What "invisible to AI" means in practice

Being invisible to AI does not mean an assistant has never heard of you. It means that when a buyer asks the question that matters, such as "which firms can help with this" or "what is the best option for that", your brand is not in the answer, or appears so rarely that most buyers never see it.

This matters because the answer is increasingly where the shortlist forms. In G2's 2026 survey of 1,076 B2B software buyers, 85% said they think more highly of a vendor when an AI tool includes it in an answer, and 69% said they had chosen a different vendor from the one they first planned after a chatbot recommendation (G2, 2026). That is a vendor survey of software buyers, so the figures will not transfer exactly to every sector, but the pattern is hard to ignore.

Consumers are moving the same way. BrightLocal's 2026 survey of 1,002 US adults found 45% had used ChatGPT or another AI tool for local business recommendations, up from 6% a year earlier (BrightLocal, 2026). OpenAI said in October 2025 that ChatGPT had more than 800 million weekly users (TechCrunch, 2025).

Why good search rankings are not enough

Search engines rank pages. AI assistants write answers, and they assemble those answers from many sources at once. A page can rank well because it targets a keyword precisely, yet still be vague about what the business does, who it serves and why it is credible. An AI tool summarising a category has little to work with if your pages do not say these things plainly.

AI answers also reduce the value of a ranking. Pew Research Center's study of 900 US adults found that when a Google results page included an AI summary, users clicked a traditional result on 8% of visits, against 15% without one. Only 1% of visits included a click on a link inside the summary (Pew Research Center, 2025). Ranking under an AI summary that does not mention you is a weaker position than it used to be.

There is some evidence that being cited helps. Seer Interactive's analysis of its own client data found that, in the third quarter of 2025, organic click-through rate was about 35% higher on queries where the brand was cited in the AI Overview than where it was not (Seer Interactive, 2025). Seer is clear that this does not prove cause and effect: brands with strong authority may simply be more likely to be cited and clicked.

AI platforms disagree with each other, and with themselves

The main AI tools do not give the same answers, and the same tool does not always give the same answer twice. SparkToro and Gumshoe asked volunteers to run 12 brand-recommendation prompts 60 to 100 times each on ChatGPT, Claude and Google's AI tools, 2,961 runs in total. The same prompt produced the same list of brands less than 1% of the time (SparkToro and Gumshoe, 2026). Gumshoe sells AI tracking tools and the work is not peer-reviewed, so read it as a strong signal rather than a final word.

The same study found something more useful. While the order changed constantly, the set of brands a tool considered for a tight category was fairly stable. In one test, 142 people wrote their own prompts about headphones, with very different wording, yet the leading brands still appeared in 55% to 77% of responses (SparkToro and Gumshoe, 2026). Brands that are strongly associated with a category keep turning up, whatever the phrasing.

The practical consequence is that one test on one platform tells you very little. Test each question several times, on each platform your buyers use, and treat the result as a frequency rather than a yes or no. For background on how this fits with search, read our page on GEO.

The three most common causes

When a brand is missing from AI answers, the cause usually falls into one of three groups. Most brands have some of each, so the audit should check all three rather than stopping at the first problem it finds. The fixes also differ in cost and speed: owned content is quickest to change, outside mentions take longest.

  • Unclear owned content. The site does not state in plain words what the business offers, for whom, where, and with what proof. Service pages are written for keywords rather than for understanding.
  • Thin third-party presence. AI tools draw heavily on what others say. Pew found Wikipedia, YouTube and Reddit were among the most linked sources in AI summaries, and government sites made up 6% of AI summary sources against 2% of standard results (Pew Research Center, 2025). A brand with few mentions in trade press, directories, review sites and community forums gives the AI little to cite.
  • Inconsistent facts. Different descriptions, old addresses, discontinued products or mismatched category labels across the web make it harder for an AI tool to describe you confidently.

Visibility varies by market, language and platform

If you sell in more than one country or language, test the questions in each one. A brand can be well known in English and largely absent in another language, because the sources an AI tool draws on differ by market. Treat each market as its own test, with its own list of questions. Translated pages help, but local press, local directories and local reviews usually matter more, because they are the sources an AI tool finds when it answers in that language.

