You have asked ChatGPT to recommend businesses in your industry. Maybe your company appeared, maybe your competitors did instead.
But what happens next?
Checking one AI response can be interesting, but it does not tell you whether your business has strong AI visibility.
If you want to understand whether ChatGPT, Gemini and other AI search experiences are helping people discover your business, you need to measure it properly.
That means looking beyond a single prompt and tracking how often your business appears, what is being said about you, whether your website is being cited, and, most importantly, whether that visibility is contributing to website visits, enquiries, and sales.
At Outrank, we believe AI visibility should be treated like any other part of digital marketing. If you are investing time and budget into improving it, you should be able to measure what is changing.
What Does AI Search Visibility Actually Mean?
AI visibility is not one single number, a business might be mentioned in an AI response without receiving a link. Another might be cited as a source but not directly recommended. A third could appear in a list of suggested businesses and receive website traffic as a result.
All three have some level of visibility, but they are not the same thing; this is why it helps to separate AI visibility into different signals.
A mention is when your business or brand appears within an AI-generated response.
A citation is when your website is referenced or linked as a source supporting the answer.
A recommendation is when the AI actively suggests your business to the user as an option.
You can then look at what happens after that visibility. Does someone click through to your website? Do they enquire? Do they eventually become customers?
Those later stages are where AI visibility starts becoming commercially meaningful.
Why One ChatGPT Search Is Not Enough
One of the easiest mistakes to make is testing a single prompt and treating the result like a Google ranking.
For example, you might ask:
“Who are the best accountants in Newcastle?”
If your business appears, it can feel like a win, if it does not, it can feel like a problem, neither conclusion is particularly reliable on its own.
AI-generated responses can vary depending on the exact wording of the question, the context of the conversation and the information available to the system at that moment.
Instead, treat AI visibility as something you measure across a group of relevant questions and over a period of time.
Rather than asking one generic question, test the different ways a genuine customer might search for your business.
A roofing company, for example, could monitor questions around roof repairs, new roofs, emergency roofing, flat roofing, costs and location-based recommendations.
That gives you a far more useful picture of your visibility than repeatedly asking the same broad question.
Start With a Set of Real Customer Prompts
Your measurement should begin with search intent. Think about what someone might genuinely ask ChatGPT, Gemini or another AI tool when they are researching a product or service like yours.
For a local business, that might include:
“Can you recommend a reliable roofing company in Middlesbrough?”
“Who installs flat roofs in the North East?”
“Which roofing companies have good reviews near me?”
“Who should I contact for an emergency roof repair in Stockton?”
For an ecommerce business, the questions could be very different.
“What is the best mattress for a side sleeper?”
“Which UK companies sell zip and link beds?”
“Where can I buy a rolled mattress with delivery in Yorkshire?”
The purpose is not to create hundreds of artificial prompts; start with the questions that represent genuine commercial opportunities for your business.
We would normally recommend grouping these around your main services, locations, customer problems and comparison searches, that gives you a consistent set of prompts you can revisit over time.
Measure Your Mention Rate
Your mention rate is one of the simplest AI visibility metrics.
It answers:
How often does my business appear when I test relevant prompts?
Imagine you test 50 prompts and your brand appears in 15 responses, your mention rate would be 30%.
You can then repeat the same test later and see whether visibility has changed, this becomes particularly useful when you compare your business against competitors.
If your business appears in 30% of relevant responses but your main competitor appears in 65%, you have a much clearer picture of where the opportunity sits.
The goal is not necessarily to appear in every answer, some prompts simply will not be relevant to your business.
The important thing is understanding whether you appear consistently when your products, services, locations and expertise genuinely match what the user is asking for.
Track Your Recommendation Rate
A mention is useful, but a recommendation is stronger, there is a difference between an AI tool saying:
“Businesses operating in this sector include Company A, Company B and Your Business.”
and:
“Based on your requirements, your business could be a good option because…”
The second response is much closer to commercial intent. Your recommendation rate measures how frequently your business is actively suggested as a suitable option.
This should be tracked separately from simple mentions, over time, you can begin to understand which types of questions are most likely to generate recommendations and where competitors are outperforming you.
Measure Citations and Website References
AI-generated answers can include links to websites used as supporting sources.
For businesses, citations can be particularly useful because they show that your website is contributing information to an AI response, even when your company is not directly recommended.
OpenAI explains that ChatGPT Search can include citations and links to web sources, while websites need to allow OAI-SearchBot if they want their pages to be eligible for inclusion in ChatGPT search results, placement is not guaranteed.
That means another useful question is:
How often is my website being used as a source?
Track the page being cited as well as the topic of the query.
You may discover that one service page or guide is frequently referenced, while another important area of your website receives little visibility, which creates a useful content opportunity.
Do Not Confuse Citations With Rankings
Traditional SEO has conditioned businesses to think in terms of positions.
