Baseline and incremental lift
Every conversion total contains a baseline — what would have happened without the marketing — and an incremental portion driven by it. Incremental lift is that incremental portion: conversions a campaign actually caused, over and above the baseline. Confusing the two leads to crediting marketing for sales it did not cause. This page defines baseline and incremental lift and explains how experiments estimate the split.
Baseline versus incremental
Baseline is the level of conversions you would get with no campaign — driven by existing demand, brand, and organic discovery. Incremental lift is the additional conversions the campaign caused on top of that baseline.
The formula is conceptually simple: incremental lift equals the treated outcome minus the baseline (counterfactual) outcome. The hard part is estimating the baseline, since you never directly observe what would have happened without the campaign.
Estimating the split
Experiments estimate the baseline by holding out a comparable group from the marketing: holdout and PSA designs, geo-experiments, and synthetic controls all build a no-marketing counterfactual. The difference between treated and control is the incremental lift.
Attribution models, by contrast, distribute credit among touches but do not separate baseline from incremental — which is why a channel can look large in attribution yet contribute little incremental lift. The two views answer different questions.
- Incremental lift = treated outcome − baseline outcome
- Baseline = conversions without the marketing
- Attribution distributes credit; it does not isolate lift
How it appears in analytics and logs
A high attributed total with low incremental lift means much of the credited volume would have converted anyway; the marketing's true contribution is the incremental part.
Diagnostic use case
Frame measurement around the question 'how many conversions did this cause?' rather than 'how many conversions touched this channel?' — separating baseline from incremental.
What WebmasterID can help detect
WebmasterID's observed conversion counts by segment and period give you the aggregated outcomes needed to estimate baseline versus incremental lift in an experiment.
Common mistakes
- Treating all attributed conversions as incremental.
- Estimating a baseline without a real holdout or control.
- Confusing attribution credit with causal lift.
Privacy and accuracy notes
Lift is measured from aggregated treated-versus-control outcomes, not individual tracking. This is educational, not legal advice.
Related pages
- Incrementality testing: what attribution cannot tell you
Incrementality testing measures the lift a channel actually causes by withholding it from a control group and comparing outcomes. It answers the question every attribution model dodges: would this conversion have happened anyway? It is causal where attribution is merely correlational, but it requires deliberate experiment design.
- Holdout-based attribution
Holdout-based attribution uses a randomized holdout — a group deliberately excluded from a campaign or channel — to estimate how much of a channel's credited conversions are genuinely incremental. By comparing the treated population against the holdout, it grounds attribution in a counterfactual rather than relying solely on observed click paths, which tend to over-credit channels that intercept already-converting users.
- Cannibalization in measurement
Cannibalization in measurement is the opposite of a halo: a channel captures conversions that another channel — often organic or direct — would have delivered anyway. Branded paid search bidding on terms users would have clicked organically is the canonical case. Attribution credits the paid click, but the incremental value may be small. This page explains cannibalization and how incrementality testing exposes it.
- Attribution analytics
Aggregated outcomes for baseline-versus-lift estimation.
Sources and verification notes
- Google Ads Help — About Conversion liftDocuments measuring incremental conversions over a control baseline.
Last reviewed 2026-06-24. Facts are checked against primary/official sources where available; uncertain specifics are marked “Data not yet verified” rather than guessed.