Marketing9 min read
Meta attribution change 2026: what moved in your data and what to do
admin
It is Monday review. Ads Manager shows fewer purchases than last month. Shopify shows orders roughly flat. Nobody touched the budgets, the creatives or the audiences.
Before you pause a campaign, check the calendar. The Meta attribution change 2026 brought two separate shifts in how Meta counts a conversion, and both can move your reported numbers without any change in what customers actually bought.
This guide explains what changed, what it does to your ROAS, and how to re-baseline your reporting this week.
The Meta attribution change in 2026 is two updates, not one. In January, Meta removed the 7-day and 28-day view windows from its Ads Insights API. From late March, Meta narrowed click-through attribution to link clicks only and moved likes, shares and saves into a new bucket called engage-through.
| Date | What changed | Where you see it |
|---|---|---|
| 12 Jan 2026 | 7-day view and 28-day view windows stop returning data in the Ads Insights API | Reporting tools, connectors and dashboards that pull from the API |
| 12 Jan 2026 | Shorter history for some breakdowns (unique counts, hourly, frequency) | Long lookback reports |
| 3 Mar 2026 (announced) | Click-through counts only conversions after a link click | Ads Manager reporting |
| Late Mar 2026 (rollout) | Engaged-view becomes engage-through; non-link interactions move there | Ads Manager reporting and delivery |
Sources: Meta for Developers, Meta newsroom.
If your conversions dipped in January, look at the first change. If they dipped from April onwards, look at the second.
From 12 January 2026, Meta's Ads Insights API no longer returns the 7-day view (7d_view) and 28-day view (28d_view) attribution windows. Meta announced this on its developer blog in October 2025 and applied it across API versions.
Here is what still works through the API, per Meta's announcement:
Meta described the change as "consistent with Ads Manager behavior". In plain terms, the API now matches what you could already choose in Ads Manager.
The same update cut how far back some breakdowns go:
unique_actions) and hourly breakdowns: 13 months of historyWho feels this most? Teams that built dashboards or agency reports on long view windows. If your Looker Studio sheet or reporting connector asked for 7-day view, those conversions simply stopped appearing. Nothing changed in the account. The column went empty.
In March 2026, Meta changed what counts as a click for attribution. On 3 March 2026, Meta announced that click-through attribution for website and in-store conversions would include only link clicks. Before this, Meta counted all click types, including likes, shares and saves (Meta newsroom).
The rollout started at the end of March 2026 for campaigns optimising for website or in-store conversions. Meta said timing would vary by advertiser and that billing would not change.
Many advertisers did not realise the old rule existed. In October 2025, Meta's own help text said clicks "may include interactions such as likes, shares and saves", which surprised a lot of media buyers when it surfaced (PPC Land).
Engage-through attribution is the new home for conversions that follow a non-link interaction with your ad. A shopper likes, shares, saves or comments on your ad, does not click through, and buys later. Meta renamed its older engaged-view attribution to engage-through and moved these conversions into it.
Two details matter for D2C accounts:
Meta made the change to bring its click numbers closer to third-party tools such as Google Analytics, which mostly count link clicks. Meta also argues that social ads drive purchases without a click: it says 46% of online purchase conversions from Reels ads happen within the first 2 seconds of attention (Meta newsroom).
Meta also said it is working with Northbeam and Triple Whale to include both clicks and views in their attribution models. And it repeated that incrementality experiments such as Conversion Lift remain the gold standard for measurement.
The change moves conversions between buckets, and some drop out of the default view entirely. Your customers bought the same products. Meta now files some of those purchases under a different label, or outside the window.
Take one illustrative shopper, not real data:
| What the shopper did | Before March 2026 | After March 2026 |
|---|---|---|
| Clicked the link, bought on day 3 | 7-day click | 7-day click |
| Liked the ad, bought the same day | 7-day click | 1-day engage-through |
| Liked the ad, bought on day 5 | 7-day click | Not counted (outside the 1-day engage-through window) |
| Watched 6 seconds of a video, bought the next morning | 1-day view at most (under the 10-second engaged-view bar) | 1-day engage-through |
This follows from the old click rule and the new 1-day engage-through window. It explains why three patterns show up in accounts after the change:
For a COD-heavy Indian brand there is a second layer. Meta's Purchase event fires at checkout, so a COD order that later goes back to origin stays counted as a purchase (Clevrr AI). Attribution changes shift Meta's count. RTO shifts what you actually collect. Keep the two problems apart when you diagnose a drop.
The default Meta attribution setting in 2026 is 7-day click, 1-day engage-through and 1-day view, on the standard attribution model (Jon Loomer). You can set engage-through or view to "none" at the ad set level.
A few points that catch teams out:
To re-baseline ROAS after the Meta attribution change, treat March 2026 as a break in your data, split click-through and engage-through, and anchor the result to Shopify orders. Do this before you change budgets or bids.
7d_view or 28d_view, change it to a window that still exists. Otherwise it reports zeros and nobody notices.Here is an illustrative example. Say a brand spends ₹8L a month on Meta and its old dashboard showed ₹28L in attributed revenue, a 3.5 ROAS. After March, click-through shows ₹23L and engage-through shows ₹3L. The brand has not lost ₹5L of sales. It has lost ₹2L of reported credit, and Shopify will tell you whether real orders moved at all.
Catch attribution shifts early by putting Meta's numbers next to your store's numbers every day, not once a month. When Meta purchases fall and Shopify orders hold, that gap points to measurement, not demand.
This is the kind of gap Clevrr AI watches for. Clevrr AI is an AI performance marketer for D2C brands. Its agent, Klev, reads Meta Ads alongside Shopify orders, stock and deliveries every morning and explains why revenue or ROAS moved, with the source of each number. A reporting change shows up as a mismatch to check, not a reason to pause your best campaign.
The Meta attribution change 2026 did not make your ads worse. It changed the labels and the windows Meta uses to count them: long view windows left the API in January, and from March only link clicks count as clicks.
Re-baseline from March, split click-through from engage-through, fix any connector still asking for 7-day view, and check every Meta number against Shopify delivered orders. Then judge performance.
Want a morning check that does this for you? Start free with 500 credits.
There is no sign the 7-day and 28-day view windows are coming back. Meta removed them from the Ads Insights API on 12 January 2026, and Meta's announcement does not mention a return. The 1-day view window remains, along with 1-day, 7-day and 28-day click. Rebuild any long-window reports on the windows that still exist.
No, the Meta attribution change does not affect what you pay. Meta said billing would not change when it announced the link-click update in March 2026. What changes is how Meta labels and counts conversions in reporting. That can still affect cost per result and ROAS targets, because those targets use the reported numbers.
Usually no, you should not change your attribution setting on a live campaign just to recover old numbers. The setting drives delivery as well as reporting, so changing it resets learning. Instead, split click-through and engage-through in your reports, re-set ROAS goals on the new baseline and test any new setting in a fresh ad set.
Join hundreds of D2C brands using Clevrr AI to automate their growth and efficiency.
Join 500+ D2C founders using Clevrr AI to automate their operations and scale faster.
Get Started Free