What is the difference between lift and attribution?
Attribution splits the credit for sales you already made among the ads and visits before them, by a rule you can change. Lift counts the extra sales your ads caused, by comparing people who could see them with a control group who could not.
By Joris van Huët, Founder & CEOUpdated 7 min read
Run the numbers for your store: the free holdout test planner.
Attribution splits the credit for sales you already made among the ads and visits before them, using a rule you can change. Lift counts the extra sales your ads caused, usually by comparing people who could see them with a control group who could not. Attribution names a winner for every sale it counts. Lift can come back with zero.
Google Analytics defines attribution as the act of assigning credit for important user actions to the ads, clicks and other factors on a buyer's path. A model does the assigning: a rule, or a data-driven algorithm. Switch models and the past changes too. GA4 says a new reporting attribution model applies to historical and future data.
Lift comes from an experiment instead. Google's Conversion Lift splits your audience into people who see your ads and a control group who don't. The difference in conversions between these 2 groups is the lift. Meta defines lift the same way: the difference in conversions caused by the presence of your ads.
| Attribution | Lift | |
|---|---|---|
| The question | Which ad or visit gets the credit? | Would the sale have happened without the ads? |
| What it needs | Clicks, visits or views on the path | A control group kept away from the ads |
| When it runs | On every sale, every day | During a test with a start and an end date |
| Change its rule | The numbers change, history included | Nothing moves: lift ignores attribution settings |
| Can it say "nobody"? | No: every sale it counts lands in some row | Yes: a flat result means no extra sales found |
Google draws the same line: Google Ads counts attributed conversions by each action's settings, such as a 7-day click-through window. Conversion Lift ignores those settings and counts every conversion in both groups.
What one store's data shows
One store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts.
| What the export shows | Share of revenue | Source cell |
|---|---|---|
| Journeys with 1 touch (0.5 days to buy) | 79.5% | Journeys sheet, 1 touch row |
| Journeys with 2 to 3 touches (12.5 days to buy) | 12.2% | Journeys sheet, 2-3 touches row |
| Journeys with 4 to 9 touches (16.9 days to buy) | 5.4% | Journeys sheet, 4-9 touches row |
| Journeys with 10 or more touches (16.0 days to buy) | 3.0% | Journeys sheet, 10+ touches row |
| All channels in the touched view, added up | 110.4% | Channels sheet, Touched column total |
A model only matters when a path has more than one touch to share the sale. GA4's paths report puts it plainly: when a path has just one touchpoint, that touch gets all the credit.
In that store's Journeys sheet, journeys with 1 touch hold 79.5% of revenue, at 0.5 days to buy. Every attribution model scores that revenue the same way. A switch from last click to data-driven can only move the rest. In the same Journeys sheet, that is 12.2% + 5.4% + 3.0% = 20.6%.
So in one store, the whole model debate covers about a fifth of revenue. A lift test questions all of it, including the one-touch sales no model can touch.
The touched view on that store's Channels sheet credits every channel a journey met, and sums to 110.4%. Credit is a bookkeeping choice, so a generous rule can hand out more than the store took.
Lift has no total to inflate. Google says a lift study measures what your campaigns add on top of any other active marketing. Each result stands on its own control group. Whatever no test claims stays in the baseline: sales that would have happened anyway.
What the export cannot show is lift itself. With no spend and no control group, it can't say what any channel added. It shows how much revenue a model change could ever move in one store.
Why does the usual answer mislead?
The usual answer is a slogan: attribution is correlation, lift is causation. The platforms have blurred that line.
Attribution models now borrow lift's tools. GA4's data-driven model uses a counterfactual approach, and Google says it trains on data from randomized controlled trials. Google Ads says its data-driven model calibrates using incrementality signals. Meta's incremental attribution uses models that predict whether a conversion is caused by an ad.
None of them holds your own customers back, though. They still work sale by sale: in GA4, the data-driven credit for a key event sums to 1.0. A model trained on someone's experiments is still a model of your paths.
Lift is not a better attribution report. A lift result covers the campaigns in the test, between its dates. Meta says its lift results are unique to your test's conditions and not meant to be compared with Ads Manager. It says little about which ad did the work, or about next quarter.
