Understand

What gets left out

Three checks run when a beacon is stored. A match is marked and skipped by the Understand reports. The row itself remains.

Latest stored beacons

0 recent beacons. A zero means the rule matched nothing in that set. The rows are still stored.

On each consented beacon, Next.js reads the User-Agent. If its crawler list matches, the row is marked crawler. That list includes names such as Googlebot and GPTBot.
If the browser string contains HeadlessChrome, PhantomJS, Puppeteer, Playwright, or Selenium, the row is marked headless. HeadlessChrome is not on the Next.js crawler list, so this check is separate.
If the same anonymous visitor key already stored 40 beacons in the previous 60 seconds, the new row is marked rate.
Marked rows stay in the collection log. Attribution, funnels, cohorts, anomalies, and the other Understand counts skip them. Pathing, recommendations, and Google Analytics do not read the mark.

This site does not read navigator.webdriver, mouse paths, or screen size to decide. It does not delete the beacon, and it does not block the request.

Reading path transitions…