Facebook ads attribution settings in 2026: what the windows mean and which to use for leads

The three attribution windows

Meta’s help center describes them this way (source):

Meta uses a default attribution setting of 7-day click, 1-day view and 1-day engagement conversion.

The third window goes by several names: “engagement”, “engage-through”, earlier “engaged-view” — it is the same thing. Most guides still say the default is “7-day click, 1-day view”; that was true before it was added. Meta notes some accounts may still be on the earlier version while the change rolls out, so check the Attribution setting column in your own Ads Manager.

It changes delivery, not only the report

The setting lives on the ad set, and Meta’s API describes it as the Conversion attribution spec used for attributing conversions for optimization. (source) In plain words: if you count only 1-day clicks, Meta learns from — and looks for — people who convert within a day of clicking. With the default, it also learns from people who converted after only seeing or engaging with the ad.

So switching the window is a real change to the campaign, not a reporting filter. To look at the same results through other windows without touching delivery, use Compare attribution settings in Ads Manager: Meta lets you compare these attribution settings regardless of the ones you chose to optimize for in your campaign. (source)

What changed, with dates

• January 2024 — Meta added an engaged-view window for video (source).
• June 2025 — the API started returning results the way Ads Manager shows them by default (source).
• 12 January 2026 — 7-day view (7d_view) and 28-day view (28d_view) attribution windows will no longer be returned in the Ads Insights API (source). The 28-day click window stays available for comparison only.
• 3 March 2026 — Meta started changing the definition of click-through attribution for website and in-store conversions to exclusively include link clicks. Likes, shares and saves now count as engage-through, and the video threshold dropped from 10 to 5 seconds. There will be no change to how advertisers are billed. (source)

If your reported website leads dropped in March 2026 while your CRM didn’t, this is the likely reason (more on reading that drop).

Lead forms vs leads on your website

Meta applies the window to actions taken off your ad, such as a purchase on your website or shop (source). A lead form is filled in inside the ad itself, so it isn’t an action “off your ad” — and Meta’s reporting API keeps a separate window for actions on the ad. Meta doesn’t spell out the consequence, but it follows: changing the window moves website leads much more than lead form leads.

Watch out for the Leads column: it counts leads submitted on Meta technologies (including forms, Instagram, and Messenger) as well as off-Meta leads tracked by your pixel or Conversions API (source). On a campaign that mixes both, a change of window shows only in part of the number. Use Website leads to see the site side alone.

Which setting to use for leads

Meta doesn’t publish a recommendation for lead campaigns. A practical order:

1. Keep the default while you learn the account. It gives Meta the most signals, which helps small budgets.
2. Compare windows on last month with Compare attribution settings: how many website leads are 1-day click, how many come only from views or engagement.
3. Check those groups against the CRM. If leads credited only to views rarely become customers, test a click-only setting on one ad set — a significant change, so give it a full week. If they buy as often as the rest, keep the default.
4. Keep one setting per campaign. Results don’t add up across ad sets with different settings, and summary rows go blank.

Some experts, such as Jon Loomer, lean towards click-only for leads (source); Meta says it would strongly encourage advertisers to leverage engaged-view attribution (source). Your CRM is the tie-breaker.

Incremental attribution

Besides the standard model, Meta offers an incremental one. Meta describes it as counting conversions we believe would not have occurred without an ad being shown and reported an average 46% lift in incremental conversions compared to their business-as-usual approach in its tests (Meta, Q1 2025 earnings call). It is a prediction model, not an experiment in your account, and practitioners report it mainly for website sales campaigns. On most lead campaigns you’ll see only the standard model — check the Attribution model field in the ad set.

Which day a lead is counted on

Meta can report a conversion by time of conversion or time of exposure when the impression was served (source). The same help page asks you to Check that your reporting tool conversion attribution windows match in both tools. That is why Ads Manager, Google Analytics and your CRM rarely agree on “leads yesterday”: different windows, different days, and the CRM counts every lead, not only those credited to ads. Compare whole weeks, not days (how to read the report).

Compare windows against sales: Variafy

Variafy connects ChatGPT, Claude and other assistants to your Meta ad accounts and reads results with the attribution window you ask for. Try: “Last month by ad set: leads with the default setting and with 1-day click only — and how many of each became customers in the CRM.” With Pipedrive or a Google Sheet connected, you see whether the leads that only views brought are worth paying for.

How we checked

Questions

Related guides

• How to read a Facebook ads report: what each column means and what to check first
• Which Facebook ads bring paying customers: matching CRM sales to ads
• Facebook ads not converting: clicks but no leads, or leads but no sales
• Why Google Ads and Meta both claim the same sale — and how to reconcile them