What ChatGPT and Claude searched in 160 tracked answers

In one seven-day observation of Reputably’s own tracker, 33 of 160 answer records disclosed 69 searches. Those records surfaced 97 distinct domains and cited 57 distinct domains. Reputably’s domain was not among those surfaced or cited domains. This is a defined sample of our own category questions, not a benchmark for every AI user or business.

The observation covers 9 September 2026 at 13:43:41 through 16 September 2026 at 13:43:41 AEST (03:43:41 UTC at both boundaries). It contains collected runs on 10, 14, 15 and 16 September. No records fell inside the window on the other dates; the export does not establish why.

The sample and the collection method

We read Reputably’s stored visibility-run records for its own business. Two parent questions asked about “best social listening tools for leads” and “best ai rank tracking tools.” Each had four variant slots. In the AI-ranking group, two slots repeated the same wording, so the eight slots represent seven distinct input texts. Several variations broadened toward traditional keyword-ranking software.

There were 32 records each for ChatGPT, Claude, Gemini, Google AI Mode and Google AI Overview. ChatGPT, Gemini and the Google surfaces were recorded through DataForSEO; Claude records were labelled direct. Reputably’s documented Claude collection uses its plan-based web-search path. These records are not recordings of personalised consumer sessions.

Missing disclosure is not evidence that an engine performed no searches. We did not impute a search count for undisclosed runs. The 69 occurrences contain 32 distinct exact search strings; these are separate from the seven input prompt texts.

What the repeated searches asked for

The shared theme is evaluation: features, comparisons, reviews and suitability. This supports investigating a focused buyer guide. It does not establish search volume, keyword difficulty or the reason any particular source was selected. The questions were chosen by Reputably and constrain what this sample can reveal.

ChatGPT’s one disclosed record also included searches restricted to Brandwatch, Sprout Social, Meltwater and Talkwalker. Those searches cannot return a Reputably URL. Their presence makes established third-party and brand coverage another plausible area to investigate, rather than a reason to repeat the same keywords more often.

Why 57 cited sites and 530 cited URLs are both correct

Within the 33 records that disclosed searches, we counted 97 surfaced domains from linked query hits plus separately recorded browsed pages, and 57 cited domains from those records’ final citation lists. Domain counts remove a leading “www.” and preserve other subdomains.

Across all 160 records, there were 1,130 citation occurrences and 530 distinct exact URL strings. The URL total includes tracking parameters. Multiple URLs can belong to one domain, and the full cohort includes answers without disclosed searches. These totals have different populations and units.

About the analysis

These findings come from an internal review of Reputably’s own tracking data. The underlying records remain private. Counts use distinct records, exact query strings and exact citation URLs unless a domain measure is explicitly named. The analysis is fixed at the stated extraction time; it is not a continuously refreshed report.

What we will change, and what we cannot conclude

Our next content priorities are a social-listening buyer guide focused on lead generation, clearer distinctions in our AI visibility comparison, and a practical fan-out workflow. These are hypotheses to test with later observations, not demonstrated ranking improvements. They follow Google’s guidance to create helpful content with original value.

Google explains that its AI search features can issue related searches, but our export contains no disclosed Google query trail. We therefore make no claim about the exact Google searches behind these answers. Nor does this sample establish how often other brands are found, what a typical user sees or how quickly new content will rank.