Meta Ads Attribution Settings in 2026: How to Choose the Right Window After Engage-Through

Meta Ads attribution settings have never been more confusing. Or more important.

In March 2026, Meta overhauled how conversions get counted. Click-through attribution now only tracks link clicks. A new "engage-through" attribution type captures social interactions and video views. And an advanced "incremental attribution" option uses holdout testing to measure true lift.

The result? Advertisers are staring at dashboards that look completely different from last month. CPAs appear higher. ROAS appears lower. Conversions seem to have vanished. And nobody knows which attribution settings to actually use.

This guide breaks down exactly what each Meta Ads attribution setting does in 2026, how the new engage-through attribution works, whether incremental attribution is worth trusting, and which settings you should use based on your business type. No fluff. Just a practical decision framework backed by real data.

What Are Meta Ads Attribution Settings?
The Three Types of Meta Attribution in 2026
Attribution Windows Explained
Engage-Through Attribution: The March 2026 Change
Incremental Attribution: Meta's Holdout-Based Approach
Which Attribution Settings Should You Use?
How Attribution Settings Affect Ad Delivery
How to Change and Compare Attribution Settings
When Your Numbers Drop: The Recovery Playbook
Frequently Asked Questions
Key Takeaways

What Are Meta Ads Attribution Settings?

Attribution settings tell Meta how to assign credit for conversions back to your ads. When someone sees or interacts with your ad and later makes a purchase, attribution settings determine whether that purchase "counts" as a result of your ad.

Two things attribution settings control:

The attribution window: how many days after an interaction Meta will credit your ad with a conversion (1 day, 7 days, etc.)
The interaction type: what kind of interaction qualifies (clicking a link, watching a video, liking the ad, or just seeing it)

Here's what trips up most advertisers: attribution settings are primarily a reporting tool, not a delivery tool. Changing your attribution window from 7-day click to 1-day click changes how conversions get counted, not necessarily how ads get delivered. But the relationship is more nuanced than that, and we'll cover the delivery impact later in this guide.

The default attribution setting in Meta Ads Manager is now 7-day click, 1-day engage-through, 1-day view. If you've never touched your attribution settings, that's what you're running.

The Three Types of Meta Attribution in 2026

Before March 2026, Meta had two attribution types: click-through and view-through. Now there are three, plus an advanced option. Here's how they work.

Attribution Type	What Triggers It	Available Windows	Status
Click-through	Someone clicks a link in your ad (goes to website/app)	1-day, 7-day	Redefined March 2026
Engage-through	Someone interacts with your ad (like, share, save, comment) or watches 5+ seconds of video, then converts	1-day only	New March 2026
View-through	Someone sees your ad (no interaction at all) and converts	1-day only	Unchanged
Incremental (Advanced)	Meta uses holdout testing to measure only conversions that wouldn't have happened without the ad	N/A (modeled)	Launched April 2025

The critical change: click-through used to include all clicks. Likes, shares, saves, comments, and link clicks all counted as "clicks" and triggered the 7-day click attribution window. Now, only link clicks count. Everything else moved to engage-through with a 1-day window.

This is why your reported click-through conversions dropped. It's not a performance decline. It's a reclassification.

Attribution Windows Explained

An attribution window is the timeframe after an interaction during which Meta will credit your ad with a conversion. Here's every window currently available.

Window	What It Means	Best For
1-day click	Conversion within 24 hours of clicking a link	Impulse purchases, low-AOV products, flash sales
7-day click	Conversion within 7 days of clicking a link	Most e-commerce (consideration purchases), standard campaigns
1-day view	Conversion within 24 hours of seeing the ad (no interaction)	Additional signal for Meta's algorithm, high-impression campaigns
1-day engagement	Conversion within 24 hours of a non-link interaction (like, share, save) or 5+ second video view	Video-heavy campaigns, Reels, social engagement campaigns

Windows that no longer exist:

Meta removed the 7-day view and 28-day view windows on January 12, 2026. Some advertisers lost 30-40% of their reported conversions overnight because those conversions fell outside the shorter windows. The 28-day click window was removed back in April 2021 after iOS 14.5 but is still available as a comparison option in reporting.

