Conversion Lift
Conversion lift is the incremental increase in conversions caused by a specific campaign, treatment, or page change, measured against a control group that did not receive it.
In depth
Conversion lift is measured with a holdout or randomized experiment: one group is exposed to the treatment and a comparable control group is not, and the difference in conversion rate is the lift. Because it isolates cause from correlation, lift answers the question attribution models cannot — how many conversions would not have happened without this campaign or change.
The biggest pitfall is treating raw before-and-after deltas as lift; without a true control, seasonality, other campaigns, and existing demand contaminate the result. Teams running quiz funnels use lift studies to prove that a new landing-page variant or retargeting flight genuinely produced more qualified leads, rather than simply reshuffling leads they would have captured anyway. This makes lift the gold standard when budget decisions are on the line.
Example in practice
Frequently asked questions
How do you measure conversion lift?
You run a controlled experiment with a treatment group and a holdout control group, then compare conversion rates. The difference, ideally tested for statistical significance, is the lift attributable to the change.
Why is conversion lift better than attribution reporting?
Attribution distributes credit for conversions that already happened, but it cannot say which ones were truly caused by a campaign. Lift uses a control group to isolate incremental, causal impact.
What sample size do I need for a lift study?
It depends on your baseline conversion rate and the minimum lift you want to detect, but smaller effects require much larger samples. A power calculation before launch prevents inconclusive results.