Run A/B tests and personalize your store without code. 14-day free trial, no credit card.
Start free trial →In-depth guides and case studies where this concept is put to work.
Incrementality measures the additional conversions (or revenue) that were caused by a specific marketing activity — the uplift that would not have happened without that intervention. It's the answer to the question: "Did this campaign actually drive new purchases, or did it just reach people who were already going to buy?" Incrementality testing isolates the causal effect of a marketing channel or campaign by comparing a group exposed to it against a control group that was not.
Incremental Lift = (Conversion Rate of Exposed Group − Conversion Rate of Control Group) / Conversion Rate of Control Group × 100
For example, if your exposed group converts at 4.2% and your holdout control converts at 3.5%:
Incremental Lift = (4.2% − 3.5%) / 3.5% × 100 = 20% incremental lift
This means 20% of conversions in the exposed group would not have happened without the marketing activity.
Attribution models tell you which channel got credited for a conversion. Incrementality tells you whether that channel actually caused it. The difference is enormous for budget decisions.
Retargeting is the classic example. A retargeting campaign reliably shows high attributed ROAS because it reaches people who already visited your site and intended to buy. Many of them would have converted anyway — via direct navigation, branded search, or email. The retargeting ad just got in the way at the right moment. Incrementality testing reveals the true uplift, which is often far lower than attributed ROAS suggests.
For D2C brands spending ₹2–20 lakh/month on paid channels, even a 10–15% reduction in spend on non-incremental retargeting can free up budget for channels that are actually growing the customer base.
Boat runs retargeting campaigns for earphones priced at ₹999–₹2,999 and sees a reported 7x ROAS. To measure incrementality, they set up a holdout test: 10% of their retargeting-eligible audience is suppressed from seeing ads for 30 days. The holdout group converts at 3.1%; the exposed group converts at 3.8%. Incremental lift: 22.6%. That means about 77% of conversions in the retargeting campaign would have happened regardless. The actual incremental ROAS is roughly 1.6x — far lower than the attributed 7x — which changes how aggressively they invest in retargeting versus prospecting.
Incrementality and A/B testing share the same causal logic: compare an exposed group to a control group and measure the difference. Holdout tests are essentially A/B tests applied to marketing channels. The statistical principles — randomization, sufficient sample size, controlling for confounders — are identical.
Run smarter A/B tests with CustomFit.ai — 14-day free trial, no credit card required.