# The best incrementality testing tools > Incrementality testing tools compared: Meta and Google lift tests, the free GeoLift package, Haus, Measured, INCRMNTAL, Lifesight and Recast. Checked 2026-10-20. - **Author**: Marxx - **Published**: 2026-11-18 - **Category**: Reporting tools - **URL**: https://marxx.ai/library-best/best-incrementality-testing-tools --- The best incrementality testing tools to start with are the free ones: Meta's lift tests, Google's Conversion Lift and Meta's open-source GeoLift package. Paid platforms such as Haus, Measured, INCRMNTAL, Lifesight and Recast run the tests for you. None of the paid ones list a price, so you book a demo. An incrementality test shows how many sales your ads caused, not just how many they were present for. It holds back some people or places from your ads and compares them with the rest. See [incrementality](/glossary/incrementality) for the basics. ## Quick comparison | Tool | Type | What it needs | Price | |---|---|---|---| | Meta lift tests | Built into Meta's Experiments tool | A Meta ad account | No fee on the pages we read. You pay for the ads | | Google Conversion Lift | Built into Google Ads | Not open to every account. User-based tests need 1,000 observed conversions and a $5,000 campaign budget | No fee on the pages we read | | GeoLift (Meta) | Open-source R package, MIT licence | Someone who can run R | Free to use | | Haus | Paid platform | Book a demo | Not listed publicly | | Measured | Paid platform | Book a demo | Not listed publicly | | INCRMNTAL | Paid platform | Book a demo | Not listed publicly. Not tied to ad spend | | Lifesight | Paid platform | Book a demo | Not listed publicly | | Recast | Paid platform | Book a demo | Not listed publicly | Checked 2026-10-20 on each company's own pages. ## The free options **Meta lift tests.** You create them in Meta's Experiments tool. Meta's getting-started guide recommends beginning with an account-level test. The tool makes a test group and a holdout group, then compares them. Meta's help page says a confidence of 90 percent or higher on a lift test is a reliable result. Meta defines conversion lift percent as the extra conversions among people who saw your ads, compared with the conversions you would have had without them. **Google Conversion Lift.** Google's help page says it is not available for all accounts and you contact your Google account representative to use it. The user-based test needs at least 1,000 observed conversions and a $5,000 minimum campaign budget, and Google allows studies as short as 7 days but typically recommends more than 14 days. The geography-based test needs campaigns that target a single country. **GeoLift.** Meta's open-source package for tests that compare regions. It helps you choose which regions to test and reads the result. See [geo-lift tests](/glossary/geo-lift-test). ## The paid platforms **Haus.** Runs incrementality experiments and also offers a causal marketing mix model and causal attribution. It lists four plans with no prices. **Measured.** Describes automated geo and audience split experiments, and a media mix model calibrated by those tests. It speaks to enterprise brands. **INCRMNTAL.** Its site says it measures incrementality without user-level data and without pausing campaigns. Its pricing page says the subscription is not related to your ad spend. It lists SMB, Pro, Enterprise and agency plans. **Lifesight.** Geo-based incrementality testing is on its Performance plan. Plans are annual subscriptions scoped to your business. **Recast.** Offers geo experiments next to its marketing mix model, with forecasting. It shows no prices. ## What a test result looks like Say you spend INR 5,00,000 in a four-week test. Meta's own report credits your ads with 500 orders. The test says that without the ads you would still have got 400 orders. 1. Incremental orders are 500 - 400 = 100. 2. Lift is 100 / 400 = 25%. 3. Cost per order as reported is INR 5,00,000 / 500 = INR 1,000. 4. Cost per incremental order is INR 5,00,000 / 100 = INR 5,000. The second price is the one that tells you if the ads earn their keep. These figures are made up. ## How to choose 1. Start with a free lift test on your biggest channel. 2. Move to a paid tool when you want repeat tests on many channels. 3. If you also want to split budget across channels over time, read the [best marketing mix modeling tools](/best/best-marketing-mix-modeling-tools). The glossary entry on [holdout tests](/glossary/holdout-test) shows the idea behind all of these. We make Marxx. It does not run lift tests. It reads your connected Meta ad account in full, with creative tagged by hook, angle, format and offer, so you can choose which creative is worth testing. For more on lift studies, read [what a conversion lift study is](/glossary/conversion-lift-study). ## FAQ ### How long should an incrementality test run? Google allows its user-based study to be as short as 7 days but typically recommends more than 14 days. Meta treats a confidence of 90 percent or higher as a reliable lift result. ### Is an incrementality test the same as an A/B test? No. An A/B test compares two ads. An incrementality test compares people who saw ads with people who did not. ### Do I need a paid tool? Not to start. Meta, Google and GeoLift are free to use. Paid tools save time when you run tests often. --- - [More Reporting tools articles](https://marxx.ai/library-best/category/reporting-tools) - [All articles](https://marxx.ai/library-best)