AI rank tracker: track your visibility in AI answers

Why AI answers need their own tracker

A normal rank tracker records your position in Google's list of results. It tells you that you are third for a keyword, or that you moved up two places this week. That is useful, but it does not tell you what an AI assistant says when someone asks a question in plain language.

AI assistants give a written answer instead of a list of links. So the question changes from "what position am I?" to "was I named at all, where in the answer, and who else was named?" A business can rank well in Google and still be missing from the AI answer. The reverse happens too.

Answers also change by assistant, by wording and over time. The same question asked of two different assistants can produce two different sets of businesses. A slight rewording can change who gets named. Because of this, tracking needs repeated checks rather than a one-off look.

For example, a dentist in Leeds might ask an assistant "best dentist in Leeds" one week and "emergency dentist Leeds" the next. The names that come back may not overlap much. Without a tracker, the dentist would never know which questions they appear in and which they do not.

What to track

Not every number matters equally. These are the measures that tell you something you can act on.

• Mention rate: the share of your tracked questions where your business is named. This is your headline number. If you track 20 questions and appear in 6, your mention rate is 30 percent.
• Assistant by assistant: ChatGPT, Gemini, Perplexity, Claude, Google AI Overviews and AI Mode can all differ. One assistant may name you often while another never does. Tracking them separately shows where the gap is.
• Competitors named instead of you: every answer shows who was named. This tells you who is winning the questions you care about, without any extra setup.
• Cited sources: the websites the assistants point to. If a directory or a review site is cited for your questions, that is a place worth being present on.
• How you are described, including mistakes: an assistant might name you but describe you as something you are not, or list an old service you no longer offer. You can only correct what you can see.
• Trend over weeks: a single check is a snapshot. The direction of travel is what tells you whether your changes are working.

For example, a small accounting firm might find it is named by one assistant for "accountant for freelancers" but never by another. That is a specific, fixable gap rather than a vague worry.

Choosing your questions

The questions you track decide what the whole exercise tells you. Choose the questions customers actually ask before they buy, not just your brand name. Tracking your own name is fine, but it mostly confirms that assistants know you exist. It does not show whether you are considered for the decision.

Include location and need in your questions. Real prompts usually carry one or both.

• "best X in [city]" for local intent, such as "best plumber in Bristol".
• "X for [situation]" for a specific need, such as "accountant for freelancers" or "web designer for a small shop".
• "X vs Y" for comparison questions, such as "Shopify vs WooCommerce for a small store".

Start with 10 to 25 questions and keep them stable. If you change the list every week, the trend line becomes meaningless because you are measuring different things. Add questions over time, but keep the original set running so you can compare like with like.

For example, a wedding photographer in Dublin might track 15 questions covering city, style and budget. After a month, the mention rate for the city questions might be higher than for the style questions, which points to where to focus next.

AI tracking and Google tracking together

The same customer question often appears as a Google search and as an AI prompt. Someone types "best CRM for a small agency" into Google, and someone else asks an assistant the same thing. Tracking only one side gives you half the picture.

Tracking both shows where you rank in Google but are missing from AI answers, and the reverse. A page that sits on page one of Google but never gets cited by an assistant is a clear signal. So is a business that gets named by assistants but ranks poorly in Google, which usually means the assistants are drawing on other sources.

Vasukai tracks AI answers and weekly Google rankings for the same searches in one account. It asks the assistants the questions your customers ask, on a schedule you pick, and shows whether you were named, who was named instead, and which websites were cited. Every answer is shown in full so you can read exactly what was said. It also gives a ranked list of actions and a "Write it for me" button that drafts the text for each fix.

For a wider view of how AI visibility and search visibility relate, see GEO vs SEO.