How to read the defaults: when Meta says "7-day click, 1-day engage-through, 1-day view," it means a conversion gets attributed to your ad if any of these conditions are met:

The person clicked a link within the last 7 days, or
The person engaged with the ad (like, share, save, video view) within the last 1 day, or
The person saw the ad within the last 1 day

If multiple conditions apply, Meta uses a priority order: click-through gets credit first, then engage-through, then view-through.

Engage-Through Attribution: The March 2026 Change

This is the biggest attribution change since iOS 14.5. And most advertisers are underestimating its impact.

What Engage-Through Attribution Actually Is

On March 3, 2026, Meta announced a fundamental redefinition of how ad interactions get classified. The change splits what used to be one category ("clicks") into two:

Click-through (redefined): Only link clicks. Someone taps your ad and lands on your website or app. That's it. Nothing else counts as a click anymore.

Engage-through (new): Everything else. Specifically:

Ad clicks excluding links: likes, shares, saves, comments, profile taps, and any other click that isn't a link click
Video views: watching at least 5 seconds of a skippable video ad (reduced from the previous 10-second threshold), or 97% of the total video length if the video is shorter than 5 seconds

The engage-through window is 1 day only. No 7-day option.

Why This Change Matters More Than You Think

Here's the part that most guides miss. Under the old system, if someone saved your ad on Monday and purchased on Thursday, that counted as a click-through conversion within the 7-day window. It showed up in your click-through column in Ads Manager.

Under the new system, that same sequence gets zero attribution:

It doesn't count as click-through (saving isn't a link click)
It doesn't count as engage-through (the 1-day window expired on Tuesday)
It doesn't count as view-through (the 1-day window also expired)

That conversion disappears from your reporting entirely. The person still bought. Your ad still influenced them. But Meta can no longer claim credit for it.

This is exactly why your ROAS numbers look worse without actual performance declining. The measurement got stricter. The results didn't necessarily change.

Who Gets Hit Hardest

Remarketing campaigns. And it's not even close.

Jon Loomer, one of the most respected voices in Meta advertising, highlighted that "one of the reasons advertisers swear by remarketing strategies is due to the misleading results." Under the old system, remarketing audiences frequently engaged with ads (likes, saves) and converted days later. All of that got credited as click-through conversions within the 7-day window.

With engage-through attribution, those same conversions either fall into the 1-day engage-through bucket or disappear from attribution entirely. If your remarketing campaigns look dramatically worse after March 2026, they probably weren't as effective as your old reporting suggested.

This doesn't mean remarketing is dead. It means the data was inflating remarketing's apparent value. Now you're seeing a more honest picture.

Why Meta Made This Change

The official reason: alignment with third-party analytics. Previously, Meta's broad definition of "clicks" created persistent discrepancies between Ads Manager and Google Analytics 4. GA4 only tracks link clicks. Meta was counting likes and saves as clicks. That gap confused everyone.

The structural benefit for Meta: engage-through feeds more conversion signals to Meta's algorithm. As one agency put it, Meta gets to "have its cake and eat it too." Stricter click-through numbers that match GA4, while still capturing social interaction value in a separate bucket that helps the algorithm optimize.

And there's a Reels angle. Meta says 46% of Reels purchase conversions happen within the first 2 seconds of attention. The shorter 5-second video threshold (down from 10 seconds) captures more of these fast-scroll conversion paths. With 90% of Meta's ad inventory now in vertical format, this matters.

Incremental Attribution: Meta's Holdout-Based Approach

While engage-through is about reclassifying interactions, incremental attribution asks a fundamentally different question: would this conversion have happened even without the ad?

How It Works

Standard attribution counts every conversion that happens after someone interacts with your ad. The problem: some of those people would have bought anyway. Brand loyalists, people who were already browsing your site, bargain hunters who would have found you through Google. Standard attribution gives your ads credit for conversions they may not have caused.

Incremental attribution uses a holdout testing methodology:

Test group (~85%): Sees your ads normally
Control group (~15%): Never sees the ad
Meta compares conversion rates between the two groups
The difference = incremental lift (conversions that genuinely wouldn't have happened without the ad)

This is based on Meta's Conversion Lift study methodology, built directly into Ads Manager. You enable it under "Advanced options" in campaign setup.

