Are my ads creating sales, or taking credit? Conversion lift: the sales your ads added, measured against a group that did not see them, and the ROAS on those sales alone.
Are my ads creating sales, or taking credit?
Full breakdown
Three fixes for your numbers
See the full breakdownThe numbers behind your answer and three fixes, also sent to your WhatsApp.
Are you losing money here? A senior marketer will find where, in 48 hours, free.
Get the free growth auditYour numbers look healthy. Book 30 minutes and we will show where to grow next.
Book a callWorked through with the typical numbers already in the calculator.
What you get
| Step | Formula | Result |
|---|---|---|
| Test group conversion rate | 2,400 ÷ 500,000 | 0.48% |
| Holdout conversion rate | 1,700 ÷ 500,000 | 0.34% |
| Extra purchases from ads | 2,400 − 0.34% × 500,000 | 700 |
| Incremental ROAS | 700 × ₹1200 ÷ ₹600,000 | 1.40 |
This is our rule of thumb for reading results. A test needs enough purchases in both groups to be trusted, so small tests are read with care.
| How we read conversion lift | |
|---|---|
| Under 5% | Ads add little. Most sales would happen anyway. |
| 5 to 20% | The ads add something, but not much. Check incremental ROAS against break-even. |
| Over 20% | Ads create a meaningful share of sales. |
This page was updated in October 2026 and reviewed by our senior marketing team.
We see these in most accounts we audit.
Platforms count sales the ad touched. Some would have happened anyway.
A tiny holdout gives a result that could be chance. Size the test before it starts.
New offers, prices or campaigns during a test muddy the result.
Incremental sales only matter if they make money after every cost.
These are the four levers we pull first, and the service behind each one.
Lever 1
We run Meta and Google conversion lift studies, or geo tests.
Fixed by tracking and measurementLever 2
We test retargeting and brand search first.
Fixed by performance marketingLever 3
We rank channels and audiences by incremental ROAS.
Fixed by E-commerce and D2C consultingLever 4
New angles reach people who would not have bought anyway.
Fixed by creatives and creatorsMeta or Google splits your audience. One group sees your ads, a holdout group does not. The difference is what ads created.
Incremental ROAS is the revenue from the extra purchases your ads caused, divided by ad spend. It is usually lower than platform ROAS, and it is the number that matters.
Platforms also count people who would have bought anyway, so their number includes sales the ads did not create.
People kept from seeing your ads on purpose. Comparing their purchases with the test group shows what the ads added.
You need the size of your test and holdout groups, purchases in each, your average order value and the ad spend in the test.
From a Meta or Google conversion lift study, or from a geo test where ads run in some regions and not in others.
As a rule of thumb, under 5% means ads add little, 5 to 20% is real but modest, and over 20% means ads create a meaningful share of sales.
Move budget towards channels and audiences with the highest incremental ROAS, and cut spend that only takes credit.
Test retargeting and brand search first, because they most often take credit for sales that would have happened anyway.
Big enough for the difference to be clear. Platform lift studies size the groups for you, and we size geo tests before you commit.
Yes, with a geo test: ads run in some regions and not in others, and the difference in sales shows what the ads added.
Five questions, no email needed. You get an honest answer on screen.
Great fit
Book 30 minutes and we will read your accounts before the call.
Book a callGood fit
You keep the findings either way.
Book a callNot yet
Our free tools will show you where your money goes today. When you cross ₹5 lakh a month, come back and we will read your accounts. If you would like advice before then, we are happy to talk.
Tell us about your brand. We read your accounts before the call, so it is about your numbers.
