Incrementality

What is Incrementality?

Incrementality is the true causal lift from advertising, measured with holdout tests (geo splits, conversion lift, or media-mix modeling) rather than platform attribution, and it answers the one question ROAS and the marketing efficiency ratio (MER) cannot: did the spend cause the sale, or just take credit for one?

How it is measured

The formula is simple: incremental lift = conversions in the exposed group − conversions in a matched holdout that saw no ads. The practical version in health is a geo split. Pause a campaign in a set of matched markets, keep running it in comparable markets, and compare new-patient volume between them. Work an example with round, illustrative numbers. A telehealth brand’s branded campaign reports 8× ROAS and 500 attributed sign-ups a month. Hold it out in matched geos and new-patient volume there falls by only 100, so 400 of those 500 would have converted anyway (from an email, a friend’s recommendation, or an earlier ad), and the campaign’s real incremental contribution is a fraction of what the platform claimed. The gap does not mean the campaign is worthless (some of those 100 truly incremental patients may still clear your CAC bar), only that its budget should be set by the 100 it actually caused rather than the 500 it was credited with.

Where platform attribution overstates

Branded search and retargeting are the clearest offenders, because both mostly reach people already on their way to converting. The three common ways to measure real lift trade rigor against effort:

Why it matters for health and DTC brands

Health funnels are multi-touch and often run partly offline through intake and clinical review, so platform attribution misses more of the journey than it does in simple e-commerce, and the double-counting compounds. A quick over-claim check runs the other way: sum every platform’s reported revenue and compare it to what the bank account actually shows (the same logic the marketing efficiency ratio uses). If the platforms together claim more than the business truly made, the overage is the size of the attribution problem, and at least one channel is taking credit for conversions it did not cause. Let the incremental gap, not the platform’s attribution, set the budget on your most-suspect campaigns (branded search and retargeting first), and read incrementality alongside blended CAC and MER so the spend ceiling is set by causation rather than credit.

Related terms