Since March 2026, many advertisers have been seeing lower click-through conversion numbers in Meta Ads Manager without any change in actual ad spend or campaign setup. If that is happening in your account, it is not a performance problem. Meta changed how it defines and counts conversions, and the shift is affecting reported numbers across the board.
This post explains what changed, what engage-through attribution is, how the new model affects your reporting, and what to check in your account now. Note that this is a different change from the attribution window removal of January 2026: that update removed longer view-through windows from the API. This one changes the definition of what counts as a click.
What Meta changed and why
Before March 2026, Meta's click-through attribution counted any click on an ad as a qualifying interaction. That included link clicks, but also likes, shares, saves, comments, profile taps, and image expansions. If someone liked your ad on a Tuesday and bought your product on Friday, Meta reported that as a click-through conversion.
Third-party tools like Google Analytics have always counted only link clicks, because those are the only interactions that actually send someone to your website. The result was a persistent gap between what Ads Manager showed and what GA4 or any other analytics tool reported, and that gap was often hard to explain to clients or stakeholders.Meta announced on March 3, 2026 that it was fixing this. Click-through attribution now requires an actual link click: a click that sends the user to a website, app, lead form, or other destination. Likes, shares, saves, and other non-link interactions no longer qualify as click-through conversions.
The conversions that used to be counted under click-through because of a like or a save did not disappear entirely. They moved to a renamed category called engage-through attribution.
What engage-through attribution is
Engage-through attribution is the new name for what Meta previously called engaged-view attribution, with an expanded scope. It now captures two types of interactions:
- Social engagements that are not link clicks: likes, shares, saves, comments, profile visits, and other non-link ad interactions followed by a conversion within one day.
- Video engaged views: someone who watched at least 5 seconds of a video ad (reduced from the previous 10-second threshold) and then converted within one day.
The conversion window for engage-through is fixed at one day. There is no 7-day option as there is for click-through. Engage-through attribution is on by default for campaigns optimizing for website or in-store conversions, and advertisers can turn it off by selecting "none" in attribution settings.
The key thing to understand about engage-through is what it represents: a user who shared or saved your ad showed a meaningful signal of interest. They noticed it, they responded to it. Meta is formalizing that as a step in the conversion journey. Whether that step actually caused the conversion is a separate question, and one worth thinking about depending on your campaign type.
How the Meta attribution change affects your reported numbers
The immediate effect most advertisers notice is a drop in click-through attributed conversions. This is not a real performance decline. It is a reclassification: conversions that previously counted as click-through because someone liked or saved an ad are now sitting in the engage-through column instead.
Some conversions will disappear from reported totals entirely. If a non-link interaction happened and the user converted on day two through seven, that conversion now falls outside engage-through's one-day window. It does not move to another bucket. It simply does not get attributed.
Billing is not affected. Meta confirmed that the changes apply only to how conversions are classified in reporting. You will not be charged differently.
The default attribution setting for campaigns optimizing toward website conversions is now 7-day click-through, 1-day engage-through, 1-day view-through. If you previously had engaged-view attribution turned off in your account, review that setting. Engage-through now covers a broader category than the old engaged-view did, so the decision to keep it on or off may be different depending on your campaign type.
When to keep engage-through on
For purchase campaigns, keeping the default 1-day engage-through on is generally reasonable. A user who saves or shares an ad for a product they later buy is showing genuine interest, and that signal has value. The attribution may not be perfectly incremental, but it reflects real engagement.
When to consider turning it off
For lead generation campaigns, the logic is less clear. If someone liked your ad but did not click through to the lead form, their subsequent conversion is harder to attribute to your ad with confidence. The same applies when running campaigns with tight remarketing audiences, where engage-through can inflate numbers in ways that obscure real performance.
Why your Meta Ads numbers may not match GA4 even after this change
Even with this update, some gap between Meta Ads Manager and GA4 will persist. The new definition of click-through attribution is much closer to how GA4 counts sessions from paid social, but the two systems still measure different things. GA4 tracks sessions that start with a click to your website. Meta tracks conversions within its attribution window regardless of what happens in between. A user might click a Meta ad, land on your site, leave, return via a direct visit three days later, and convert. GA4 might attribute that to direct. Meta will attribute it to the ad click.
The gap will be smaller after this change, but it will not be zero. The right approach is to use both data sources together rather than expecting them to match. If you want a deeper look at how Meta Ads and GA4 measure things differently, we cover that in detail in our Meta Ads and GA4 integration guide.
What to check in your account now
- Do not compare pre- and post-March 2026 data directly. You are comparing two different measurement systems. A drop in click-through conversions between February and March does not mean performance declined. Set a new baseline from mid-March 2026 onwards and evaluate from there.
- Review your attribution settings. Go to Ads Manager, open a campaign, and check the attribution setting at ad set level. Confirm whether engage-through is on or off and whether the current setting makes sense for your campaign objective.
- Use Compare Attribution Settings in Ads Manager. This feature lets you see how your conversions distribute across different attribution windows without changing your live settings. Use it to understand how much of your reported volume comes from engage-through before deciding whether to keep it on.
- Be careful when changing attribution settings on live campaigns. Switching from 7-day click to 1-day click mid-campaign can trigger a learning phase reset if it drops your weekly conversions significantly. If you want to test different settings, do it at the start of a new campaign or a new flight period.
- Update your reporting to separate click-through and engage-through columns. Presenting total conversions without this breakdown makes it impossible to evaluate the quality of each attribution type and will cause confusion when stakeholders compare the numbers to GA4.
Quick reference: the 3 Meta attribution changes in one place
Bottom line
If your Meta Ads numbers look different since March 2026, the attribution change is almost certainly the reason. Click-through conversions will be lower because the definition is now stricter and closer to what third-party tools have always measured. That is a good thing for reporting accuracy, even if the short-term adjustment is uncomfortable.
The most important thing to do right now is not to change settings reactively. Understand where your conversions are coming from using the Compare Attribution feature, set a new baseline, and then make deliberate decisions about whether engage-through attribution makes sense for each of your campaign objectives.
For further reading, Meta's original announcement was covered in detail by Search Engine Land, and Brainlabs has a useful breakdown of the creative strategy implications of the change.
