Three measurement methods, explained with a football analogy.
Attribution, incrementality and marketing mix modelling (MMM) are often treated as interchangeable. They are not. Each one answers a different question, needs different data and fits a different point in the campaign cycle.
At Adsquare's Outcomes Era evening in Düsseldorf, Kolja Brosche, Co-Founder and Managing Director of SkillProjects, asked Adsquare CEO Tom Laband what agencies and advertisers can measure today. Tom's answer started with a comparison he had read on the train that morning.
The analogy: Three questions after a win
Your football team won on Saturday. Now there are three questions to answer:
- Attribution: Who gets the credit for the goal?
- Incrementality: Would we have won without our striker?
- MMM: How should we spend next season's transfer budget across the whole squad?
"With attribution, you look at whether you won because your striker actually scored. Incrementality asks whether you would have won the game without him. And MMM is about how you spend your entire transfer budget on the team next season."
Tom LabandCEO and Co-Founder, Adsquare
Attribution: Who scored?
Attribution gives credit for an outcome to the touchpoints that came before it. Exposure data is linked to outcome data by device or campaign ID. The result shows which campaign, publisher or creative the converters saw.
It works during and right after a campaign. That makes it useful for shifting budget between partners and creatives while there is still time to act.
The limit: attribution cannot tell you whether the outcome would have happened anyway. A loyal customer visits the store every Saturday. If she also saw the ad, attribution still credits the ad. Whether the ad caused the visit is a question for incrementality.
Incrementality: Would we have won anyway?
Incrementality compares people who saw a campaign with a similar group who did not. The difference between the two groups is the lift: the outcomes the campaign actually caused.
This is the question CFOs now ask. Not "who saw the ad before buying?" but "what did the spend add?" The usual methods are holdout groups and synthetic control groups built from comparable audiences.
The limit: an incrementality test usually measures one campaign or channel at a time. It also needs enough scale for the difference to be real rather than noise.
MMM: Planning the season
Marketing mix modelling works top-down. It takes aggregated spend and sales across all channels, usually weekly and over two years or more. It then estimates what each channel contributed, alongside factors like price, promotions and seasonality.
MMM answers planning questions. How much should go to TV, CTV, digital and out-of-home next year? It also shows where spend hits diminishing returns. A second star striker adds less than the first, and doubling spend on one channel rarely doubles its results.
The limit: MMM is slow and aggregated. It will not tell you which publisher to cut on Tuesday.
Closing the loop: Measuring offline outcomes
Most transactions still happen in stores. Ecommerce makes up an estimated 20.5% of global retail sales in 2025, which leaves about 79.5% offline (eMarketer). As Tom put it in Düsseldorf, the industry spent the last decade optimizing for online outcomes. Store visits and in-store purchases were much harder to measure.
Store visits can now be tied to the campaigns people saw, using pseudonymized device data. That puts offline outcomes into the same toolkit:
- Attribution, in flight: Link store visits to the campaigns, partners and creatives people saw, and shift budget while the campaign runs.
- Incrementality, after the campaign: Compare actual store visits with how many would have happened without the campaign. The difference is what the spend added.
- MMM, at planning time: MMM already works with total sales, including in-store sales. It sets next year's channel split.
A coach wants to know who scored, whether the striker is worth keeping and where to spend in the summer window. The mistake is using one method to answer another method's question.
Adsquare covers attribution and incrementality for store visits today. A daily Visit Data Feed supports optimization while a campaign runs. An end-of-campaign Attribution Report measures the incremental uplift in store visits.
Want to see what attribution and incrementality look like for your offline outcomes? Speak with one of our experts.