The $1.05M Reality Check

Meta claims incremental attribution delivered "more than 20% improvement in incremental conversions" across 45 advertisers in testing. That sounds impressive. But independent testing tells a more nuanced story.

Seer Interactive tested incremental attribution across 6+ accounts representing $1.05 million in ad spend during April 2025. Their findings:

Meta reported 87% of conversions were incremental (only 13% would have occurred without ads)
When cross-referenced with GA4, only 67% were incremental (33% would have converted anyway)
That's a 20-percentage-point gap between Meta's self-assessment and third-party measurement
Best incremental performance came from refined mid-funnel audiences
Worst incremental performance: broad targeting and hyper-narrow retargeting

As Seer Interactive put it: "Don't blindly trust Meta's numbers when platforms report their own incrementality." Fair point. Any platform measuring its own effectiveness deserves scrutiny.

When to Use Incremental Attribution

Good fit:

Prospecting and new customer acquisition campaigns (highest incremental value)
Scaling budgets where you need to understand true lift
Always-on campaigns where organic and paid demand overlap
Teams with flexible budgets and a testing mindset

Not recommended:

Retargeting campaigns (by definition, these audiences have pre-existing intent, so incremental lift will be lower)
High-performing existing campaigns (don't disrupt what's working)
Campaigns with fewer than 50 conversions per week (not enough data)
If you rely on cost caps (not yet available with incremental attribution)

Important limitation: when you enable incremental attribution, you lose the ability to edit standard attribution settings. It's an either/or choice, not an add-on.

Which Attribution Settings Should You Use?

Here's the decision framework. Your ideal settings depend on what you sell, your sales cycle, and your campaign objectives.

E-Commerce: Low AOV (Under $50)

Recommended: 7-day click + 1-day engage-through + 1-day view

Impulse purchases happen fast. Most sub-$50 conversions occur within 1-2 days of the first interaction. The 7-day click window gives you a safety net for people who click and come back, while engage-through captures video-driven impulse buys.

When to narrow to 1-day click: flash sales, limited-time offers, or products with near-zero consideration time (think $15 accessories). This gives you cleaner data but fewer attributed conversions.

E-Commerce: Mid-Range ($50-$200)

Recommended: 7-day click + 1-day engage-through + 1-day view (default)

This is where the default settings work best. Mid-range products have a 3-5 day consideration window. Someone clicks, compares options, reads reviews, comes back. The 7-day click window captures that full path.

Keep engage-through ON, especially if you run video ads or Reels. It feeds Meta's algorithm more conversion signals, which helps optimization for scaling your e-commerce campaigns.

E-Commerce: High Ticket ($200+)

Recommended: 7-day click + 1-day view. Consider turning engage-through OFF.

High-ticket products require more consideration. A 7-day click window is essential. But engage-through attribution for high-ticket items is debatable: if someone liked your ad but didn't click through, and then bought a $500 product the same day, was the "like" really the driver? Maybe. Maybe not.

Jon Loomer recommends evaluating engage-through's contribution using the Compare Attribution Settings feature (more on that below). If engage-through conversions represent less than 5% of your total, you're not losing much by turning it off. And your remaining data becomes cleaner.

Lead Generation

Recommended: 7-day click + 1-day view. Turn engage-through OFF for most campaigns.

For lead gen, the reasoning is straightforward: if someone didn't click through to your landing page, they didn't fill out your form. Attributing a lead to a "like" or a 5-second video view is a stretch.

Exception: video-heavy lead gen campaigns (webinar signups promoted via Reels, video sales letters). In these cases, engage-through captures legitimate video-driven conversions.

B2B / Long Sales Cycles

Recommended: 7-day click + 1-day view + offline conversion integration.

B2B is where Meta's attribution windows hurt the most. Your prospect clicks an ad on Monday, downloads a whitepaper, talks to their team, gets budget approval, and converts three weeks later. The 7-day click window misses everything after day 7.

The fix isn't a settings change. It's infrastructure. Implement the Conversions API (CAPI) with CRM integration to send offline conversions back to Meta. Use UTM parameters and post-conversion surveys ("How did you hear about us?") to capture attribution that Meta's windows miss.

