How to Track AI Overview and ChatGPT Rankings

[Operator’s read] I get asked some version of “what tool tracks my ChatGPT ranking” every week now, usually from an agency owner who just sold a GEO retainer and needs to prove movement. I’ve run all three of the tools below against my own sites for months. None of them is a rank tracker in the SEO sense — there’s no fixed SERP to check position on — but they solve the real problem, which is knowing whether you’re showing up at all.

Table of contents

Open Table of contents

• Why a normal rank tracker can’t do this
• Start manual — it’s a legitimate baseline
• The three tools worth paying for
• What to actually track once you’re paying for it
• Pricing this into a retainer
• The honest limits
• FAQ
  • Do I need a paid tool to sell a GEO retainer?
  • Which of these tracks Claude?
  • Can these tools track my competitors, not just my own site?
  • Is a $29/month tool actually enough, or is it a toy?

Why a normal rank tracker can’t do this

A classic rank tracker works because Google’s results page is a stable, orderable list — position 4 today, position 6 tomorrow. AI answers don’t have that structure. ChatGPT, Perplexity, and Google AI Overviews generate a fresh response per query, per session, sometimes per user. Ask the same question twice and you can get two different sets of cited sources. There’s no “rank” to track — there’s a probability that you get cited at all, for a given prompt, on a given day.

That’s why the tools in this space call themselves AI visibility or citation trackers, not rank trackers, even when their dashboards look like one. What they’re actually doing: running your tracked prompts against each engine on a schedule, logging whether you appear and where, and charting that over time. It’s sampling, not a ground-truth crawl of every real user query — the same honesty caveat I cover in how to measure AI search traffic applies here.

Start manual — it’s a legitimate baseline

Before you pay for anything, run this by hand for two to four weeks:

1. Write down the 15–25 questions your buyers actually ask that your business should be the answer to.
2. Run each one against ChatGPT, Perplexity, and Google AI Overviews (search the query and see if an AI Overview triggers).
3. Log, per question and per engine: cited or not, and roughly where in the answer.
4. Repeat weekly, same day, same time.

A spreadsheet is enough. I recommend this as the standard baseline in GEO audit pricing for exactly this reason — you don’t need a subscription to start proving whether GEO work is landing, and a client can see the method for free before you ask them to pay for the tooling.

Manual breaks down at volume: past 20–30 prompts, or once you’re running this for more than one client, the weekly pass turns into hours you can’t bill. That’s the point to pay for a tool — not before.

The three tools worth paying for

I checked current pricing directly against each vendor’s own site rather than trust a comparison blog, because this category is full of stale numbers.

Tool	Entry price	What entry gets you	Engines tracked	Best for
Otterly.AI	$29/month (Lite)	15 prompts, refreshed on a schedule	ChatGPT, Google AI Overviews, Perplexity, Copilot (Gemini and AI Mode are paid add-ons)	Solo operators, first GEO client
Semrush AI Toolkit	$99/month per domain	25 prompts/day, 300 reports/day	ChatGPT, Google AI, Gemini, Perplexity	Anyone already paying for Semrush
Ahrefs Brand Radar	$50/month (custom prompts) or $199/month (83 prompts/day, 2,500+ checks)	Custom-prompt tier is add-on pricing; the $199 tier is the full dashboard	AI Overviews, Gemini, Perplexity, ChatGPT, Copilot, AI Mode, Claude	Agencies that need Claude coverage

Otterly.AI is the honest entry point. $29/month for 15 tracked prompts across the four engines that matter most (ChatGPT, Google AI Overviews, Perplexity, Copilot) covers a single small client or your own site without forcing you into enterprise pricing you don’t need yet. Gemini and Google’s AI Mode are separate add-ons here, not included — check that before you quote a client “full coverage.”

Semrush’s AI Toolkit only makes sense as an add-on. At $99/month per domain for 25 daily prompts, it’s not cheaper than Otterly on its own — the case for it is that you’re likely already paying for Semrush’s core toolkit for keyword and backlink work, and bolting AI visibility onto a bill you’re already carrying beats adding a fourth vendor login.

