AI search has no page-one ranking to check. Learn what rank means when customers ask AI for recommendations and how to find your position.
By Heather Laskin · Published July 6, 2026 · Updated July 24, 2026
Business owners keep asking a version of the same question: where do I rank in AI search, and how would I even see the number. The instinct comes from years of checking a Google position. AI search does not work the same way, so the number you want does not exist in the form you expect. What exists is better, and you read the answer in an afternoon.
AI search has no page-one ranking to check. Your position is how often AI names your business, how accurately, and whether you land first in the answer. You see these insights by running the prompts your customers ask across AI engines and scoring the results. The pattern shows where you stand and what to fix.
In a Google result, rank is your spot in a list. A customer scrolls until something catches their eye. In an AI answer, there is no list to scroll. The customer asks a question, reads a short paragraph, and acts on the two or three businesses named. Your rank is whether you make the short list, and where you sit inside the answer.
Three numbers describe your position:
A business named first, accurately, on most of its prompts holds a strong position. A business named last, with a wrong service listed, on a handful of prompts holds a weak one. These three numbers, tracked over time, are your rank insights.
You get a baseline by hand. Open a spreadsheet and list the ten questions a customer who wants to hire you would ask. Use the shape "best [service] near me," "best [service] in [your city]," and "[specific high-value service] in [your city]." Run each question through the AI engines your customers use, and for every run record three things: were you named, was the description right, and where did you land.
Add up the results. If you tested ten prompts across four engines, you ran forty checks. Appearing in twelve is a 30 percent mention rate. Now you hold a baseline, and every change you make afterward moves against a real starting point instead of a guess.
A worked example shows the shape. Say a plumber tests ten prompts across three engines, thirty checks, and appears in nine. That is a 30 percent mention rate, almost always behind a competitor, with the service described correctly. After adding a page for each service and asking recent customers for reviews that name the specific job, the next month's run of the same thirty checks shows fifteen appearances and two first-place finishes. Mention rate moved from 30 to 50 percent, and the number, not a guess, tells the plumber the work is paying off.
Mention rate tells you reach. A low number means AI rarely brings you up at all, and the fix is usually structural: missing service pages, a thin Google Business Profile, or reviews with no detail.
Accuracy tells you trust. If AI names you but describes a service you dropped or a city you left, AI is reading stale or scattered signals. The fix is consistency across your site, your profile, and the directories AI reads.
Placement tells you preference. If you appear but land last while a competitor leads, AI has a reason to prefer them: deeper content, more specific reviews, or stronger local signals. Placement is the slowest number to move and the one worth the most once you do.
Three numbers are easy to read one at a time. To track progress, you want them rolled into a single figure you watch month over month. A structured audit does this, combining your results into an AI Visibility Score from 0 to 100.
| Number | What it measures | Weak reading | Strong reading |
|---|---|---|---|
| Mention rate | How often AI names you at all | Named in a handful of prompts | Named in most of your buyer-intent prompts |
| Top-spot rate | How often you are the first recommendation | Almost always behind a competitor | Frequently the first name in the answer |
| Accuracy | Whether the description matches what you do | Wrong service or city listed | Described correctly every time you appear |
A score near the bottom of the range means AI rarely names you, or names you wrong. A score near the top means you are the default answer for your service in your area, described accurately, across the engines your buyers use. The value is not the number itself. The value is watching it climb.
Some answers name a competitor but pull a fact from your site. That is a separate signal worth tracking, called your website-cited rate: the share of prompts where an engine used your page as a source, whether or not it named you. A high cited rate with a low mention rate is a specific, telling pattern. The engine trusts your content enough to read it, but is not yet naming you as the answer. That usually means your pages are useful while your business identity is thin or inconsistent across the web, so the engine borrows your facts and credits someone else.
Sort your prompts by the widest gap. The prompts where competitors get named and you do not are your priority list. For most local businesses, the fixes behind those gaps fall in a predictable order:
Work the list top to bottom. Each fix gives AI more to name you with, and your next measurement shows the effect.
Re-run your prompt set every 30 days. AI answers change as models update and as your own content and reviews change, so a single snapshot ages fast. A monthly rhythm turns a one-time score into a trend, and a trend tells you whether your work is moving your position or standing still.
For a structured version of this measurement across more prompts and more engines, with a repeatable score you track month over month, see a sample AI Visibility report.
See also: The AI Visibility Gap Most Businesses Don't Know Exists for why this gap forms, AI Search for Local and Small Businesses for how AI decides who to name, and the AI Visibility Checklist for the full set of fixes.