App Install Campaigns

Recommended: Default settings + Apple SKAdNetwork for iOS.

For app campaigns, engage-through only counts video views (not social interactions). iOS tracking remains limited by App Tracking Transparency. Use SKAN integration and Meta's Advanced Mobile Measurement (AMM), which was re-enabled in June 2025 with row-level attribution data.

How Attribution Settings Affect Ad Delivery

This is where the "attribution is just reporting" narrative falls apart. While attribution settings primarily change how conversions are counted, they also influence how Meta delivers your ads.

How it works: Meta's algorithm optimizes toward conversions within your attribution window. If you use 7-day click attribution, Meta looks for people likely to convert within 7 days of clicking. If you switch to 1-day click, Meta narrows its optimization to people likely to convert within 24 hours. That's a different audience.

The practical impact:

Shorter windows = fewer data points for optimization. Meta needs approximately 50 conversion events per week per ad set to exit the learning phase. If switching from 7-day to 1-day click cuts your attributed conversions by 40%, you might drop below that threshold. The algorithm gets noisier. Performance gets volatile.
Engage-through ON = more signal for the algorithm. Keeping 1-day engage-through enabled gives Meta additional conversion touchpoints. For video-heavy campaigns, this can meaningfully improve optimization by telling the algorithm "these types of viewers are converting."
View-through feeds the algorithm too. Even if you don't trust view-through conversions for your reporting, leaving 1-day view ON gives Meta more data to learn from. This matters most for accounts with lower conversion volume.

The bottom line: don't change attribution settings casually. Switching attribution mid-campaign can trigger a learning phase reset, temporarily tanking performance. If you need to change settings, do it on a new campaign or at a natural reset point (new creative batch, budget change).

How to Change and Compare Attribution Settings

Changing Attribution Settings

Open Ads Manager and navigate to your campaign
At the ad set level, scroll to "Optimization & Delivery"
Click "Show more options" (easy to miss)
Under "Attribution setting," select your preferred window
For incremental attribution, look under "Advanced option"

Critical note: changes only apply to new campaigns or ad sets. You can't retroactively change attribution on existing data.

Using Compare Attribution Settings (the most underrated feature in Ads Manager)

This is the tool most advertisers don't know exists. It lets you view your conversion data across every attribution window simultaneously, without changing your actual settings.

Go to Ads Manager and select your campaigns
Click "Columns" and then "Compare Attribution Settings"
You'll see conversions broken out by: 1-day click, 7-day click, 28-day click, 1-day view, and 1-day engagement

Why this matters: before changing any settings, use Compare Attribution to see how your conversions distribute across windows. If 90% of your conversions happen within 1-day click, you're not gaining much from the 7-day window. If engage-through conversions represent 30% of your total, turning it off would significantly reduce your reported numbers (and your algorithm's data).

This feature is purely for reporting. Selections here do not change ad delivery or optimization. Use it to diagnose, then decide.

When Your Numbers Drop: The Recovery Playbook

If your Meta Ads dashboard looks worse after March 2026, here's the step-by-step recovery process.

Step 1: Stop Comparing to Pre-March Baselines

This is the most important step and the one most advertisers skip. Comparing March 2026 metrics to February 2026 metrics is comparing two different measurement systems. You didn't gain weight. The scale changed.

Set March 15-31 as your new baseline period. Evaluate April performance against this baseline, not against pre-attribution-change numbers.

Step 2: Add Engage-Through Columns to Your Reports

Go to Ads Manager, customize your columns, and add "1-day engagement" conversion columns. This shows you the conversions that moved out of click-through into the new engage-through bucket. In many cases, your total conversions (click + engage + view) haven't changed much. They're just distributed differently.

Step 3: Implement CAPI If You Haven't

This is non-negotiable in 2026. The Meta Pixel alone captures roughly 40-60% of actual conversions. iOS privacy restrictions, browser limitations, ad blockers, and consent banners eat the rest.

The Conversions API (CAPI) sends conversion data directly from your server to Meta. No browser dependency. No cookie blocking.

CAPI improves attribution accuracy by 25-40%. That's not a nice-to-have. It's the difference between your algorithm optimizing on 60% of reality versus 90%.