Position one.
Position three.
Page one.
AI search does not always work in the same way.
Microsoft makes this distinction clear in its AI Performance reporting. Bing Webmaster Tools can show how often a website has been cited in supported AI experiences, but it specifically states that citation counts do not indicate ranking, page importance or placement within an answer.
This is an important distinction; being cited is evidence that your content is being used.
It does not necessarily mean the AI system considers your website the “number one” source.
Instead of forcing AI visibility into a traditional ranking model, measure the different signals separately.
That gives you a much more accurate view of what is happening.
Compare Your Visibility Against Competitors
AI visibility makes much more sense when you put it into context. If your business appears in 20% of the questions you test, is that good?
The answer depends on what is happening around you. If your nearest competitors appear in 5%, you may already have strong visibility. If they appear in 70%, there is clearly more work to do.
Track the businesses that repeatedly appear alongside you. You may also discover competitors you had not previously considered.
AI-generated recommendations can expose companies that compete for the same customer questions, even if you do not consider them direct traditional search competitors.
Look at who is being recommended, which businesses are cited and which websites appear to support those recommendations.
This can help shape future SEO, content, digital PR and reputation work.
Check Whether the Information Is Accurate
Visibility is not automatically positive, an AI tool could mention your business but display outdated, incomplete or incorrect information.
That could include an old location, services you no longer provide, incorrect opening hours or a poor understanding of what your company actually specialises in.
Accuracy should therefore form part of your measurement.
When your business appears, check whether the response correctly describes:
- Your products and services.
- The locations you cover.
- Your specialist areas.
- Your contact information.
- Your brand positioning.
- The information should reflect the business you actually operate today./li>
If the same inaccuracies continue appearing, investigate where those details might be coming from across your website and wider online presence.
Measure Your Share of AI Visibility
Once you have enough data, you can start looking at your overall share of visibility.
Instead of simply asking whether your business appears, calculate how much of the relevant AI landscape your brand occupies compared with competitors.
Microsoft introduced Citation Share within Bing Webmaster Tools in 2026. It shows the percentage of citations attributed to a website for a particular grounding query compared with all citations shown across sites for that same query. Microsoft describes it as an observational metric rather than a ranking system.
You can apply a similar principle more broadly to your own testing.
If five businesses consistently dominate the answers around your most commercially important topics, understanding your share among those businesses can be more useful than looking at an isolated mention count.
Use Google Search Console to Measure AI Search Visibility
Google has made measuring AI visibility significantly easier.
In June 2026, Google introduced dedicated Generative AI performance reporting within Search Console, and by 31 August 2026 those reports had been rolled out worldwide.
The report covers visibility within Google AI Overviews and AI Mode.
Businesses can use it to monitor how often their pages receive impressions within Google’s generative AI features, which pages are appearing, where those impressions originate and how performance changes over time.
This gives you much stronger evidence than manually searching Google and hoping to see your website within an AI Overview.
Google also continues to include AI Mode and AI Overview activity within the wider Search Console performance data.
It is worth making an important distinction here. This reporting relates to generative AI features within Google Search.
It is not the same thing as measuring whether your business is being mentioned inside the standalone Gemini experience.
Different AI platforms therefore still require different measurement approaches.
Track AI Referral Traffic in Google Analytics
Visibility is only one stage of the customer journey.
The next question is:
Are people clicking through to my website from AI platforms?
Google Analytics introduced a dedicated AI Assistant channel in May 2026.
Traffic from recognised assistants such as ChatGPT, Gemini and Claude can now be grouped under the AI Assistant channel, with an ai-assistant medium used for recognised referral traffic.
You can use this to monitor how many visitors reach your website from AI assistants and how that audience behaves once they arrive.
Google’s current channel definitions also make a useful distinction: the AI Assistant channel covers external assistants such as ChatGPT, Gemini, Copilot and similar platforms, while traffic from Google’s AI Overviews and AI Mode is not included in that particular AI Assistant category.
This means you should avoid treating every AI-driven interaction as one channel.
Google AI Search visibility, external AI assistant traffic and traditional organic search may appear differently within your reporting.
Measure What AI Visitors Do Next
Referral traffic alone does not tell you whether AI visibility is valuable.
A website could receive thousands of AI-referred visits and generate no commercial return.
Another could receive only 50 visits and generate several strong enquiries.
Look at the actions those visitors take.
- Do they view important service pages?
- Do they complete an enquiry form?
- Do they call?
- Do they purchase?
- Do they return later through another channel and convert?
- This is where your existing GA4 conversion and event tracking becomes important.
AI visibility should ultimately connect back to the same commercial measures as the rest of your marketing.
Leads.
Sales.
Revenue.
Appointments.
Qualified enquiries.
Whatever matters to your business.
