20.5 per cent of results overall.
- In that dataset informational queries triggered AI Overviews 21.4 per cent of the time, commercial queries 4.3 per cent, transactional queries 2.1 per cent and navigational queries 0.9 per cent.
- Longer queries trigger AI Overviews more often: 46.4 per cent for queries of seven or more words against 9.5 per cent for single words.
- Semrush found that the share of commercial results with an AI Overview grew 71 per cent between November 2025 and April 2026, so buying queries are not permanently safe.
- The published studies are cross-industry, so fashion teams should test their own query sets rather than assume the averages apply.
Which types of search trigger AI answers most often?
Informational searches trigger AI answers most often, and transactional and navigational searches rarely do. In an Ahrefs study of 146,122,391 desktop search results collected in September 2025, AI Overviews appeared on about 20.5 per cent of results. By intent, the share was 21.4 per cent for informational queries, 4.3 per cent for commercial, 2.1 per cent for transactional and 0.9 per cent for navigational.
These figures are not specific to fashion, they cover desktop results, and they are a snapshot from one month. They are useful as a shape, not as a forecast for a particular retailer.
| Query intent | Fashion example | Share with an AI Overview |
|---|---|---|
| Informational | How should a wool coat fit across the shoulders? | 21.4% |
| Commercial | Best waterproof jackets for hiking | 4.3% |
| Transactional | Buy navy chinos size 34 | 2.1% |
| Navigational | Brand name plus login or store | 0.9% |
Does query length make a difference?
Yes. The same Ahrefs data shows AI Overviews rising steadily with the number of words: 9.5 per cent for one word, 9.9 per cent for two, 14.0 per cent for three, 19.5 per cent for four, 27.6 per cent for five, 36.6 per cent for six and 46.4 per cent for seven or more. Conversational, specific questions are therefore far more likely to receive a generated answer than short product terms.
This matters because AI interfaces encourage longer questions. Google describes AI Mode shopping as letting people describe what they want in natural language instead of using filters or keywords. A shopper who types a full sentence about occasion, budget and fit is exactly the kind of query that the length data associates with more AI answers.
Teams should not read this as a reason to ignore short queries. Short product and brand terms still drive large volumes of search, and the lower AI answer rates there mean that conventional listings and shopping results keep most of their prominence. The point is that the mix is shifting at the conversational end, which is where new shoppers often begin.
Are buying queries really safe from AI answers?
Not necessarily. Semrush analysed more than 600,000 keywords from its US desktop database across 10 industries between November 2025 and April 2026. It reported that the share of commercial results with an AI Overview grew 71 per cent over the period, while the share of transactional results fell 5 per cent. These are changes in share, not absolute levels, and the study does not publish an overall percentage for those intents.
The direction is the point. Commercial research queries, such as comparisons and roundup searches, appear to be where AI answers are expanding, and these are close to the purchase. Retailers that rely on informational or comparison content to attract early-stage shoppers should expect more of that journey to happen inside an answer.
Which fashion query types are most exposed?
The studies do not break results out by fashion, so the mapping below is reasoned from the intent patterns and should be tested. Questions about fit, fabric, care, sizing conversion, dress codes and trends are informational and conversational, which are the profile most associated with AI answers. Brand and category searches with clear buying intent show lower rates in the published data.
- Higher exposure: how to style, what to wear to an event, fabric and care questions, size guides and comparisons between materials.
- Medium exposure: best of lists and product comparisons, where the share of AI answers was growing.
- Lower exposure: navigational brand searches and specific product purchases, though shopping features inside AI assistants are changing how these are fulfilled.
What does this mean for fashion content?
Google says there are no additional requirements to appear in AI Overviews or AI Mode, and that being indexed and eligible for a snippet is the condition. OpenAI says that ChatGPT search responses may include citations, that placement is not guaranteed and that results can be incomplete or outdated. Content strategy therefore rests on being a source worth citing, not on a special format.
- Separate content by intent and look at which pages answer informational questions that may now be summarised.
- Make those pages specific and verifiable: measurements, materials, care instructions and clear authorship.
- Keep product pages and feeds accurate, since shopping surfaces pull from them.
- Track visits and conversions by landing page, not only rankings, because fewer clicks may follow impressions.
- Re-test a fixed set of your own queries every month and note when an AI answer appears and which sources it cites.
Editorial fashion publishers face a sharper version of this problem than retailers. Their pages answer informational questions, which are the profile most often summarised, and their value lies in originality: tested fit notes, reporting and photography that a summary cannot reproduce. Retailers, by contrast, can lean on the fact that buying queries show lower AI answer rates, while recognising that comparison queries are moving. Both should check where their traffic actually comes from before changing strategy.
How should teams test their own query set?
Build a list of 50 to 100 queries drawn from search console data and customer service logs, tag each by intent and length, and record whether an AI answer appears, which sources are cited and whether the brand is named. Repeat monthly, from the same market and device. This produces fashion-specific evidence that the cross-industry studies cannot supply, and it shows where content investment is exposed.
Interpret results with care. A query that triggers an AI answer in one session may not in another, and the answer itself may change. Use the log to see trends in three measures: how often an answer appears, how often the brand is cited and how organic traffic to the same pages moves. If traffic falls where AI answers rise, the case for changing the content on those pages is clearer than any industry average could make it.
Frequently asked questions
What percentage of searches show Google AI Overviews?
An Ahrefs study of 146 million desktop results from September 2025 found AI Overviews on about 20.5 per cent. The rate varied sharply by intent, from 21.4 per cent for informational queries to 0.9 per cent for navigational ones.
Do shopping searches trigger AI Overviews?
Less often than informational ones in Ahrefs data: 4.3 per cent for commercial and 2.1 per cent for transactional queries. Semrush found the commercial share growing 71 per cent between November 2025 and April 2026.
Do longer search queries get more AI answers?
Yes. In the Ahrefs data the share rose from 9.5 per cent for one word queries to 46.4 per cent for queries of seven or more words.
Do these figures apply to fashion?
Not directly. The cited studies cover many industries on US or desktop data, so fashion teams should run their own tests on a fixed set of queries.
One edition every weekday morning. Read in five minutes. Free for industry professionals.
