Measure the tag's real cost on your own storefront

Each site's settings now has a Performance tab where you can measure what the tag actually costs on your own storefront, not just our fixture.

Enter a URL (pre-filled from the site's own address) and click Measure now. We measure the page twice through Google PageSpeed Insights (once with the tag running, once with the tag's own kill switch turned on): the same page, at the same instant, with the tag as the only difference. Each side runs three times and is reduced to a median, so normal measurement noise doesn't skew the result; the individual runs stay visible if you want to check the spread yourself. The delta covers Performance score, CLS, TBT, and FCP.

If the two sides don't hold up as a fair comparison (the kill switch didn't actually suppress the tag, the pages came back with different HTTP statuses, or Google's own measurement was incomplete on one side), the run is marked not comparable and tells you why, instead of publishing a number you couldn't trust.

Alongside the lab comparison, the same run pulls your storefront's real-shopper history from Google's Chrome UX Report: LCP, CLS, INP, and TTFB as weekly 75th-percentile figures, with a collapsible view of past weeks. If your site doesn't have enough real Chrome traffic yet, you'll see that plainly instead of a guessed number.

A run needs its target URL to be on the site's own registered origin allowlist (Settings → Public access).