What Is the Attribution Setting in Meta Ads?

When someone buys after seeing or tapping your ad, Meta needs a rule for whether that sale counts as yours. The attribution setting is that rule. You set it per ad set, and because Meta optimizes delivery around the conversions it can credit, the choice affects who your ads are shown to as well as the numbers you read in reporting.

Which attribution models can you choose?

Meta lets you pick an attribution model at the ad set level:

• Standard optimizes delivery for the time windows and behaviors you select, and lets you credit conversions based on ad impressions, clicks and video plays.
• Incremental optimizes for incremental conversions, using models that predict whether a conversion was caused by an ad.
• Custom lets you share attribution data from your external analytics tool so delivery optimizes toward your own attribution logic.

Only the standard model uses attribution settings. Meta warns that results can’t be compared in the campaign overview table across ad sets with different models, because each model counts conversions differently.

What windows does standard attribution offer?

Setting	What it counts	Window
Click-through	Events after a link click on your ad	1 day or 7 days
View-through	Events after an impression of your ad	1 day
Engage-through	Events after a click that isn’t a link click; for video, a 5-second play, or 97% of a video shorter than 5 seconds	1 day

These windows apply to website and in-store conversions. Meta says some accounts may still use earlier versions of click-through and engage-through while the change rolls out. Some Meta pages still call engage-through by an older name, engaged-view. iOS 14 and later app campaigns follow different rules.

How does the setting change the Results column?

Meta gives a worked example. An ad set uses 1-day click-through, 1-day view-through and 1-day engage-through, and records 5 click-through, 15 view-through and 3 engagement conversions. The Results column shows 20, which is 5 + 15: with all three selected, Results counts click-through and view-through conversions. If you select only click-through and engage-through, Results counts click-through plus engagement conversions.

How do you compare windows in Ads Manager?

1. Go to Ads Manager and click the Columns: Performance dropdown menu.
2. Select Performance, then scroll down and select Compare attribution settings.
3. Select the settings to compare, such as 1-day view, 1-day click, 7-day click and 28-day click.
4. Choose All conversions or First conversion as the conversion count, then click Apply.

You can compare these windows regardless of the setting the ad set optimizes for. You need administrator, advertiser or analyst access to the ad account. If a campaign has run for less than 28 days, the 28-day click view only covers the days it has run.

What if ad sets in one campaign use different settings?

Meta calls this multiple attribution settings. Results for the conversion event are then added together across the different settings, such as 1-day view, 1-day click and 7-day click. To see each ad set’s choice, add the Attribution setting column through Customize columns.

FAQ

Can I see 28-day click results?

Yes, as a reporting comparison. Meta’s Compare attribution settings feature includes 28-day click alongside 1-day view, 1-day click and 7-day click. If the campaign hasn’t been live for at least 28 days, that view only shows partial data covering the campaign’s length.

Can I compare ad sets that use different attribution models?

Meta warns against doing it in the campaign overview table. Each model counts conversions differently, so comparing across models leads to inaccurate conclusions. Meta recommends the Compare attribution settings feature instead, or evaluating performance in your external analytics tool if you use one.

Does the attribution setting change who sees my ads?

Yes. Meta says attribution models inform ad delivery, and that standard attribution optimizes delivery for the time windows and user behaviors you select. The setting is not only a reporting filter; it also shapes which conversions the system works to get.