# [AI visibility tracking that measures your own server](https://rebilder.com/solutions/ai-visibility-tracking)

> Most AI visibility tracking samples assistant answers, which is a lagging indicator you cannot move. Your request log is the leading one, and it is exact.

- **Updated:** 2026-08-17

Sampling what assistants say is a survey. Reading your own request log is a measurement. Only one of them tells you what to fix on Monday.



## What you get

- Every agent fetch, by platform, from your own server
- Answer rate: what you served against what you missed
- The miss log, ranked by how often agents asked
- A deterministic readability score you can re-run and compare

## Leading and lagging indicators

Tracking products in this category mostly run prompts against assistants on a schedule and count how often you appear. That number is real, but it is downstream of everything you actually control, and it moves for reasons that have nothing to do with you: a model updates, a prompt is phrased differently, a competitor publishes.

| Signal | Kind | What you do with it |
| --- | --- | --- |
| Agent fetches by platform | Leading, exact | Confirm you are being read at all |
| Answer rate | Leading, exact | Find where you return nothing useful |
| Miss log | Leading, exact | Rank the next fix by real demand |
| Readability score | Leading, exact | Compare before and after a change |
| Mentions in answers | Lagging, sampled | Report a trend, never a target |

The practical consequence: if you buy one thing, buy the leading indicators. A site that is never successfully fetched will show zero mentions forever, and a mention tracker will describe that situation accurately without ever explaining it.

## What gets tracked, and where it comes from

Everything below is derived from requests that hit your own infrastructure, which is why it is exact rather than sampled. There is no panel, no proxy and no third-party measurement in the path.

- **Who fetched you.** Agent traffic classified from real header samples, with the fixture corpus behind the classification kept as data rather than test scaffolding.
- **What you served.** The format, the size, and how long the render took, so a regression shows up as a number instead of a feeling.
- **What you missed.** Requests for URLs that produced nothing. This is the list worth working through in order.
- **Whether crawlers still see HTML.** A standing check that search engines keep receiving canonical markup, so the reassurance is observed rather than asserted.

> **This traffic is invisible to your analytics** Agent fetches run no JavaScript and fire no beacon, so a page-view script cannot record them however it is configured. A tracking plan built on one will report that nothing changed no matter what happened.

## Take the baseline before you change anything

1. **Score one page per template** against the published spec and keep the numbers. This is the comparison you will want in three months and cannot reconstruct later.
2. **Install the gateway** so fetches start being recorded. Free tier is 1,000 agent visits a month, and the SDK itself is unmetered.
3. **Wait a fortnight**, then read the miss log rather than the totals. The totals tell you the weather; the miss log tells you what to build.
4. **Re-score the same pages** after each change, so every claim you make internally is a diff between two deterministic numbers.

## What this does not do

- It does not watch assistants answer questions on your behalf. What it measures is your own server: who asked, what you served, and what you missed.
- It cannot make an assistant mention you. Assistants are not deterministic and no vendor controls their output, so a guaranteed citation is not a thing anyone can sell.
- It does not show you an average position. There is no position in a generated answer to hold, so any dashboard displaying one has invented the number.
- It does not track your competitors’ scores. Nothing about a named third party is published without that domain owner opting in.

## What is AI visibility tracking?

Monitoring how AI systems interact with your site over time. It divides into leading indicators taken from your own server (who fetched you, what you served, what you missed) and lagging indicators sampled from assistant answers.

## Do I need an AI visibility tracker?

Not before you know your pages can be read. The first useful measurement is a free scan of one page per template; a tracker is worth adding once the serving layer is fixed and you want to watch it hold.

## How is this different from a rank tracker?

A rank tracker measures a position in a list. There is no equivalent in a generated answer, so this measures the things that do exist: whether the fetch succeeded, how much you sent, and whether the facts were reachable.

## Can I export the data?

Yes. It is your own request data and it stays exportable. The command-line scanner also ships zero telemetry, so a local audit never uploads a scan, a URL or a hostname anywhere.

## Related

- [AI visibility: the half you control](https://rebilder.com/solutions/ai-visibility)
- [AI visibility tools: how to choose one](https://rebilder.com/learn/ai-visibility-tools)
- [AI brand visibility](https://rebilder.com/solutions/ai-brand-visibility)
- [ChatGPT traffic](https://rebilder.com/solutions/chatgpt-traffic)
- [How the gateway classifies agent traffic](https://rebilder.com/docs/classification)