What Should Your AI Visibility Report Include?
We would avoid presenting clients with one vague “AI score”.
Instead, build a simple dashboard that tracks the different parts of visibility individually.
A useful monthly AI visibility report could include your mention rate, recommendation rate, citation rate, competitor visibility, accuracy of information, prompt coverage, AI referral sessions and conversions originating from AI traffic.
It should also show change over time.
The important question is not just:
“Are we visible?”
It is:
“Is our visibility improving in the areas that matter to the business?”
That distinction keeps the reporting connected to strategy rather than turning AI visibility into another vanity metric.
Can Bing Webmaster Tools Help Measure AI Visibility?
Yes, although much of the current conversation focuses on ChatGPT and Gemini, Bing Webmaster Tools now provides some of the most detailed first-party AI visibility reporting available to website owners.
Its AI Performance dashboard can show total citations, cited pages, grounding queries and changes in citation activity over time across supported Microsoft AI experiences.
Microsoft has continued expanding the reporting throughout 2026, adding data around intents, topics, Citation Share and competitor comparison.
In October 2026, it also introduced Topics Portfolio in beta, which groups the queries where a website appears into broader subject areas and identifies potential opportunities where other websites are appearing but yours is not.
For businesses already using Bing Webmaster Tools, this gives you another useful layer of evidence around how your website is participating in AI-generated answers.
Should You Use AI Visibility Tracking Tools?
Third-party platforms are increasingly offering AI visibility tracking.
These can be useful when you need to test large numbers of prompts, compare competitors and monitor multiple AI platforms consistently.
But treat the numbers carefully, Google itself warns businesses to be wary of tools claiming access to internal Google AI or ranking metrics. Third-party platforms can help organise and monitor testing, but they do not have access to Google’s internal ranking systems.
The same principle applies more broadly, use tracking tools to identify patterns, do not mistake a proprietary visibility score for an official ranking metric.
The methodology behind the score matters.
Ask what platforms are being tested, how often prompts are repeated, what counts as a mention and whether the responses reflect the locations and customer journeys that matter to your business.
How Often Should You Measure AI Visibility?
AI search changes quickly, but checking your prompts every day is unlikely to tell you anything useful.
For most businesses, consistency matters more than frequency, establish a baseline first.
Use a fixed group of commercially relevant prompts and record your current mentions, recommendations, citations and competitors.
Then repeat the process using the same methodology, monthly reporting is a sensible starting point for many businesses because it gives you enough time to make changes and observe meaningful patterns.
If you are actively testing a new content strategy or conducting an AI visibility experiment, you may want to monitor more frequently.
The important thing is to keep the methodology consistent, changing your prompts, platforms and scoring system every month makes comparisons far less useful.
What Does Good AI Visibility Look Like?
There is no universal percentage that defines good AI visibility, a local roofing company and a national ecommerce retailer operate in completely different search environments.
Your benchmark should therefore begin with your own market.
- Are you appearing more frequently for relevant commercial questions?
- Are you closing the gap on competitors?
- Are AI systems describing your business accurately?
- Are more of your important pages being cited?
- Is AI referral traffic increasing?
- And does that traffic contribute to meaningful actions?
Those questions provide a much more useful definition of success than trying to achieve an arbitrary AI visibility score.
What Should You Do If Your AI Visibility Is Low?
Do not start by creating dozens of pages designed purely for ChatGPT.
Start with the basics, lok at the prompts where competitors appear and you do not, then review whether your website genuinely provides the information needed to answer those questions.
- Is the relevant service clearly explained?
- Does your website demonstrate expertise?
- Do you have useful evidence, case studies and reviews?
- Is information about your locations and services consistent?
- Can search engines and AI crawlers access your important pages?
Google’s current guidance remains clear that strong SEO, valuable original content and a sound technical structure remain the foundation for visibility within its generative AI search experiences.
AI visibility should strengthen your wider search strategy, not pull you away from it.
From AI Visibility to Commercial Results
AI search measurement is still developing.
The reporting available today is already far stronger than it was even a year ago, and platforms are continuing to provide businesses with more information about how content appears within AI-generated experiences.
But the principle should remain simple, do not measure AI because AI is fashionable.
Measure it because customers are using these platforms to discover, research and compare businesses.
Track whether you are appearing, understand why competitors are appearing, measure whether customers reach your website, then connect that visibility to enquiries, sales and growth.
That is when AI visibility becomes useful marketing data rather than another number on a dashboard.
Want to Know How Visible Your Business Is in AI Search?
At Outrank, we help businesses understand how they appear across traditional search and emerging AI search experiences.
We can look at your visibility across relevant customer questions, compare your presence against competitors and identify where your website, content and wider online authority could be strengthened.
If you want to understand whether ChatGPT, Gemini and AI Search are finding your business, speak to the Outrank team.