Do you need both?
Yes: they do different jobs. Attribution steers inside a channel, every day, on every sale. Lift checks how far to trust that steering, a few times a year.
Meta names the second job outright. Among the questions a lift test can answer, it lists whether your current attribution model aligns with the results of a controlled experiment.
Google suggests testing before major budget decisions, and says most advertisers run about 1 to 2 studies per year.
The working link is a ratio: what the test found, divided by what attribution credited the same campaign over the same dates. Read it as a rough exchange rate, not an exact overcount, since the two count over different windows. The step-by-step version does it in GA4, Google Ads and Meta.
What to do this week
- Measure what a model switch could move. In GA4, click Advertising, then Key event attribution paths under the Key events dropdown. Note the purchase revenue in the top row, then filter Path length to greater than 1 touchpoint and note it again. Pass: you can say what share of revenue sits in paths of two or more touches. Fail: the top row shows no purchase revenue, so check that purchase is marked as a key event.
- Find the channel whose value depends on the rule. In GA4, click Advertising, then Attribution models under Attribution. Compare Data-driven with Paid and organic last click and read the revenue % Change column. Pass: no channel swings much, so model choice barely matters. Fail: one channel swings hard, so its value rests on a rule; make it your first lift test.
- Ask for a lift test on your biggest claim. In Meta, open the Experiments tool and start a Conversion Lift test for the campaign with the largest attributed result. In Google Ads, open the Lift studies tab under Campaigns, then Experiments, and select the plus button. Pass: the tool lets you set up the test. Fail: the campaign misses Meta's guide of $5,000 USD in spend and 500 conversions, so plan a regional holdout instead.
Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework
Sources, 1 October 2026: Get started with attribution (Analytics Help); Select attribution settings (Analytics Help); Key events attribution paths report (Analytics Help); Key event attribution models report (Analytics Help); About Conversion Lift (Google Ads Help); Understand your Conversion Lift based on users measurement data (Google Ads Help); Set up Conversion Lift based on users (Google Ads Help); About Conversion Lift (Meta Business Help Center); Differences between Conversion Lift test results and other reporting tools (Meta Business Help Center); About incremental attribution (Meta Business Help Center); Best practices to get started with Experiments (Meta Business Help Center)
Related answers
Frequently asked questions
Is data-driven attribution a kind of lift measurement?
No. GA4's data-driven model learns from paths and, Google says, from randomized trials. It still shares out the credit for each sale among the touches on its path. Nobody in your audience is held back. A lift test keeps a control group away from your ads and counts the gap.Can attribution and lift ever agree?
Yes, when the sales a campaign was credited with really were mostly extra. You only find that out by testing. Agreement for one campaign at one budget says little about the next, so test again when spend, creative or the season changes.Do I need a lift test if I already use multi-touch attribution?
Yes, if you want to know what the ads added. Multi-touch models spread credit across more touches, but every touch they score is one a buyer already had. None of them compares buyers with people who never saw the ads. Only a control group does that.
Go deeper: Causal attribution, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Keep reading
Terms in this article
- Attribution ModelAn Attribution Model defines how credit for conversions is assigned to marketing touchpoints. It dictates how marketing channels receive credit for sales.
- Attribution ReportAttribution Report shows which touchpoints or channels receive credit for a conversion. It identifies which campaigns drive desired actions.
- Causal AttributionCausal Attribution uses causal inference to determine which marketing touchpoints genuinely cause conversions, not just correlate with them.
- Control GroupControl Group is a segment of an audience intentionally not exposed to a marketing campaign, used to measure the campaign's true causal impact.
- CounterfactualCounterfactual is a hypothetical outcome that would have occurred if a subject had received a different treatment.
- Google AnalyticsGoogle Analytics is a web analytics service that tracks and reports website traffic.
- IncrementalityIncrementality measures the true causal impact of a marketing campaign. It quantifies the additional conversions or revenue directly from that activity.
- Incrementality TestingIncrementality Testing measures the additional impact of a marketing campaign. It compares exposed and control groups to determine causal effect.