Attribution in 2026: what changed for measurement software

These dated changes moved what measurement software can read, and none of them changes what buyers did. Meta said its Ads Insights API would stop returning its 7-day view and 28-day view windows on 12 January 2026, and announced on 3 March 2026 that click-through would count only link clicks for website conversions. GA4 let you set attribution per conversion from 16 January 2026 and any click lookback from 1 to 90 days from 11 August 2026. Compare the same metric for four weeks either side of each date before you blame a channel.

What changed in how Meta reports attribution?

Meta's developer blog said on 16 October 2025 that from 12 January 2026 the Ads Insights API would no longer return the 7-day view and 28-day view windows. The 1-day click, 7-day click, 28-day click, 1-day engaged view and 1-day view windows would remain. The same post said unique-count and hourly breakdowns would be limited to 13 months of history and frequency breakdowns to 6 months, while totals stay available for up to 37 months. A connector that asks for a removed window, or for more history than those limits allow, won't get it.

On 3 March 2026 Meta announced that click-through attribution for website and in-store conversions would count link clicks only. Conversions from a share, save or other non-link click action move into engaged-view attribution, which Meta renamed engage-through, and an engaged view of a video ad now takes 5 seconds instead of 10. Meta said the changes would begin later that month and could reach advertisers at different points.

Meta's stated reason is alignment: it says many third-party platforms primarily attribute only to website link clicks, and that the difference can make Ads Manager and third-party tools disagree. The announcement gives no size for the effect on reported conversions, so measure it in your own account.

What changed in GA4?

GA4's release notes list four changes that touch attribution:

• 16 January 2026 (Beta). Conversion attribution settings became adjustable independently for every conversion, and a Conversion attribution analysis report appeared in the advertising workspace.
• 13 May 2026. Traffic from recognized AI assistants got its own AI Assistant channel in the Default Channel Group, so totals for other channels can move when you compare across that date.
• 11 June 2026. A new Source Group field consolidates source values for Facebook, Instagram and TikTok, and some classifications in the existing Source Platform field were updated, so check any report that groups by it.
• 11 August 2026. Click-through conversion windows became any integer from 1 to 90 days, where they had been limited to preset values of 1, 7, 14, 30, 60 or 90, and engaged-view windows became any integer from 1 to 30 days, where they had been fixed at 3.

Google notes that the Beta features may not be available to every property.

What changed in the browser?

On 17 October 2025 Google said it would retire the Attribution Reporting API in Chrome and Android, along with other Privacy Sandbox technologies, after evaluating ecosystem feedback and in light of their low levels of adoption. The post recalls that Chrome will maintain its current approach to offering users third-party cookie choice, and says Google will keep working on an interoperable attribution standard with other browser makers through the web standards process. Anything built on that API has to be rebuilt on something else.

What do the platforms now say about experiments?

Both platforms also pushed experiments. Meta's announcement of 3 March 2026 says incrementality experiments are the best way to answer what an ad or campaign caused that would not have happened otherwise, and calls them the gold standard in measurement. Google Ads said on 11 November 2025 that one incrementality experiment which once might have cost upwards of $100,000 can now be done for $5,000. Both statements are published by companies that sell the ads and aren't independently audited.

How do I find out which change moved my numbers?

1. List the change dates above that touch your setup.
2. For each date, pull four weeks before and four weeks after, using the same weekdays, for three counts: Shopify orders, GA4 purchases and the platform's own reported purchases. Keep the attribution setting fixed across the period.
3. Divide the platform's purchases and GA4's purchases by Shopify's orders for each week.
4. Pass, a reporting change: the ratios step up or down on the date while Shopify orders stay on their usual path. Fail, not a reporting change: orders moved and the ratios did not, which points at demand. If both moved, widen the window before you conclude anything.

Write each date next to your chart so the next review starts from it. Neither Meta nor Google publishes how much these changes moved advertisers' reported conversions, and this post cites no such figure.

Sources, 30 September 2026: Ads Insights API Metric Availability Updates (Meta for Developers, 16 October 2025); Simplifying Ad Measurement for a Social-First World (Meta for Business, 3 March 2026); What's new in Google Analytics (Google Analytics Help, 2026); Update on Plans for Privacy Sandbox Technologies (Google Privacy Sandbox, 17 October 2025); Strengthen media measurement and ROI clarity with incrementality testing improvements (Google Ads Help, 11 November 2025).

Related answers

• What multi-touch attribution can and cannot see now
• Meta Ads ROAS dropped: what changed and what to check
• Meta Ads vs Shopify revenue: why the totals differ
• Ad platform attribution windows, read from their own docs