Attribution is how Meta gives credit to an ad for a conversion. While that’s generally what it is, there are several layers of attribution to understand if you’re going to truly know how it works.
When you create an ad set that utilizes the Website conversion location and performance goal to maximize the number of conversions, the default attribution setting is 7-day click-through, 1-day engage-through (changed from engaged-view), and 1-day view-through.
But that’s not it. The default attribution model is Standard…
…and the default conversion count is All Conversions.
That’s a lot, and these layers make attribution far more complicated now than it used to be. Once you thoroughly understand these layers, you can intelligently evaluate results and make better optimization decisions.
In this post, I’ll help you understand each of these elements of attribution as they pertain to Meta advertising and how to approach them in 2026.
1. Click-Through Attribution
Definition
Click-through is when someone clicks your ad link and converts within the period of time that you select. Meta learns from these results and will show your ads to people likely to click through within the same period of time.
Background
Meta changed the definition of click-through attribution in March of 2026. Prior to this change, Meta considered any click on your ad prior to converting under click-through attribution. The old definition included social clicks and other clicks on ad elements other than links.
What Are Link Clicks?
It would be easy to assume that a “link click” and an “outbound click” are the same thing. While an outbound click is a type of link click, a link click isn’t always an outbound click.
Meta’s definition of link clicks includes clicks on links to websites. But it also includes several other actions like click to call, click to message, lead forms, and more.
While it may seem like a long list of potential loopholes, these “other” link clicks should be exceedingly rare when using the Website conversion location and the performance goals to maximize the number or value of conversions.
Conversion Windows
The default click-through conversion window is 7 days. In other words, Meta will count a conversion if it happens within seven days of clicking your ad.
Advertisers have the option of 7 or 1-day click-through attribution in the ad set. This will impact both how results are reported by default as well as how ads are delivered.
Years ago, the default click-through attribution window was 28 days. While it’s no longer available as an optimization element, advertisers can use the Compare Attribution Settings feature to view conversions that happened within the 28-day click-through window.
How to Approach
If your goal conversion event is a purchase, it’s generally recommended to stick with the default 7-day click-through attribution window.
If offering something for free that doesn’t require deliberation, 1-day click-through is recommended to better reflect the typical customer journey. Though the impact on volume when switching from 7-day click to 1-day click should be considered.
2. Engage-Through Attribution
Definition
Meta counts an engage-through conversion when someone converts after clicking anything other than a link on your ad (likes, reactions, comments, shares, saves, and all “other” non-link clicks). It also includes engaged-views.
Background
Engage-through attribution replaced engage-view attribution when Meta made changes to click-through attribution in March of 2026. Prior to this change, the previously named engaged-view attribution only included the situation where someone watched a video for at least five seconds and then converted within a day.
Meta updated click-through attribution to only include conversions that happened after a click on an ad link. Conversions that resulted from social clicks and all other clicks not included under click-through attribution moved to the new engage-through attribution. This new attribution still includes engaged-views.
Conversion Windows
Engage-through attribution is on by default with a 1-day conversion window. Advertisers have the option of turning engage-through attribution off by selecting “none.”
The screenshot above was taken after engage-through attribution was announced, but prior to the completed rollout.
How to Approach
When the goal conversion event is a purchase, it’s recommended to keep the default 1-day engage-through attribution on. This action reflects interest and awareness. Even if the eventual purchase was driven by another channel, the initial click shows that the ad had impact.
On the other hand, it may not make sense to keep 1-day engage-through on when optimizing for any other type of event, like a lead. If the person didn’t click through to get the free thing, your ad’s impact is debatable. You may also want to remove engage-through when restricting by a remarketing audience.
3. View-Through Attribution
Definition
Meta counts a view-through conversion when someone converts within a day of being served an impression. This assumes they never clicked the ad, otherwise it would fall under either click-through or engage-through attribution.
Background
View-through is the most controversial of the attribution settings, and it leads to the most confusion. It’s also the most likely source of inflated results.
While click-through attribution reflects intent and engage-through suggests interest, the value of view-through results is cloudy. While view-through conversions can suggest that your ad impacted a customer decision, we also can’t prove that the impression was even seen.
The most common scenario where this comes up is via remarketing. Someone who is on a customer email list or who engages with a brand on other channels is served an ad that they may not have seen. Within a day, they act on another marketing effort and convert.
Did the ad have an impact? In some cases, even if the impact is less than if a click were made. But in others, that impression may have had no impact at all.
Conversion Windows
View-through attribution is on by default with a 1-day conversion window. This can be turned off by selecting “none.”
How to Approach
The most common example of a view-through conversion that has value is when someone is shown your ad for a product, the person doesn’t click, but it inspires them to later Google your business or product to complete the purchase. While your ad doesn’t deserve full credit, it positively impacted results.
