Thanks for the clarification. This helps a lot. Regards, Vibhuti
On Tue, 3 Nov 2026 at 20:20, Ian Clelland <[email protected]> wrote: > Hi Vibhuti - > > I haven't been involved in the origin trials program for quite some time > now, but I don't think the criteria has changed - usage is based on feature > use, not on the presence or absence of the token in headers. It should be > fine, for most trials, to ship the header on every page view, as long as > you're not triggering the feature on every page load. We count the > number of page loads where the feature is used at least once. > > That said, the 1% limit applies across *all* page loads measured by > Chrome; generally you would have to be an extremely large popular site to > trigger that by yourself. A very large number of smaller sites could > potentially cause the trial to reach this limit as well, though that is > less likely for an origin trial, given that each site has to sign up for > their own token. > > A rough but mostly-correct proxy for this calculation is the feature > popularity metrics we publish at > https://chromestatus.com/metrics/feature/popularity - if that number were > to climb above 1% for a feature while it was still in an origin trial, that > would likely trigger the usage limits. > > This is mostly documented in > https://github.com/GoogleChrome/OriginTrials/blob/gh-pages/developer-guide.md#19-are-there-any-usage-limits-on-experimental-features > > Hope this helps, > Ian > > On Tue, Oct 20, 2026 at 11:57 AM Vibhuti Mahajan < > [email protected]> wrote: > >> Hi Blink Dev Community, >> >> We're planning to enable an Origin Trial for our web application and >> would appreciate clarification on how the usage restrictions are enforced. >> >> We have two questions: >> >> 1. >> >> *What counts toward the usage limit?* Is usage measured based on >> pages receiving a valid Origin-Trial response header, or only when >> the experimental API is actually invoked? Specifically, can we send the >> token to all users but restrict actual feature usage to a small subset, or >> should we limit token distribution itself? >> 2. >> >> *How is the usage percentage calculated?* Our trial disclosure >> mentions a 1% usage limit. Is this calculated against our application's >> traffic, or against total Chrome page loads globally? >> >> Any clarification or pointers to relevant documentation would be greatly >> appreciated. >> >> Thanks! >> >> -- >> You received this message because you are subscribed to the Google Groups >> "blink-dev" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion visit >> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/a886dbf9-b11b-4290-a518-30d1bdbe17b4n%40chromium.org >> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/a886dbf9-b11b-4290-a518-30d1bdbe17b4n%40chromium.org?utm_medium=email&utm_source=footer> >> . >> > -- You received this message because you are subscribed to the Google Groups "blink-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAND%3DjrkaWp7nSFMis316qio5gMAwk7bvF9UqTPGrojYyvFQfXQ%40mail.gmail.com.
