Your Meta Ads are running. Conversions are coming in. But do you actually know what those numbers mean — or whether they reflect what’s happening in your real business?
If you’ve never reviewed your attribution settings in Meta Ads Manager, you may be optimising campaigns towards numbers that don’t reflect your actual performance.
This guide covers every major attribution setting available in 2025, how each one works, and a clear framework for deciding which settings are right for your campaigns.
What Is Attribution in Meta Ads?
Attribution is the process of determining which conversions should be credited to your ads — and on what basis.
When someone sees your ad on Monday and buys from your website on Friday, did your ad cause that conversion? Meta has to make a decision about this, and that decision is governed by your attribution settings.
Get your attribution settings right, and Meta’s algorithm will optimise towards the outcomes that actually matter to your business. Get them wrong, and you’ll be chasing numbers that look impressive in Ads Manager but don’t match your revenue or lead quality in the real world.
The Five Core Meta Ads Attribution Settings (2025)
- Click-Through Attribution Window
The click-through window is the number of days after someone clicks your ad during which a conversion can still be credited to that ad.
The default is 7 days. You can shorten it to 1 day.
A 7-day window is appropriate for most businesses where purchase decisions aren’t made immediately — people research, compare options, and return to buy later. For impulse purchases or lower-ticket items, a 1-day window may give you a cleaner picture.
- View-Through Attribution Window
View-through attribution credits conversions to users who saw your ad but did not click it, within a set time window. The default is 1 day.
This is the most debated setting in Meta Ads. If someone sees your ad and then buys from you the next day without clicking, is that because of your ad? Sometimes yes. Often, especially for always-on campaigns with broad reach, it is simply a pre-existing customer who happened to see an ad before converting.
When to shorten to 0 (off): e-commerce and lead generation campaigns where your audience already knows your brand, or when your view-through numbers dramatically exceed your click-through numbers without a corresponding lift in actual revenue.
When to keep it on: video-first brand awareness campaigns where recall and influence matter more than direct click tracking.
- Engaged-View Attribution
Engaged-view sits between a full click and a passive impression. It credits conversions from users who watched at least 10 seconds of your video ad (or 97% if the video is under 10 seconds) and converted within 24 hours.
This is a meaningful signal for video campaigns. A user who watched three-quarters of your testimonial video before moving on is demonstrably more engaged than someone who scrolled past a static image.
Activate this when you run video-first campaigns. Keep view-through off (or minimal) alongside it to avoid double-counting.
- Incremental Attribution
Incremental attribution is the most significant update to Meta’s attribution system in recent years, launched in 2025.
Where standard attribution asks ‘did a conversion happen after someone interacted with my ad?’ — incremental attribution asks ‘would this conversion have happened without my ad?’
Meta uses modelling to answer that question and only reports (and optimises towards) conversions it estimates were genuinely influenced by your advertising.
The tradeoff: you will likely see fewer reported conversions, even if your actual business outcomes remain the same. This can feel alarming, but it’s a more honest signal.
Who should use it: advertisers with consistently high conversion volumes and established campaigns that already exit the learning phase. If you’re still building volume, the stricter signal can make optimisation harder.
- First Conversion vs. All Conversions
All Conversions (the default) counts every conversion within the attribution window — including multiple conversions from the same person. If one customer clicks your ad and makes three purchases in a week, Ads Manager reports three results.
First Conversion limits Meta to counting only the first conversion per person within the attribution window.
In terms of optimisation, switching to First Conversion tells Meta not to value a repeat purchaser any more than a one-time buyer — which matters depending on your business model.
- E-commerce with high repeat purchase rate: All Conversions is appropriate
- Lead generation: First Conversion is almost always the right setting — you don’t want to reward duplicate form submissions
- Subscription models: First Conversion focuses Meta on acquisition, not re-engagement within the same window
Attribution Setting vs. Reporting Window — Don’t Confuse These
This is one of the most common sources of confusion for advertisers and their agency teams.
Your attribution setting (configured in the ad set) is what Meta uses to count conversions and optimise your campaigns. Your reporting window (configured in Ads Manager’s columns view) is simply a filter on what data you see in your dashboard.
You could have a 7-day click attribution setting but be viewing your dashboard on a 1-day reporting window — and see very different numbers without anything in your campaign having changed.
Best practice: always ensure your reporting window matches your attribution window when reviewing campaign performance. This is particularly important when presenting results to clients or comparing performance across campaigns.
Common Mistakes to Avoid
- Changing attribution settings mid-campaign — this resets learning and makes before/after comparison meaningless
- Using incremental attribution when you struggle to exit the learning phase — the reduced volume makes optimisation harder
- Leaving the 1-day view-through window active on always-on campaigns without checking if view-through numbers are inflating results
- Comparing Ads Manager results to CRM data without accounting for the reporting window difference
- Customising attribution settings without a specific, documented reason — the defaults are defaults for good reason
Conclusion: A Decision Framework for Attribution Settings
Before changing any attribution setting, ask yourself:
- What specific problem am I trying to solve?
- Do my current results look suspicious compared to my business outcomes?
- Am I already generating enough conversion volume for the algorithm to learn?
- Will this change make it harder to compare performance over time?
Start with Meta’s defaults. They are sensible baselines for most campaigns. Make changes intentionally, one at a time, with a documented reason — and always align your reporting window with your attribution setting when reviewing results.
Attribution isn’t glamorous. But it’s the difference between campaigns that optimise towards real growth and campaigns that optimise towards impressive-looking dashboards.
FAQs
1.What is the default attribution setting in Meta Ads?
The default is 7-day click and 1-day view-through. This means Meta credits conversions that happen within 7 days of a click or 1 day of a view of your ad.
2.What is incremental attribution in Meta Ads?
Incremental attribution, launched in 2025, uses predictive modelling to estimate whether a conversion would have happened without your ad. It only credits conversions that were genuinely influenced by your advertising — making it a more accurate but typically lower-volume attribution method.
3.Should I use first conversion or all conversions in Meta Ads?
For lead generation, use First Conversion to avoid counting duplicate form submissions. For e-commerce with repeat buyers, All Conversions is appropriate as it gives Meta signal on total purchase value within the attribution window.
4.Why do my Meta Ads results not match my CRM?
The most common reasons are: view-through conversions being counted that your CRM cannot track, the reporting window in Ads Manager not matching your attribution window, or multiple conversions from the same person being counted under All Conversions. Check each of these before assuming there is a tracking error.