But such an example is limited to purchases and would be rare for any other event. Might someone be interested in your free lead magnet, not click, and then Google it later to complete the form? Sure. But, it’s not normal behavior.
Consider removing 1-day view-through attribution for any non-purchase event. And since inflated results are most often caused by this type of conversion when remarketing, remove view-through attribution in that case.
4. Attribution Model
Attribution Model is set to “Standard” by default, but you do have another option…
Definitions
Standard: Optimize ad delivery for a selected time window and engagement type based on your attribution settings. For example, 7-day click, 1-day engage-through, and 1-day view-through.
Incremental: Optimize delivery for conversions using models that predict whether a conversion is caused by an ad.
Background
Incremental attribution rolled out in 2025. When selected, you lose the ability to edit attribution settings. That makes sense when you understand how each works.
Ultimately, standard attribution doesn’t consider whether someone saw your ad or was influenced by it at all. Sure, that’s difficult to avoid if you get a conversion following a click-through, but it’s certainly plausible that a view-through conversion wasn’t incremental.
The point of incremental attribution isn’t to eliminate view-through conversions. You’ll actually see view-through results when using incremental attribution. The hope is that your results will be limited to those that were actually influenced by your ads.
How to Approach
In theory, incremental attribution should lead to a more accurate reflection of the impact of your ads. And, in theory, it should lead to higher quality results, even if that’s at the expense of volume. And these are tradeoffs most advertisers are willing to make.
Personally, I’ve seen little difference between my standard and incremental results. This may be because I approach attribution settings differently depending on the conversion goal. I prevent Meta from leaning into view-through conversions when they’re otherwise likely to be a problem, for example.
Does it make sense to use incremental attribution and should you prefer it in most cases? Yes. But does it make the difference you’d expect it to make? I just haven’t seen it. And because of that, I’m not all that motivated one way or the other.
I’d consider incremental attribution the best option by default if you deal in high budgets and already have no problem getting high volumes of results. Otherwise, I wouldn’t recommend intentionally limiting results if you already struggle to get out of the learning phase (exiting the learning phase is a general rule of thumb, not a requirement).
5. Conversion Count
Conversion Count is set to “All Conversions” by default, but you do have another option…
Definitions
All Conversions: Meta will count every conversion that happens after someone clicks or views your ad and takes an action within your attribution window.
First Conversion: Meta will only count the initial conversion that happens after someone clicks or views your ad and takes an action within your attribution window.
Background
Meta introduced first conversion in 2024, but only as a reporting option at the time. Meta eventually launched the ability to customize how your ad set is optimized by conversion count in late 2025.
The ability to separate “first” from “all conversions” is valuable because it highlights a common point of confusion for advertisers. Your default results can include multiple conversions by the same person.
For example, let’s assume your performance goal is to maximize the number of conversions where the conversion event is a purchase and you’re using the 7-day click-through attribution setting. Someone clicks on your ad and immediately goes to your website and makes a purchase. They then decide to make another purchase five days later (the product doesn’t matter). Both purchases would be reported when using “all conversions.”
But when using first conversion, Meta would only report the first purchase. This would impact both the raw number of results and all value-related data — the value would reflect the value of the first product, not the second.
Refining your results to only the first conversion can also help clear up reporting conflicts. We may assume that Ads Manager is inflated because it doesn’t line up with back-end or third-party reporting. But Ads Manager results can include multiple conversions by the same person.
How to Approach
While I’m a big fan of first conversion for adding context to results, I’m not sure I understand the argument behind customizing conversion count to impact optimization in the ad set. It feels counterintuitive.
By changing the conversion count in the ad set, Meta will only report on the first conversion and make decisions based on that. Meta will not value a customer who makes two purchases more than a customer who makes one. When using first conversion, Meta sees both customers as equals.
But don’t we prefer the scenario where a customer makes multiple conversions? Don’t we want Meta to learn from that? And don’t we want to see that in our results?
The use cases I can think of aren’t particularly strong. Maybe if you’re optimizing for leads and you don’t want to reward a situation where someone completes your form twice in a few days. Or maybe you’re optimizing for an engagement-related website event and you don’t want Meta to lean into inflated results of people performing actions over and over.
I guess it wouldn’t hurt in either case to limit conversion count to first in the ad set. But it’s a limited scope.
General Approach to Attribution
Hopefully this post provides a basic understanding of all of the attribution-related elements so that you can make informed decisions.
The main thing I’d say is that you should normally use the default settings unless there’s a specific problem to be solved. This applies to attribution just as it does to targeting, placements, creative enhancements, and just about anything else.
The default settings are going to be better for the majority of situations, which is why they’re the defaults in the first place. Know what you’re doing and why when making customizations.
Your Turn
Do you have any questions about attribution that aren’t answered here?
Let me know in the comments below!
The post How Meta Ads Attribution Works in 2026 appeared first on Jon Loomer Digital.
