Core question
The core question incrementality answers is:
Did this marketing actually cause the conversion, or would it have happened anyway?
How it works
Incrementality is typically measured through holdout experiments (also called lift tests):
Split your audience into two groups randomly.Test group: Exposed to your marketing (for example, sees your ads).Control group: Not exposed (held out from seeing ads).Compare conversion rates between both groups.Incrementality formula
Incrementality = (Test\ Group\ Conversions - Control\ Group\ Conversions) \div Test\ Group\ Conversions
If your test group converts at 5% and your control group at 3%, your incremental lift is 2 percentage points. This means 40% of your conversions were truly incremental, while 60% would have happened regardless.
Why it matters
Common cases where traditional attribution can be misleading:
Brand search ads: Someone searches your brand name and clicks your ad. They may have converted organically anyway.Retargeting: You retarget a user who already had items in their cart. The ad may not have caused the purchase.Coupon sites: A user searches for a discount code at checkout. The affiliate gets last-touch credit, but they did not actually drive the sale.Incrementality helps you separate true causal impact from conversions that would have happened without the marketing touch.
Limitations
Requires enough traffic volume to run statistically valid experiments.Can be complex to design and implement correctly.Platform-run lift tests (for example, Meta or Google) may have conflicts of interest.Typically measures specific campaigns or channels, not the entire customer journey.