In this guide
- A conversion lift study compares people who saw your ads with a comparable group who did not, and measures the difference in conversions.
- Meta and Google offer built-in lift tests, and you can run a geo or holdout test yourself.
- Use lift for big budget decisions. Use attribution for day-to-day optimisation.
Every attribution model is a guess about causation. A lift study answers the causal question directly: would these sales have happened without the ads? That makes it the strongest evidence you can get, and the most expensive to run. This page explains how lift studies work, when to run one and how to read the result. It connects to marketing attribution models and marketing mix modelling.
The idea
Split a population into a test group that can see your ads and a control group that cannot. Run the campaign. Compare conversions per person in each group. The difference is the lift: the extra conversions caused by the ads. Because the groups are chosen randomly or matched carefully, other factors even out.
Check your tracking in one pass
This is an example. Run the check on your own site to see where yours stands.
Ways to run one
| Method | How | Good for |
|---|---|---|
| Platform lift test (Meta, Google) | The platform holds out a random share of eligible users | Measuring one platform's ads, with platform-run mechanics |
| Geo holdout | Pause ads in some regions and compare sales with similar regions | Any channel, using your own sales data |
| Time-based on and off | Switch a channel off for a period and watch the effect | Quick, rough checks, with caution about seasonality |
| Ghost ads or PSA tests | Show a neutral ad to the control group | Larger advertisers |
What you need
- Enough volume. Small effects need many conversions to detect.
- A clean setup. No other big changes during the test.
- A clear question. For example, does paid social drive incremental purchases for new customers?
- Reliable outcome data. Your own orders by region or group, ideally.
3 things to remember
Scroll and the points stack up.
A conversion lift study compares people who saw your ads with a comparable group who did not, and measures the difference in conve
Meta and Google offer built-in lift tests, and you can run a geo or holdout test yourself.
Use lift for big budget decisions. Use attribution for day-to-day optimisation.
Reading the result
| Result | Meaning |
|---|---|
| Positive lift, cost per incremental conversion acceptable | The channel works at this spend |
| Small or no lift | The channel mostly credits itself for sales that would have happened |
| Lift with wide uncertainty | You need more data or a longer test |
Look at the cost per incremental conversion, not only the lift percentage. A large lift on a tiny base can still be expensive.
How tracking quality matters
A lift test is only as good as the outcome data. If your conversions are under-reported, the effect is harder to detect. Server-side events, good matching and your own order data make the results more reliable. Open Tracked's analytics can show orders and revenue by source and region from first-party data. See attribution with server-side tracking.
Use the result
- Compare incremental cost per conversion with your margin and target.
- Calibrate: if the channel's platform ROAS was 4.0 and incremental ROAS is 1.5, apply that ratio to future reports.
- Re-run on a schedule for large channels, for example yearly.
Orders from your store, back to the ad that earned them
Your store sends the order once. Open Tracked passes it on to each ad platform, cleaned and deduplicated.
Frequently asked questions
What is a conversion lift study?
A test that compares conversions in a group that saw your ads with a comparable group that did not, to measure what the ads caused.
How is it different from attribution?
Attribution splits credit among touchpoints. Lift measures causation.
Can I run one without a platform tool?
Yes, with a geo holdout using your own sales data.
How much does it cost?
Platform tests are usually free but need budget and volume. Geo tests cost the sales you give up in the holdout regions.
How often should I run them?
For large channels, about once a year, or when you plan a big budget change.
Do I need spreadsheets to read my results?
No. Once tracking is live, the Open Tracked analytics shows spend, revenue, ROAS and profit, and you can ask Ari, the built-in AI assistant, questions about your own numbers in plain words. Ari's allowance depends on your analytics plan.
Is Open Tracked different from Open Track or Open Space Track?
No. Open Tracked is the official name of our server-side tracking platform, and the website is opentracked.io. Many people search for it as Open Track, OpenTrack or Open Space Track. All of them lead to the same product: server-side Google Tag Manager hosting, first-party analytics and a CRM pipeline.
Fix your tracking at the source
Open Tracked hosts your server-side Tag Manager, keeps first-party cookies alive and shows what each ad is really worth.
The Open Tracked team
We build Open Tracked: server-side tracking hosting, analytics and the tools around them, for stores that spend on ads. Questions about this guide, or an idea for the next one?
Comments
Loading comments...
Guides · 5 min read
