# What is incrementality in advertising?

> Incrementality is the extra sales your ads cause that would not have happened without them. Learn how it differs from attribution, with a worked example.
- **Author**: Marxx
- **Published**: 2026-10-26
- **Category**: Measurement
- **URL**: https://marxx.ai/library-glossary/incrementality

---

Incrementality is the extra result your ads cause, compared with what would have happened with no ads at all. If 1,000 people buy after seeing your ad, but 700 would have bought anyway, only 300 sales are incremental.

## Attribution vs incrementality

| | Attribution | Incrementality |
|---|---|---|
| Question | Which ad did the buyer touch? | Did the ad cause the sale? |
| Method | Tracks clicks and views | Compares people who saw ads with people who did not |
| Common result | Generous to ads | Often lower than attribution |

Attribution counts a loyal customer who clicked a retargeting ad before buying. Incrementality asks if they would have bought anyway.

## A worked example

You split an audience of 2,00,000 people at random.

- Group A (1,00,000) sees your ads. 2,400 buy.
- Group B (1,00,000) sees no ads. 1,800 buy.

Incremental sales = 2,400 − 1,800 = **600**. Lift = 600 / 1,800 = 33%.

If the ads cost INR 1,50,000 and each sale is worth INR 1,000, incremental revenue is INR 6,00,000. That is an incremental ROAS of 4.0. Ads Manager may have reported 2,400 purchases and a ROAS of 16. The test tells you the real return.

## How teams measure it

- **[Holdout tests](/glossary/holdout-test):** hold back a random group from seeing ads.
- **Geo tests:** turn ads off in some cities and compare them with similar cities.
- **Meta conversion lift tests:** Meta splits people into a test group and a control group for you, in its Experiments tool, on eligible accounts. As a guide, Meta asks for a campaign from the past year with $5,000 or more in spend and 500 conversions (checked 2026-10-20).
- **Marketing mix modelling:** a statistical model across months of spend and sales.

## FAQ

### Why does incrementality matter?

Because budgets follow results. Retargeting can look best in attribution and weakest in incrementality, since it reaches people who were likely to buy anyway. Without a test you may move money to ads that cause little.

### Can a small brand test incrementality?

Yes. A simple geo test, with ads off in a few similar cities for three to four weeks, needs no special tools. Compare total orders by city against the weeks before.

---
- [More Measurement articles](https://marxx.ai/library-glossary/category/measurement)
- [All articles](https://marxx.ai/library-glossary)