Ahrefs Brand Radar is the one to reach for once Claude coverage matters — it’s the only one of the three that tracks Claude natively, alongside AI Overviews, Gemini, Perplexity, ChatGPT, Copilot, and AI Mode. The custom-prompts tier starting at $50/month is cheaper than it looks if you only need a handful of tracked questions per client; the $199/month AI Visibility Index tier is where you go once you want the full dashboard and competitive view.

What to actually track once you’re paying for it

A dashboard full of numbers isn’t the deliverable. Watch four things:

1. Citation coverage — the percentage of your tracked prompts where you’re cited, per engine. This is the number from how to measure AI search traffic that actually moves before traffic does.
2. Position within the answer — cited first versus cited fifth in a list of sources reads very differently to a user, even though a classic rank tracker has no equivalent metric for this.
3. Competitive share — who else gets cited for the same prompts. All three tools show this; it’s the fastest way to show a client movement against a named competitor, not just an abstract score.
4. Week-over-week change — a single snapshot tells you almost nothing in a category where the same prompt can answer differently two days apart. The trend across four to six weeks is the real signal.

Pricing this into a retainer

If you’re running this for clients, the subscription cost has to show up in what you charge — it doesn’t come out of your margin quietly. Ahrefs Brand Radar’s $199/month tier across ten client sites is $1,990/month in tooling alone; that has to be priced into the retainer structure, not absorbed. I cover the retainer math in GEO audit pricing — the short version is that ongoing citation tracking sits at the high end of that formula, 50%+ of the original audit fee, precisely because the tool cost is real and recurring.

Don’t multiply seats blindly, either. Otterly’s per-prompt pricing (15 prompts at $29, 100 at $189) means it’s often cheaper to run one account with a shared prompt budget across small clients than a separate subscription per client, as long as your billing and reporting can keep the prompts attributed correctly.

The honest limits

None of these tools give you a number you can defend as precisely as a Google Search Console position. They’re sampling a fixed prompt list against engines whose answers vary by session, location, and model update — the same limitation covered in where to spend your GEO effort across Perplexity, ChatGPT, and Google AI Overviews. Treat the output as a trend line, report it as a trend line, and don’t let a client push you into promising a specific citation rate by a specific date — nobody selling this tooling can guarantee that outcome, including the vendors themselves.

FAQ

Do I need a paid tool to sell a GEO retainer?

No. The manual spreadsheet method is a legitimate, sellable baseline for the first month or two of any engagement — see GEO audit pricing for how I scope that. Pay for a tool once the prompt count or client count makes the manual pass too slow to run weekly.

Which of these tracks Claude?

Ahrefs Brand Radar, natively, at both its custom-prompts tier and the $199/month AI Visibility Index tier. Otterly.AI offers Claude as a paid add-on rather than including it by default. Semrush’s AI Toolkit tracks ChatGPT, Google AI, Gemini, and Perplexity, but not Claude as of this writing.

Can these tools track my competitors, not just my own site?

Yes — competitive citation share is a core feature of all three, not an upsell. That’s usually more useful for a client conversation than your own coverage number in isolation, because “you’re cited 40% of the time” means little without knowing what the competitor sitting at 70% is doing differently.

Is a $29/month tool actually enough, or is it a toy?

For a single site or your first paid GEO client, it’s enough. Fifteen tracked prompts across ChatGPT, Google AI Overviews, Perplexity, and Copilot is a real baseline, not a demo. Outgrow it when you need more prompts, more clients on one dashboard, or Claude coverage — not before.

---

Related: How to measure AI search traffic · GEO audit pricing: what to charge clients · Perplexity vs ChatGPT vs Google AI Overviews

Want this set up and reported for you? Get in touch — I run GEO consulting projects for operator teams that need to prove AI-search visibility to a client or a boss.