SEsearchable
Traffic

Get crawl-to-visit correlation

Changed on

Per page: how often AI crawlers fetched it versus same-window AI-referred human sessions. groupBy=platform|botCategory|botName adds the requested crawl-side breakdown; it does not attribute sessions to a crawler. Rows are ranked by AI-referral sessions. limit defaults to 50, max 200.

get/api/mcp/projects/{projectId}/traffic/correlation

Request

  • Base URL
    https://app.searchable.com
  • Auth
    HTTP bearer

Path parameters

projectIdstring uuid required

The project id.

Query parameters

daysinteger

Lookback window in days. Default and maximum vary by endpoint — see the operation description.

fromstring
Example:2026-03-01

Start of an absolute window (inclusive, UTC). Accepts YYYY-MM-DD or a full ISO timestamp. Requires to, and is mutually exclusive with days — sending both returns 400 invalid_argument. A range longer than the endpoint's maximum is rejected rather than clamped.

tostring
Example:2026-03-31

End of an absolute window (inclusive, UTC). Accepts YYYY-MM-DD or a full ISO timestamp. Requires from.

hoststring
Example:blog.acme.com

Comma-separated hostnames to narrow the report to. Useful on multi-domain projects; omit for every host in the project.

groupBy'platform' | 'botCategory' | 'botName'

Optional crawler-breakdown dimension on traffic overview, bots, and correlation. Omit for the legacy response. platform requests byPlatform, botCategory adds byCategory, and botName adds byBot. Overview and bots retain their existing byPlatform array. These rows classify crawl requests only; AI-referral sessions are not attributed to a crawler. Unknown values return 400 invalid_argument.

limitinteger

Page size. Defaults and caps vary by endpoint (see each operation's description); big list endpoints (cited domains/URLs, prompts, opportunities) cap at 1000, issues at 5000. Values above an endpoint's cap are clamped server-side, never rejected.

Response

Per-page crawl and AI-referral correlation rows.

groupBy'platform' | 'botCategory' | 'botName'

Present when the request selected a crawl-side grouping.

pagesConsiderednumber required

Size of the ranked population — a floor on the true page count, not a site-wide total.

populationTruncatedboolean required

True when the upstream per-set cap (500) was reached, so pages are certainly missing.

Changes