What Is Incrementality?
The Causal Truth Behind Your Marketing
Incrementality in marketing answers the question that attribution alone cannot: what did your advertising actually cause? Controlled geo testing establishes the causal truth for specific channels, then feeds that evidence back into your model to make every budget decision sharper.
Geo testing is the method. Incrementality is the outcome. MMM is the backbone that puts it all into strategic context.
What is incrementality?
The causal impact of advertising
Incrementality meaning is simple: it is the revenue your marketing genuinely caused. What would not have happened without it. Incrementality in marketing is the answer to “did our advertising actually drive that sale, or would the customer have bought anyway?” MMM estimates it statistically. Geo testing helps validate it.
How do you validate it?
Randomised geo experiments
Randomly assign geographic markets to test and control groups. Run the campaign in test regions and hold control regions at normal activity. The sales difference between groups — after controlling for everything else — is the incrementality. No individual tracking required.
MMM as the backbone. Incrementality as calibration.
Incrementality evidence from experiments anchors your model’s estimates in causal reality. MMM provides the strategic context that gives those estimates planning power.
Geo Testing
Randomised controlled trials providing direct causal measurement of incrementality, iROAS, incremental lift, and response curve shape that anchor MMM coefficients in observed reality
Always-ON Analytics
Continuously refreshed model outputs for tactical decisions. Budget shifts and channel reallocation informed by a model that updates with each new geo testing result
What You Will Get
Strategic Planning
Full model outputs inform annual budget allocation and channel strategy
Tactical Optimisation
Always-on analytics and incrementality inputs guide continuous adjustments
Integrated Measurements
Reliable, continuous insights to anticipate shifts in market conditions and make evidence-based strategic decisions.
The Gap Between Correlation and Causation is Where the Expensive Mistakes Happen.
Privacy-safe by design
Geo testing uses aggregated regional sales data. No individual identifiers, no cookie dependency. It is fully resilient to any privacy regulation change.
Defensible to finance
Budget recommendations anchored in causal incrementality evidence are substantially more defensible than model outputs alone. Finance teams respond to proof, not correlation.
Step 1: Identify the highest-uncertainty channels
MMM outputs reveal where confidence intervals are widest, channels with low spend variation, limited history, or heavy concurrent activity. These are the first candidates for geo testing.
Step 2: Design the geo experiment using MMM guidance
MMM informs the experiment design: how many weeks are needed, which regions provide the best test-control match, what spend variation maximises the chance of detecting a real signal. MMM makes the geo testing more likely to succeed.
Step 3: Run the experiment with proper controls
Randomly assign regions to test and control. Run media at planned weight in test regions; hold control regions at normal activity. Monitor delivery weekly to confirm treatment separation holds throughout.
Step 4: Calibrate the MMM with incrementality evidence
Experiment results, iROAS with confidence intervals, uplift curves, and response curve shape are integrated into the model as Bayesian priors or coefficient constraints. The model is re-run. Recommendations improve. The cycle begins again.
The Incrementality Flywheel
Build or refresh MMM
Identify low-certainty channels
MMM guides experiment design
Run geo experiment with controls
Measure incremental ROAS (iROAS)
Calibrate MMM for sharper, more confident recommendations
What calibrated incrementality measurement delivers
When experiments and MMM work together
+26%
Increase in measured incremental sales contribution after MMM calibration with experiment results
+24%
Improvement in measured ROI after incrementality evidence integrated into the model
+33%
Increase in budget share to tested channel following incrementality-calibrated model recommendations
£1B+
Incremental revenue identified from branded search; invisible to attribution, revealed by geo experiment
When Incrementality Evidence Changes the Decision
Budget defence
Before you cut, measure what it’s actually driving
A channel that looks low-ROI in your model may be driving significant incremental revenue through a pathway attribution can’t track. An experiment reveals the true causal impact before the budget decision is made.
New channel evaluation
Establish incrementality before you scale
Before committing serious budget to a new channel: CTV, audio, or retail media, a targeted geo test establishes the causal incrementality baseline your MMM needs to model it accurately.
Hard-to-measure media
Prove what attribution ignores
TV, radio, OOH, and brand advertising drive incremental sales through pathways digital attribution misses entirely. Geo experiments capture the full causal picture and calibrate the model to reflect it.
MMM validation
Prove your model’s recommendations work
Run a structured A/B test across markets to validate whether MMM‑driven budget optimisation delivers in practice, independent of the team that built the model. This provides real proof for finance teams and boards.
Model calibration
Give your MMM causal ground truth
Incrementality evidence from experiments anchors uncertain modelled coefficients in observed causal reality, narrowing confidence intervals, stabilising forecasts, and improving the reliability of all subsequent recommendations.
Continuous improvement
Build a library no competitor can buy
A sustained incrementality testing programme accumulates causal benchmarks across channels, markets, and seasons, creating proprietary evidence that makes your model more accurate with every test you run.
Article series
Getting Started with Incrementality Measurement
Six articles taking you from “our model estimates this” to “we have causal evidence for this.” Read the full series or start with the article most relevant to where you are now.
The Incrementality Gap: What Your Model Can’t Tell You
Article 1
MMM is one of the most powerful tools in marketing measurement. Yet there is one question it cannot answer on its own, and it is the one that matters most when budget is on the line: did our advertising actually cause those sales?
7 min read
How Geo Experiments Measure Incrementality
Article 2
Controlled experiments are the most rigorous method available for establishing what advertising truly caused. Here’s what a geo experiment is, why it works, and what separates a result you can act on from an expensive piece of noise.
8 min read
The $1 Billion Incrementality Question
Article 3
A major advertiser nearly cut branded search. An incrementality experiment proved it was driving over £1B in revenue through a conversion path attribution had missed entirely. The principle: measure twice before you cut once.
7 min read
How Incrementality Measurement Makes Your MMM Sharper
Article 4
Incrementality evidence from controlled experiments calibrates MMM coefficients — replacing uncertain statistical estimates with observed causal truth. Three approaches, and what changes in the model when they work.
8 min read
Which Channels Need Incrementality Testing First?
Article 5
Not every channel needs an experiment right now. The goal is to build a testing programme that creates a growing library of causal benchmarks, starting with the areas where incrementality uncertainty is highest and the cost of being wrong is greatest.
7 min read
Building an Always-On Incrementality Programme
Article 6
A single incrementality test is only a data point. A sustained programme builds a causal evidence library that no competitor can replicate. This is what Integrated Marketing Measurement looks like when it is working, and how to get there.
8 min read