Book a call
Services

Measurement strategy that starts with the decision

Most measurement arguments start with the method. MMM or MTA. Incrementality or ROAS. They should start with the decision someone needs to make, and pick the model second.

We help media owners, ad tech platforms and retail media networks build measurement their buyers actually believe.

The problem

Your buyers have their own models. When your numbers don’t line up with theirs, you get a polite nod in the meeting and your deck is ignored afterwards. More reporting rarely fixes that. Measurement designed around the buyer’s decision usually does.

What we do

  • Measurement audit. What you report today, what your buyers believe, and where the gap is.
  • Marketing mix modelling. Scoping, building or pressure-testing MMM, including open-source approaches.
  • Incrementality testing. Test design (geo, holdout, matched market), sizing, and reading results honestly.
  • Attribution. Where it helps, where it misleads, and how it sits alongside MMM and testing.
  • Measurement roadmap. How the methods fit together, and what to put in front of buyers at each stage.
Three methods, three decisions Method The decision it suits Marketing mix modelling Budget across channels, over time Incrementality testing Did this activity cause the result? Attribution In-flight optimisation within digital Testing checks the other two
Three methods, three decisions Marketing mix modelling → Budget across channels, over time Incrementality testing → Did this activity cause the result? Attribution → In-flight optimisation within digital Testing checks the other two
Each method answers a different question. Testing is how you check the other two.

When the data needs work before any model can run, our data integration and predictive analytics services cover that layer.

How it works

  1. Decision first. What does the buyer need to decide, and what evidence would change their mind?
  2. Audit. Your data, your current methods and your buyers’ methods.
  3. Design. The right mix of models and tests for that decision.
  4. Build and test. With your team, or alongside your measurement partner.
  5. Report in the buyer’s language. Results a buyer can check against their own numbers.
How a mix model splits a year of sales Base Marketing-driven Base 62% Paid search 14% TV 11% Social 8% Email 5% Illustrative example.
How a mix model splits a year of sales Base Marketing-driven Base 62% Paid search 14% TV 11% Social 8% Email 5% Illustrative example.
Marketing mix modelling separates what your spend actually caused from what would have happened anyway.

Why work with us

Adaptiv’s founder, Carl Carter, spent 15+ years on the other side of the table. At IRI and then Circana, where he ran media measurement and analytics across EMEA, he built the models hundreds of global brands used to judge their media partners. We’ve seen which numbers a buyer believes, and which ones get ignored.

Along the way: IPA and Digiday awards, and a seat on the advisory board at Dragonfly AI.

FAQ

Should we use MMM or attribution?

Usually both, for different decisions. MMM suits budget allocation across channels over time. Attribution suits in-flight optimisation within digital channels. Testing is how you check either one. The right answer depends on the decision in front of you.

Do you build the models or advise?

Both. Some clients want a strategy and a brief for their existing partner. Others want the model built.

Do you only work with media sellers?

Mostly media owners, ad tech platforms and retail media networks. Brands with a specific measurement decision are welcome too.

What decision are you trying to make?

Talk about measurement