Nostimates

Google AI Overviews answer text tracking

google_ai_overviewsanswer_textAnonymous

Every AI Overview summary we collect is stored verbatim. When a client asks why their share moved, the answer is the paragraph, not the percentage.

How it renders

The summary block itself, captured before parsing.

Sampling

Every sample is stored, not just the modal one. Divergence between samples is the raw material for the volatility metrics.

Gotcha

AI Overviews frequently rewrites the same answer between samples with no change to the citation set. Storing text is what lets you prove that a share change was framing, not sourcing.

What the API returns

A single collection call against Google AI Overviews for the answer text field, sampled n times and returned as one object per prompt.

{
  "engine": "google_ai_overviews",
  "field": "answer_text",
  "prompt": "best ai visibility tracking tool",
  "country": "GB",
  "language": "en",
  "n": 30,
  "samples": [
    {
      "sample_index": 1,
      "tokens": 214,
      "text": "Several APIs return AI answer data…",
      "captured_at": "2026-08-03T09:14:22Z"
    }
  ]
}
GET /v1/collect?engine=google_ai_overviews&field=answer_text

Response fields

KeyTypeNotes
textstringVerbatim answer, markup stripped.
htmlstringOriginal markup, for citation position reconstruction.
tokensintegerApproximate length, useful for drift detection.
captured_attimestampUTC collection time for this sample.
sample_indexintegerWhich of the n samples this is.

What it costs

Included. Evidence retention defaults to 90 days and is configurable.

See credit pricing →

Google AI Overviews answer text FAQ

Do you store every sample or just one?
Every sample. The divergence between them is the data, not noise to be averaged away before storage.
How long is answer text retained?
Ninety days by default, configurable per project for clients with longer reporting cycles.

Ship evidence with every AI Overview number

Agencies win renewals by showing the sentence. Let's scope evidence retention for your prompt set.

No newsletter. One reply from a human.

Across every engine

Compare how answer text behaves on every engine we track.

Answer text across all engines →