AI Visibility / GEO
Search queries
See the web searches AI models run while answering your prompts
View as MarkdownWhen an AI model answers a prompt, it rarely answers from memory. It first runs its own web searches, reads what comes back, and writes the answer from that. Those searches decide which pages the model reads, and therefore which brands can appear in the answer at all.
The Search queries page collects them. Every search your project's prompts triggered, across every model and location you track, in one list.
Search queries and prompts are not the same thing#
A prompt is the question you configured. A search query is what the model typed into a search engine while working on it.
One prompt usually produces several search queries, and the model writes them itself. It rephrases, splits a broad question into narrower ones, and adds words you never used. That gap is the point of the page: prompts tell you what you asked, search queries tell you what the model went looking for.
What each row shows#
| Column | What it means |
|---|---|
| Search query | The search itself. The most frequently run wording names the row. |
| Times run | How often models ran it in the selected period, with the change against the period before. |
| Phrasings | How many near-identical wordings are grouped into this row. |
A row with more than one phrasing can be opened. Underneath you see each wording and how often it ran on its own.
How grouping works#
Models rarely phrase the same search identically twice. Left ungrouped, one intent would fill dozens of rows that differ by a plural or a reordered word.
aclipp compares the searches in the selected period and puts near-identical wordings in one row, named after the most frequently run of them. Wordings only group when they are similar to every other wording already in the row, not just to the one that named it. This keeps a long search and a short search out of the same row when they only have a middle wording in common.
Grouping is deliberately cautious. Two searches that mean the same thing in different words, such as "aclipp reviews" and "aclipp experiences", stay in separate rows, because nothing in the wording says they belong together.
Filters and export#
The date range, AI model, location and topic filters at the top of the page all apply. The search box narrows the whole list rather than only the rows currently on screen, so a search reaches queries further down than the page displays.
The export gives you the ungrouped searches with their run counts, as CSV or Excel. Grouping exists to make the page readable, and a spreadsheet does not need it.
How to use it#
- Find the wording your market is actually searched with, and write for those words rather than your own.
- Watch for search queries that are new in a period. They show a topic entering the model's research before it shows up anywhere else.
- Compare a search query against the domains and URLs cited for the same period to see which pages that search surfaced.