Platform choice matters too. In the UK, about 30% of Google searches now show an AI Overview, and 53% of adults say they see these summaries often (Ofcom, 2025). Google has said that in markets such as the US and India, AI Overviews drove an increase of more than 10% in usage for the types of queries that show them (Google, 2025). That is Google's own internal data, but it suggests AI Overviews deserve a place in any test alongside standalone assistants such as ChatGPT.

What to fix first

Work through this list in order. The early steps are cheap and help both search and AI answers.

  • Measure. Write down the twenty questions your best customers ask before they buy. Test each one several times in the main AI tools and record who is named and which sources are cited.
  • Fix your own pages. Make sure your site states your offer, your customers, your locations and your proof in plain language, on pages that are easy to crawl. Our content work focuses on this.
  • Make the facts consistent. Update directory listings, review profiles, social bios and partner pages so they describe you the same way.
  • Earn outside mentions. Target the trade press, review sites and communities that AI answers cite in your category. Our digital PR work is built around this.
  • Retest monthly. Repeat the same questions every month and log what changes.

How to measure progress and what good looks like

Measure mention rate for each question: the share of runs in which your brand appears. Track it by platform and against two or three competitors. Alongside that, record whether the description is accurate, and which sources are cited. Watch AI referral traffic in your analytics: Adobe found AI-referred visitors to US retail sites viewed 12% more pages and bounced 23% less than other visitors, although they converted slightly less often (Adobe, 2025).

Good looks like a rising mention rate on the questions that matter commercially, an accurate description on every major platform, and a growing number of credible sources that mention you. When you talk to a supplier, ask how many runs they use per question, which platforms they cover, and how they will link visibility to enquiries rather than reporting mentions alone.

Put this into practice

Book a discovery call and we will review your channels and results first, then agree what to test.

See how your brand appears in AI answers

Questions and answers

Why does one AI tool mention my brand but another does not?

Each tool uses different sources, training data and retrieval methods, so their answers differ. Answers also vary between runs of the same tool. Test several platforms, run each question more than once, and track the results over time rather than relying on a single check.

Does this replace SEO?

No. Search still drives a large share of research, and Google is used by 82% of UK adults according to Ofcom. AI visibility sits alongside SEO, and the two share many of the same foundations, such as clear pages and credible sources.

How quickly can we become visible in AI answers?

Fixes to your own pages and listings can be picked up within weeks by tools that search the web live. Earning third-party mentions takes longer, often several months. Retest monthly with the same questions so you can see which changes are working.

Do reviews affect whether AI tools recommend us?

They can. AI tools draw on third-party sources, and review platforms are among them. BrightLocal's 2026 US survey found 82% of consumers read AI-generated review summaries, so the content of your reviews may shape how you are described as well as whether you appear.

Is ranking position in an AI answer worth tracking?

Not on its own. Research by SparkToro and Gumshoe found AI tools almost never repeat the same list in the same order. How often your brand appears across many runs is a more reliable measure than where it appears in any single answer.

References

  1. G2, "In the answer economy, don't win the click: win the answer", 2026
  2. BrightLocal, "Local Consumer Review Survey 2026", 2026
  3. TechCrunch, "Sam Altman says ChatGPT has hit 800M weekly active users", 2025
  4. Pew Research Center, "Google users are less likely to click on links when an AI summary appears in the results", 2025
  5. Seer Interactive, "AIO impact on Google CTR: September 2025 update", 2025
  6. SparkToro and Gumshoe, reported by Search Engine Journal, "AI recommendations change with nearly every query", 2026
  7. Ofcom, "From apps to AI search: how the UK goes online in 2025", 2025
  8. Google, "AI in Search: going beyond information to intelligence", 2025
  9. Adobe, "Adobe Analytics: traffic to US retail websites from generative AI sources jumps 1,200 percent", 2025

This article is adapted from ContxtAI's article on hidden gaps in AI visibility, rewritten for Insight Labs readers.

← Back to all articles