Understanding Meta’s new Incremental Attribution
What It Means, Where It Helps, and How to use it to your advantage
Meta recently introduced a new Incremental Attribution setting in Ads Manager—aimed at answering a critical question in digital advertising: “Did this ad actually cause the conversion, or would it have happened anyway?”
This model is Meta’s attempt to move beyond traditional last-click attribution and towards a more meaningful evaluation of Ad impact. But while it’s a step forward, the real challenge lies in interpreting results within Meta’s closed ecosystem and balancing it with third-party validation.
Here's what you need to know.
What Is Incremental Attribution?
Traditional attribution models like First Click, Last Click, or Linear credit parts of the user journey but don’t ask whether the user would have converted anyway without the ad.
Incremental attribution flips this around—it focuses on causality.
Nikita, the manager of a Superstore decided to increase her Sales by hiring an external sales person Greg. To track Greg’s sales, she gave him a Code to share with all customers referred by Greg.
After one week, she sees that Greg is contributing an astounding 100 Customers per day. But her total customers per week had stayed flat at 220 per day.
The results made no sense.
She paid Greg’s commission, but she decided to visit the store to see what was happening. What she saw shocked her. Greg was standing near the checkout counter and giving his code out to people who were already waiting in line to complete their purchase.
Greg was getting lots of Sales credited to him, but Greg was giving no incremental results. This is where Incremental Attribution comes in.
BTW, if you think the above example is contrived, checkout a real life multi million dollar scam doing exactly this.
How does Meta implement Incrementality?
According to Meta’s release notes →
Meta’s Incremental Attribution Model uses machine learning, trained on data from controlled lift tests, to predict how likely it is that a given conversion happened because of the ad rather than organically.
Meta’s implementation is still a black box, but from the little they have released we know that
Meta compares ad-exposed users with similar unexposed users (a synthetic control group).
It predicts whether a conversion is truly incremental.
Campaigns can be optimised directly based on these incremental conversions instead of all conversions.
Using Incremental Attribution in Meta Ads
Meta has provided a helpful guide to use and analyse results from Incremental Attribution.
How to use incremental attribution in Meta Ads Manager
How to view results for incremental attribution in Meta Ads Manager
Setting up Incremental Attribution in Meta Ads Manager
You can switch to Incremental Attribution in just a few clicks:
Create Your Campaign: Choose your campaign objective as Conversions, Product Catalog Sales, or Sales (purchase optimised).
Access More Options: In the campaign setup within Ads Manager, locate “Show more options” under your performance goal section.
Select Incremental Attribution: Choose the Incremental Attribution setting and let Meta’s AI-driven model do the rest.
Reporting Incremental Conversions
Once your campaign is live, the Incremental Attribution results will appear directly in Ads Manager. Initially, these metrics will appear only in the results column, but Meta plans to expand functionality, enabling you to compare incremental conversions to rules-based conversions.
What are the benefits of this?
It’s a native Meta setup. Easy to setup, and easy to track. It is available directly within Ads Manager. No additional tools or manual setup is required.
Since this focuses only on conversions caused by ads, removing organic purchases that mislead ROAS can help advertisers move spend to campaigns that actually drive net-new business.
Meta’s initial claims suggest that Ads optimised toward incrementality may improve campaign efficiency by 15–25% .
Better Signal, Less Noise → Smarter Budget Allocation → Automatic Optimisation
What are some Limitations that Meta is not talking about?
Walled Garden Syndrome
Meta is performing these experiments in it’s own walled garden, completely ignoring the impact of other factors. It only measures it’s own impact, and does not track or even have the visibility of the influence of other channels (Google, email, influencers, etc.).
When there are no other channels everything will be hunky-dory.
But the moment there are other channels involved. Meta’s conclusions about incrementality can get murky.
It is liable to under-estimate it’s effect in the presence of other activity.
But it is also likely to over-estimate or it’s own Impact.
Other Limitations
There are a few limitations to this approach that Meta’s release notes have downplayed or ignored.
Black Box Modelling
Meta has not exposed the exact models or lift study assumptions used. We could not find any reference to the underlying maths → leaving no room for independent validation.
Bias Toward Meta’s Own Interests
At BooleanMaths, we are extremely cynical of any claims, unless we see the underlying data. In this scenario, as Meta benefits from looking impactful, its definition of "incremental" may be technically true, but also maybe a bit generous.
Retargeting can get complicated quickly.
Our assumption is that this setting would not be right for certain campaigns.
For example - Retargeting campaigns if run on Email or other channels might give Meta a false signals.
Meta might reduce budgets to non-retargeted cohorts as it sees less incrementality which will be extremely counterproductive and reduce new customer discovery.
On the other hand for targeted cohorts, Meta will get a false signal of incremental results and might over allocate budgets resulting in Ad spamming.
💡 Smart Strategy: Combine Meta + Independent Tests
Meta's incremental attribution model gives directional insight within the platform. But to make smarter cross-channel marketing decisions, you should validate it independently.
Suggested Hybrid Approach:
Enable Meta’s Incremental Attribution for insight and optimisation.
Run your own Incrementality experiments outside Meta to validate lift. We will cover how to do that in our next post.
Use reliable third-party tools for full-channel attribution.
Compare and calibrate results to find a more honest ROI number.
✍️ Final Thoughts
Meta’s Incremental Attribution Model is a step in the right direction—putting causality at the heart of performance measurement. But as always in digital marketing, no single platform should grade its own homework.
To truly understand ad effectiveness, brands must blend platform-native tools with independent experiments. That’s how you make the invisible visible—and your budget accountable.
In case you want to read more about different attribution models and the maths behind it - our three part series covers everything in nerdy detail.
👉 Attribution Maths - Heuristic Models
Really enjoyed reading this post, we've just published a handful of our experiments with Incremental Attribution including one alongside a holdout study we ran ourselves. Definitely inconclusive so far.
https://read.earlystagegrowth.com/p/incremental-attribution