RevenueLab.fyi
Revenue Rex logo mark
💰 Financial · Rex's Toolbox

Incrementality Lift Calculator

Convert a holdout or geo test result into true incremental revenue and iROAS.

Revenue Rex peeking

Rex says

Money math without the spreadsheet headache. Plug in your numbers and I'll show you exactly where the dollars land.

Try a scenario

Click to load — tweak from there.

Inputs

Result

Relative lift from exposure

15.0%

Incremental conversions

150

Incremental revenue

$12,750

Incremental ROAS (iROAS)

0.71

Incrementality factor vs. platform-reported

11%

Revenue Rex peeking

Psst — share this and help Rex grow

One click, a permanent link with your numbers baked in.

More financial

How to use this

  1. 1Enter test (exposed) group size.
  2. 2Enter conversions in test group.
  3. 3Enter control (holdout) group size.
  4. 4Enter conversions in control group.
  5. 5Enter average order value ($).
  6. 6Enter media spend on test group ($).
  7. 7Enter platform-attributed conversions.
  8. 8Read your relative lift from exposure on the right — it updates as you type.
  9. 9Hit Share to keep the scenario or send it to someone.

About this calculator

Platform-reported conversions almost always overstate what an ad channel actually causes, because they credit purchases that would have happened anyway. An incrementality test — a geo holdout, PSA/ghost-ad test, or matched-market pair — measures the real difference between an exposed group and an unexposed control. This calculator takes the conversion or revenue rate in your test group versus your control group, computes the lift, and translates that into incremental revenue and incremental return on ad spend (iROAS), which is almost always lower than platform-attributed ROAS. It also derives the 'incrementality factor' — the ratio of true incremental conversions to platform-reported conversions — which you can apply as a haircut to that channel's reported numbers going forward until you re-test. A factor of 0.4, for example, means only 40% of what the platform claims is actually incremental; the rest is conversions you'd have gotten anyway from brand search, direct traffic, or other channels picking up the slack. Use this after any properly randomized holdout, not on unmatched before/after comparisons, which conflate seasonality and trend with the channel's real effect.

FormulaLift = (test rate − control rate) ÷ control rate; incremental conversions = lift × control rate × test group size; iROAS = incremental revenue ÷ spend.

Worked example

Using the values the calculator loads with:

Inputs

  • Test (exposed) group size: 50000
  • Conversions in test group: 1150
  • Control (holdout) group size: 50000
  • Conversions in control group: 1000
  • Average order value: 85 $
  • Media spend on test group: 18000 $
  • Platform-attributed conversions: 1400

Results

  • Relative lift from exposure: 15.0%
  • Incremental conversions: 150
  • Incremental revenue: $12,750
  • Incremental ROAS (iROAS): 0.71
  • Incrementality factor vs. platform-reported: 11%

What each field means

Inputs

Test (exposed) group size
The test (exposed) group size used in the calculation. Starts at 50000 so you have a working example on load.
Conversions in test group
The conversions in test group used in the calculation. Starts at 1150 so you have a working example on load.
Control (holdout) group size
The control (holdout) group size used in the calculation. Starts at 50000 so you have a working example on load.
Conversions in control group
The conversions in control group used in the calculation. Starts at 1000 so you have a working example on load.
Average order value ($)
The average order value used in the calculation, measured in $. Starts at 85 $ so you have a working example on load.
Media spend on test group ($)
The media spend on test group used in the calculation, measured in $. Starts at 18000 $ so you have a working example on load.
Platform-attributed conversions
The platform-attributed conversions used in the calculation. Starts at 1400 so you have a working example on load.

Results

Relative lift from exposure
Returned as a percentage and shown as the headline result. It recalculates instantly whenever you change an input, so you can compare scenarios without reloading.
Incremental conversions
Returned as a whole number. It recalculates instantly whenever you change an input, so you can compare scenarios without reloading.
Incremental revenue
Returned as a money amount in US dollars. It recalculates instantly whenever you change an input, so you can compare scenarios without reloading.
Incremental ROAS (iROAS)
Returned as a decimal number. It recalculates instantly whenever you change an input, so you can compare scenarios without reloading.
Incrementality factor vs. platform-reported
Returned as a percentage. It recalculates instantly whenever you change an input, so you can compare scenarios without reloading.

FAQ

How big does the control group need to be?

Big enough to detect the lift you expect with confidence, same logic as an A/B test. For most brand or upper-funnel channels expect lifts of 5-20%, which typically needs tens of thousands of users per arm. Use the A/B test sample size calculator with your expected lift to size it properly before launching.

Why is iROAS almost always lower than platform ROAS?

Platforms credit any conversion within their attribution window that touched their ad, including people who would have bought anyway through organic search, direct navigation, or another channel. iROAS only counts the conversions caused by the exposure, which by definition is a subset of what's attributed.

What's a normal incrementality factor?

Brand search commonly runs 10-30% incremental because most of that traffic would convert anyway. Prospecting and upper-funnel video tends to run higher, 40-70%, since it's driving people who wouldn't otherwise have found you. Retargeting varies wildly, 20-60%, depending on how tightly it's scoped to recent high-intent visitors.

Can I run this without a true randomized holdout?

You'll get a directional number from a geo or time-based comparison, but it will be contaminated by seasonality, competitor activity, and macro trends. Randomized user-level or matched-market holdouts are the only way to isolate the channel's true causal effect.

Accuracy and limitations

  • Results are estimates before tax, fees, and inflation unless an input explicitly covers them.
  • Rates are treated as fixed for the whole period — variable-rate products will drift from this projection.
  • This is educational maths, not financial advice. Check anything contractual with the lender or your accountant.

Related tools

A/B Test Sample Size Calculator

How many visitors per variant you need before you look at results.

Attribution Window Shift Calculator

See how much your reported CAC and ROAS change when you shorten or lengthen the attribution window.

Cite this calculator

Writing about this topic? Grab a citation — every link helps keep these tools free.

APA
RevenueLab. (2026). Incrementality & Lift Test Calculator. Retrieved from https://www.revenuelab.fyi/toolbox/incrementality-lift-calculator
HTML
<p>Source: <a href="https://www.revenuelab.fyi/toolbox/incrementality-lift-calculator" target="_blank" rel="noopener">Incrementality & Lift Test Calculator — RevenueLab</a> (2026).</p>
Markdown
Source: [Incrementality & Lift Test Calculator — RevenueLab](https://www.revenuelab.fyi/toolbox/incrementality-lift-calculator) (2026).
Advertisement