Why Attributed Revenue Is Not the Same as Incremental Revenue
A customer receives an email, clicks, and purchases. The platform attributes the revenue to the email. The report appears complete.
What the report does not answer is whether the customer would have purchased anyway.
Attribution Describes a Sequence
Attribution assigns credit based on a rule: last click, last touch, view-through, or a platform-defined window. It connects marketing exposure to an observed transaction.
That information is useful for understanding customer paths and channel interaction. It does not prove causation.
Incrementality Estimates Causation
Incremental revenue is the difference between what happened with the treatment and what would have happened without it. Because the counterfactual cannot be observed for the same customer, teams estimate it through randomized control groups or credible experimental designs.
Why Lifecycle Attribution Is Especially Inflated
Lifecycle messages often target customers already likely to act: active subscribers, recent browsers, replenishment-ready buyers, and high-intent cart abandoners. These audiences produce strong attributed revenue even when the message changes little behavior.
Frequent messaging also creates more opportunities for a platform to claim credit for demand generated elsewhere.
Use Holdouts That Match the Question
A no-message holdout measures whether the program creates action. A current-state control measures whether a redesign improves performance. An untreated offer control measures whether an incentive changes behavior or subsidizes customers who would have converted.
Measure Profit, Not Only Revenue
Incremental revenue should be adjusted for discount, product margin, shipping, returns, and program cost. A treatment can produce positive incremental revenue and negative incremental contribution margin.
Use Attribution and Incrementality Together
Attribution helps teams operate channels and understand response. Incrementality helps leadership allocate investment and determine causal value. The two measures answer different questions and should not be forced into one number.
The Executive Takeaway
Attributed revenue tells you which touchpoint was present. Incremental revenue tells you whether the touchpoint mattered. Senior lifecycle teams report both—and make investment decisions with the second.
How to design an incrementality test
Start by defining the decision the test will inform. For example, should a brand continue a weekly promotional email to recent purchasers, or would those customers have purchased without it? Randomly assign eligible customers to a treatment group and a persistent no-send holdout before the campaign begins. Keep eligibility, attribution windows, and other channel exposures consistent.
Measure incremental orders and contribution margin per eligible customer, not just revenue attributed to recipients who clicked. Subtract the cost of incentives, messaging, and any fulfillment differences. If the treatment produces more attributed sales but little additional margin versus holdout, the apparent performance may reflect demand that already existed.
Use a sample-size and test-duration plan before launching. Customers with different purchase cycles require different observation windows, and a small experiment may not detect modest lifts. Report the estimated lift alongside uncertainty rather than declaring victory from a directional difference.
A practical reporting table should show treatment and holdout counts, orders, net revenue, contribution margin, unsubscribe rate, and the resulting incremental difference. Segment results by new versus repeat customers only when each subgroup has enough data to support a credible comparison.
Comments