One of the most common questions we are asked in 2026 is some version of "how do I track my ChatGPT rankings over time?" It is a reasonable question, and it is also slightly the wrong one, because ChatGPT does not have rankings in the way Google does. Understanding why is the first step to measuring AI visibility properly.
If you have not yet checked whether you appear in AI at all, start with how to check if your business shows up in AI, then come back here to turn that one-off check into ongoing measurement.
Why ChatGPT has no "ranking position"
Google returns a ranked list. Position 1 is position 1, it is stable for hours or days, and a rank tracker can read it. ChatGPT returns a generated answer. Ask the same question twice and you can get two different responses, with different brands named and different sources cited. There is no slot to occupy and no number to track. So the familiar idea of "rank 3 for this keyword" does not transfer.
What does transfer is the underlying question a business actually cares about: when someone asks an AI for a recommendation in my category, how often does my brand come up, and how is it described? That is a measurable thing. It is just measured as a rate, not a position.
In Google you track position. In ChatGPT you track presence rate: how often you appear across repeated runs of the questions your customers ask.
What to actually measure
A useful AI visibility measurement has four parts, and none of them is a single position number:
Presence rate. Out of all the relevant prompts you test, in what percentage does the AI mention your brand at all? This is the headline number and the one to trend over time.
Citation rate. Of those mentions, how often are you actually linked or named as a source, rather than just referenced in passing? Citations are worth more than mentions.
Share of voice. When competitors are named alongside you, what is your share of the recommendations? This tells you your position relative to the market, not in absolute terms.
Framing and accuracy. How does the AI describe you? Correct, current, and flattering, or out of date and vague? An inaccurate description is a visibility problem even when you are mentioned.
How to build a tracking system you can repeat
The method is straightforward, and you can start it manually in a spreadsheet before reaching for any platform.
1. Define a fixed prompt set. Write down the real questions your customers ask, the ones that should surface you. "Best [your service] in [your city]", "who should I use for [problem]", "alternatives to [competitor]". Keep this set stable so your measurements are comparable over time.
2. Sample, do not snapshot. Because answers vary, run each prompt several times in fresh sessions and record how often you appear. One run tells you almost nothing; ten runs tell you a rate.
3. Use clean sessions. Turn off personalisation, memory and account history, or you will measure your own relationship with the tool rather than what a new customer sees.
4. Track across engines. Run the same set through ChatGPT, Google's AI Overviews and AI Mode, Perplexity, Gemini and Claude. A brand can be strong in one and absent in another.
5. Log it on a schedule. Weekly or fortnightly is enough for most businesses. The single most valuable output is a trend line: is presence rate climbing, flat, or falling?
What actually moves your AI visibility
Once you are measuring, the obvious next question is what to change. The factors that influence whether an AI surfaces and recommends a brand are closer to entity architecture and digital PR than to classic keyword SEO:
Entity clarity. AI systems need to understand who you are as a distinct entity. Clean, consistent structured data, a clear identity across all your properties, and unambiguous naming all help the model resolve and trust you.
Third-party citations. Mentions on sources the model already trusts, industry publications, directories, credible editorial, carry far more weight than anything you say about yourself.
Consistency everywhere. If your name, role and description differ across your website, LinkedIn, and other profiles, you give the model conflicting signals. Identical, repeated descriptions reinforce the entity.
Crawler accessibility. If AI crawlers such as GPTBot and Google-Extended cannot reach your content, you are invisible to the systems that read the web to answer questions. This is a common, silent blocker.
Measuring presence across engines, repeatedly, over time, and turning it into a single trend is exactly what the Periodic Table of Digital Authority™ (PTODA) framework was built to do, and what OG01 automates as a repeatable Authority Rating. You can run the manual version above today; the platform version removes the spreadsheet.
The honest summary
You cannot track ChatGPT rankings, because there are none. You can track ChatGPT visibility, and you should, because it is measurable, it trends, and it responds to work. Define your prompts, sample repeatedly across engines, log the rate over time, and fix the entity and crawler signals underneath. That is how you turn "are we showing up in AI?" from a guess into a number.
Book a diagnostic to measure how often ChatGPT, Gemini, Perplexity and Google's AI surfaces recommend your brand, and what is holding your AI visibility back.
Tracking ChatGPT & AI rankings
Doug Lord is a Digital Authority & AI Visibility Strategist and founder of Digital Dominator. He created the Periodic Table of Digital Authority™ (PTODA), an independent research framework for measuring digital authority, AI visibility and crawler accessibility, and is co-founder of OG01, where he serves as COO and CPO.