In March 2026, we upgraded our measurement methodology to give you an even more accurate view of consumer behavior, ad performance, and sales.
Our goal is to give you the best in class measurement methodology. With these changes, we are verifying which products actually came into a consumer’s view and converted to sales. This helps you tell a clearer performance story and keeps us aligned with industry standards and our MRC‑accredited approach to measurement.
Ad campaign impact
With more refined data, we’re counting more valid impressions and no longer giving credit to accidental or low‑quality clicks that rarely led to sales. The clicks that remain are higher‑intent and better aligned with real consumer behavior.
Because of this methodology change, you may notice an impact to the following metrics—
Lower Click through rate (CTR)
Higher Return on ad spend (ROAS)
Note: Previously, certain order adjustments (ex. replacements, out-of-stock, returns) could result in ad-driven sales being undercounted sales. By capturing orders at the time they’re placed, we now capture the order value more consistently, providing a more complete view of ad impact.
Impressions: We now require viewable impressions before counting an ad unit click. Previously, accidental clicks (ex. Viewed while fast-scrolling) might have been counted even if the ad wasn't viewed. We have filtered these out to align attribution with viewability.
View and clicks: Lower quality clicks and views (non-viewable clicks that likely never led to conversions) disappear from your reporting. The remaining clicks represent higher-intent shoppers.
We’re not changing auctions, delivery, optimization, or how budgets are spent. We’re changing the rules and data we use to count impressions, clicks, and sales so reporting lines up better with what is already happening on the platform.
Your campaigns are performing the same. We’re simply improving our measurement methodology to align to our commitment to industry-leading measurement.