Performance Marketing

Meta Ads Attribution Settings in 2026: What Changed and What to Trust

Meta removed long view windows and redefined click-through in 2026. Learn what each setting counts, why numbers changed and how to judge results.

TL;DR
  • In January 2026 Meta removed the 7-day view and 28-day view attribution windows from the Ads Insights API. The longest view window is now 1-day view.
  • In March 2026 Meta narrowed click-through attribution to link clicks only, and moved likes, shares, saves and similar interactions into a separate engage-through category.
  • Reported conversions may look lower than before, but real sales did not change. Compare Meta results with store data and lift tests rather than with old reports.

What does an attribution setting do in Meta Ads?

An attribution setting decides which conversions Meta counts for an ad and how long after an interaction a purchase can still be credited. It affects two things:

  • Reporting: the conversions and ROAS you see in Ads Manager.
  • Optimization: which conversions Meta’s delivery system learns from when it decides who should see your ads.

The default for most conversion campaigns is 7-day click plus 1-day view. Other common options are 7-day click only and 1-day click only. The attribution window you choose does not change how many people buy from your store. It changes how many of those purchases Meta claims.

What changed in 2026?

Several changes landed close together. Dates below are based on Meta’s developer notices and industry reporting.

Date Change Practical effect
January 12, 2026 7-day view and 28-day view windows removed from the Ads Insights API Reports and tools that used longer view windows show fewer conversions
March 2026 Click-through attribution counts only link clicks Purchases after a like, comment, share or save are no longer counted as click-through
March 2026 New engage-through category for non-link interactions and engaged video views, with a 1-day default window Some conversions moved from “click” to a separate bucket
March 2026 Engaged-view threshold for video lowered from 10 seconds to 5 seconds and folded into engage-through Video-driven conversions reported differently

If your dashboards or third-party tools pulled 7-day view or 28-day view data, historical comparisons will show a drop around January 2026. If your team compares click-through conversions before and after March 2026, part of the change comes from the new definition, not from weaker ads.

Which setting should an eCommerce store use?

There is no perfect choice, but a few rules help.

  • 7-day click plus 1-day view remains a sensible default for optimization in most prospecting campaigns, because it gives the system enough conversion signals.
  • 7-day click only is more conservative and often closer to what GA4 or your store data credits. Some brands report on it even if they optimize on the default.
  • 1-day click suits impulse purchases and very short buying cycles, but it can starve campaigns of learning data.

Whatever you choose, keep it consistent across campaigns you compare, and document it. Changing attribution settings in the middle of a reporting period makes trends meaningless.

Why do Meta and GA4 still disagree?

Even with shorter windows, Meta and GA4 measure different things:

  • Meta counts view-through and engage-through conversions. GA4 does not credit views at all.
  • Meta can match conversions across devices using logged-in users. GA4 depends on cookies and consent.
  • GA4 uses its own attribution model, typically data-driven, which may credit the last channel before purchase, such as email or branded search.

So Meta will usually report more conversions than GA4 gives it credit for, and both are partial views. The GA4 attribution settings guide explains the GA4 side.

How should you judge Meta performance now?

Use several layers together:

  1. Platform data, with a consistent attribution setting, for day-to-day optimization.
  2. Store data: total orders, new customer orders and blended efficiency. Check whether Meta spend changes move total sales.
  3. Conversions API quality: strong event match quality helps Meta attribute conversions it can still count. See Meta Conversions API on Shopify.
  4. Lift tests: Meta conversion lift studies or geo tests show how many sales Meta actually adds. See incrementality testing.
  5. Post-purchase surveys: “How did you hear about us?” captures influence that click tracking misses.

Worked example: explaining a “drop” in reported ROAS

(Illustrative numbers.) A fashion brand’s reporting dashboard shows Meta ROAS falling from 3.4× in December 2025 to 2.6× in February 2026, with similar spend. Leadership asks whether Meta stopped working.

Analysis:

  • The dashboard used 7-day click plus 7-day view. After January 12, 2026, the 7-day view data no longer arrived, and the tool fell back to 1-day view.
  • Store orders and new customers in January and February are in line with seasonal expectations.
  • A geo holdout run in February shows Meta still drives a meaningful incremental lift in new customer orders.

Conclusion: the drop is mostly a reporting change. The team rebuilds the dashboard with 7-day click plus 1-day view for all periods where possible, adds blended MER and new customer counts from the store, and stops comparing against pre-2026 view-through numbers.

Checklist: Meta attribution in 2026

  • [ ] Attribution setting documented for every campaign
  • [ ] Dashboards and third-party tools checked for removed view windows
  • [ ] Historical comparisons adjusted for the January and March 2026 changes
  • [ ] Engage-through conversions reviewed separately from link-click conversions
  • [ ] Conversions API set up with good event match quality
  • [ ] Meta results compared with store orders and new customers
  • [ ] Blended efficiency (MER) tracked alongside platform ROAS
  • [ ] Lift test or geo test run at least twice a year
  • [ ] Post-purchase survey includes Instagram and Facebook options
  • [ ] Attribution setting not changed in the middle of a test or reporting period

FAQ

Did Meta remove view-through attribution completely?

No. The 1-day view window remains available. The longer 7-day view and 28-day view windows were removed from the Ads Insights API in January 2026.

What is engage-through attribution?

It is a category Meta introduced in 2026 for conversions that happen after someone interacts with an ad without clicking the link, such as liking, sharing, saving, commenting or watching an engaged video view. It uses a short default window and is reported separately from link-click conversions.

Which attribution setting is most accurate?

None is fully accurate, because each counts conversions differently. The most reliable view of Meta’s impact comes from comparing platform data with store results and running lift tests.

Turn this into a plan for your store

The Free Profit Audit scans your store, sizes your biggest profit leaks and builds a 90-day plan.

Free Profit Audit