Three implementation options:

Option	Cost	Technical Effort
CAPI Gateway (no-code)	$10-400+/month	Low
Google Tag Manager Server-Side	Platform cost	Medium
Custom Developer Integration	Dev time	High

Whichever you choose: run both Pixel and CAPI together, send event matching parameters (email, phone, IP, click ID), configure event deduplication, and target an Event Match Quality score of 8.0+ in Events Manager. If you're running e-commerce and want to reduce your CPA, CAPI implementation is one of the highest-leverage moves you can make.

Step 4: Cross-Reference With GA4 and Your CRM

Meta's numbers should now align more closely with GA4 for click-through conversions (since both now count only link clicks). If they still don't match, check your CAPI setup and event deduplication.

For the full picture, use a simple framework:

Meta Ads Manager: what Meta claims credit for
GA4: independent measurement of traffic and conversions from Meta
Your CRM/Shopify: ground truth (actual revenue)
Post-purchase survey: qualitative attribution ("How did you hear about us?")

No single source tells the whole story. Triangulate.

Step 5: Brief Your Clients Before They See the Drop

If you manage ads for clients, get ahead of this. Send a proactive message explaining the attribution changes before your client logs into Ads Manager and panics. Show them the Compare Attribution view so they can see where conversions moved (not disappeared). Frame it as "Meta is now reporting more honestly" rather than "performance dropped."

Frequently Asked Questions

What is the default attribution setting in Meta Ads Manager in 2026?

The default is 7-day click, 1-day engage-through, and 1-day view. This means Meta will attribute a conversion to your ad if someone clicked a link within 7 days, engaged with your ad within 1 day, or viewed your ad within 1 day before converting. You can change this at the ad set level under "Optimization & Delivery."

Does changing attribution settings affect how Meta delivers my ads?

Yes, indirectly. Meta's algorithm optimizes toward conversions within your attribution window. A shorter window means fewer conversion signals, which can limit optimization. Switching from 7-day to 1-day click may push your ad set back into the learning phase if attributed conversions drop below 50 per week. Always make attribution changes on new campaigns rather than mid-flight.

What is 1-day engagement (engage-through) attribution?

Engage-through is a new attribution type introduced in March 2026. It counts conversions that happen within 1 day after someone interacts with your ad through a non-link click (like, share, save, comment) or watches at least 5 seconds of your video ad. Previously, these interactions counted as "clicks" in the click-through window. Now they have their own separate 1-day bucket.

Should I turn engage-through attribution ON or OFF?

It depends on your campaign type. For e-commerce with video ads or Reels: keep it ON. It feeds more conversion data to Meta's algorithm and captures legitimate video-driven purchases. For lead gen where form fills require a landing page visit: consider turning it OFF, since a "like" didn't drive a form submission. Use the Compare Attribution Settings feature to check what percentage of your conversions come from engage-through before deciding.

Why did my Meta Ads conversions drop in March 2026?

Most likely because of attribution reclassification, not actual performance decline. Social interactions (likes, shares, saves) that previously counted as click-through conversions within a 7-day window now only count as engage-through conversions within a 1-day window. Conversions from social interactions that happened 2-7 days before the purchase disappeared from attribution entirely. Check your engage-through columns in Ads Manager to see where those conversions moved.

What is the difference between standard and incremental attribution?

Standard attribution counts every conversion within your attribution window, regardless of whether the person would have purchased anyway. Incremental attribution uses holdout testing (15% of your audience never sees the ad) to measure only conversions that genuinely wouldn't have happened without the ad. Independent testing by Seer Interactive on $1.05M in spend found a 20-percentage-point gap between Meta's self-reported incrementality (87%) and GA4's assessment (67%).

When should I use incremental attribution instead of standard?

Use incremental for prospecting and new customer acquisition campaigns where you want to understand true lift. It works best with 50+ weekly conversions and flexible budgets. Don't use it for retargeting (lower incremental value by nature), campaigns that rely on cost caps (not available with incremental), or proven campaigns you don't want to disrupt. Think of incremental as a testing tool, not a permanent replacement for standard attribution.

Does attribution affect my ad spend or billing?

