Incrementality
Also known as: incremental lift, incremental impact
Incrementality measures the additional conversions that happened because of a marketing effort, versus what would have occurred anyway without it. It is usually tested with controlled experiments, such as holdout groups or geo tests, that compare an exposed audience against a comparable unexposed one.
Key Takeaways
- Incrementality measures the additional conversions caused by a marketing effort versus what would have happened without it.
- It is usually tested with controlled experiments such as holdout groups or geo tests that compare exposed and unexposed audiences.
- Standard attribution can credit channels for conversions that would have occurred anyway, overstating their value.
- Branded search and retargeting often show high attributed conversions but lower true incremental lift.
- Incrementality protects budget by revealing which spend actually creates new demand rather than claiming existing demand.
How It Works
Incrementality isolates cause from coincidence. Instead of counting every conversion that touched an ad, it asks how many conversions happened only because of the marketing. The standard method is an experiment: withhold ads from a random holdout group, or run a geo test where some regions see the campaign and comparable regions do not, then compare outcomes.
The difference between the exposed and unexposed groups is the incremental lift. This matters because an Attribution Model assigns credit based on touchpoints, which can flatter channels that merely intercept buyers already headed to purchase.
At larger scale, Marketing Mix Modeling estimates incrementality across channels using aggregate data and statistical modeling. Either way, the goal is to spend where added Conversion Rate is real. Combined with Customer Lifetime Value, incrementality tells you not just which campaigns get credit, but which ones genuinely grow the business.
Why It Matters
Standard attribution often credits channels for conversions that would have happened regardless. Incrementality reveals the true added value of spend, protecting budget from going to campaigns that merely take credit rather than create demand.
Example
A SaaS site suspects its retargeting ads look better than they are. It runs a holdout test, withholding retargeting from a random slice of eligible users. The exposed group converts only slightly more than the holdout, revealing that most of those buyers would have returned anyway. The team shifts budget toward channels that show a larger gap between exposed and unexposed groups.
Common Mistake
Confusing attributed conversions with incremental ones. A channel showing many attributed conversions, like branded search, may add little on top of what customers would have done anyway, overstating its real contribution.
Frequently Asked Questions
How do you measure incrementality?
Run a controlled experiment. A holdout test withholds ads from a random group and compares their conversions to an exposed group. A geo test compares matched regions with and without the campaign. The difference in outcomes is the incremental lift.
How is incrementality different from attribution?
Attribution assigns credit for conversions to the touchpoints along a path. Incrementality asks whether those conversions would have happened without the marketing at all. A channel can win attribution credit while adding little true incremental value.
Why do branded search and retargeting overstate value?
They tend to reach people already intending to buy or return. Attribution credits them for the conversion, but an incrementality test often shows those customers would have converted anyway, so the true added lift is smaller than it appears.
Is incrementality testing only for big advertisers?
No. Small advertisers can run simple geo holdouts or pause a channel in a controlled window and watch total conversions. The math is less precise at low volume, but the principle of comparing exposed and unexposed outcomes still applies.