No. Attribution settings affect only how conversions are reported. Your actual costs (CPM, CPC, total spend) are determined by the auction, not by attribution. Changing from 7-day click to 1-day click won't reduce your spend. It will only change how many conversions appear in your reports.

Why are my Shopify sales different from Meta's reported conversions?

This is one of the most common questions in e-commerce advertising. The gap exists because Meta and Shopify use completely different measurement methodologies. Meta uses last-touch attribution within its attribution windows. Shopify tracks the literal checkout session. Add iOS privacy restrictions, ad blockers, and cross-device behavior, and discrepancies of 20-50% between platforms are common. Implementing CAPI and using UTM-based tracking in Shopify can reduce the gap significantly. One agency reported a 53% reduction in the Shopify-Meta discrepancy after proper CAPI setup.

How does the Conversions API (CAPI) improve attribution accuracy?

CAPI sends conversion data directly from your server to Meta, bypassing browser-level tracking limitations. The Meta Pixel alone only captures 40-60% of conversions due to iOS restrictions, ad blockers, and cookie blocking. CAPI fills the gap, improving attribution accuracy by 25-40%. It also improves your Event Match Quality score, which helps Meta's algorithm optimize more effectively. Run both Pixel and CAPI together for the best results.

Can I still see 28-day click data in Meta Ads Manager?

Yes, but only through the Compare Attribution Settings feature. The 28-day click window was removed as a campaign setting in April 2021, and 28-day view was removed in January 2026. However, you can still compare how your conversions would look under 28-day click by using the comparison tool in your reporting columns. This is useful for understanding how many conversions fall outside the 7-day window.

What happens if I switch attribution settings on a running campaign?

Changing attribution settings mid-campaign can trigger a learning phase reset if the change significantly affects your conversion volume. For example, going from 7-day click to 1-day click might cut your attributed conversions by 30-40%, dropping you below the 50 weekly conversion threshold. Your CPA will likely spike temporarily. Best practice: make attribution changes on new campaigns or at natural break points (new creative launch, budget restructure), not mid-flight on a performing campaign.

Key Takeaways

Three attribution types now exist: click-through (link clicks only), engage-through (social interactions + 5s video views, 1-day window), and view-through (impressions, 1-day window). Plus incremental attribution as an advanced option.
Your click-through conversions dropped because of reclassification, not performance. Social interactions moved from click-through to engage-through. Check both columns before panicking.
Remarketing campaigns are hit hardest. The old system inflated remarketing attribution by counting social interactions as clicks within a 7-day window. The new system exposes a more honest picture.
Default settings (7-day click + 1-day engage-through + 1-day view) work for most e-commerce brands. Don't change settings without using Compare Attribution to understand the impact first.
Incremental attribution is promising but imperfect. Independent testing found a 20-point gap between Meta's self-reported incrementality and GA4's measurement. Use it as a directional signal, not a final answer.
CAPI is non-negotiable. Pixel-only tracking captures 40-60% of conversions. CAPI closes the gap by 25-40%. Without it, your algorithm optimizes on incomplete data.
Don't change attribution settings casually. Mid-campaign changes can trigger learning phase resets and temporarily tank performance.

The Honest Truth About Attribution in 2026

No attribution model is perfect. Meta is still a platform measuring its own effectiveness. GA4 has its own blind spots. Your CRM misses touchpoints it can't track. Post-purchase surveys are biased by what customers remember.

The March 2026 changes are a step toward more honest reporting. Click-through now means what most people always assumed it meant (link clicks). Engage-through gives video and social interactions their own measurement lane. Incremental attribution tries to answer the hardest question in advertising: "Did this ad actually cause this sale?"

The advertisers who will win in this new landscape aren't the ones chasing a single "correct" attribution number. They're the ones building a measurement stack that triangulates across sources, setting realistic baselines under the new rules, and investing in creative diversity that makes the algorithm's job easier.

If your Meta Ads numbers have you confused, we audit e-commerce accounts and separate real performance issues from measurement shifts. We help brands build the tracking infrastructure and campaign strategy that actually works in 2026, backed by a 3x ROAS guarantee in 90 days.

Book a free discovery call and let's figure out what your numbers are really telling you.