Google opened Conversion Lift studies to Search and Performance Max campaigns, and eligible advertisers can now run them without going through a Google rep. This is the test that tells you whether your campaigns created conversions or just took credit for them. The catch is the eligibility thresholds, and most small accounts will not clear them.
What Changed
Conversion Lift has been around for a while inside Google Ads on Display, Video, Demand Gen, and App campaigns. Search Engine Land reported on October 7, 2026 that Search and Performance Max are now supported too.
The bigger change is access. Before this, advertisers running Search and Performance Max needed a Google representative to set a study up for them. Now eligible advertisers can start one themselves.
Keep in mind, Google's documentation notes that Conversion Lift is not available to every advertiser and account availability varies. Alpha and beta campaign types may still need a rep.
How A Lift Study Actually Works
A Conversion Lift study splits your audience. One group is held out and never sees the campaign you are measuring. The other group sees your ads normally. Google compares the two and estimates how many conversions the ads actually caused.
That is a different question than the one your conversion tracking answers every day. Standard attribution tells you which touchpoint to credit. It assumes the conversion was going to be credited to something. A lift study asks whether the conversion would have happened at all.
There is a tradeoff built into the design. A larger holdout gives you a better sample and a more reliable result. It also means withholding your ads from more people for the length of the study.
The Requirements You Have To Clear
- At least 1,000 observed conversions. Conversions that rely on supplementary data do not count toward this.
- A minimum campaign budget of $5,000. This applies to the campaigns participating in the study.
- At least one compatible conversion action. Not every conversion action you have set up will work with Conversion Lift.
Read those numbers carefully before you get excited. A thousand observed conversions is a real volume requirement. If you are running a local service business booking forty leads a month, this is not a tool you can use yet.
Google also recommends picking conversion actions that happen often enough to produce a statistically useful result. Measuring a conversion action that fires twice a week will not tell you anything, no matter how long you run it.
My Take On Who Should Run One
I think this is the most useful thing Google has opened up to self-service in a while, and I also think most accounts should not rush into it.
The reason is simple. A lift study costs you something. You are deliberately not advertising to a portion of your audience to learn something. That is a fair trade when you have a real question and enough volume to answer it. It is a waste when you are just curious.
The accounts that get the most out of this are the ones spending heavily on branded search, on remarketing, or on Performance Max, because those are the three places where the incrementality question is genuinely unanswered. If you have ever argued with somebody about whether your branded campaigns are buying conversions you would have gotten for free, this is how you settle it.
One of the biggest mistakes I see is running a test without deciding in advance what you will do with each outcome. Before you start, write down what you will change if lift comes back low and what you will change if it comes back high. If the answer to both is nothing, do not run the study.
What To Do With The Result
- Pick one campaign and one question. Do not test four things at once. Pick the campaign where you genuinely do not know the answer.
- Confirm you clear the thresholds first. Check observed conversions and budget before you build anything.
- Let it finish. Reading a lift study halfway through is how you end up making a decision on noise.
- Compare it against your reported numbers. The gap between what attribution credits and what the study measures is the actual finding. Track it alongside your other key metrics.
From there, you have something most advertisers never get. A number for how much of your reported performance is incremental, measured rather than assumed. That is worth more than another round of A/B testing on ad copy.
Question to Answer:
Which single campaign in your account would you genuinely struggle to defend if someone asked you to prove it created conversions rather than collected them?
Related Reading
In Summary
Conversion Lift is now self-service for Search and Performance Max if you clear 1,000 observed conversions, a $5,000 campaign budget, and have a compatible conversion action. That is the whole story, and the thresholds are the part that decides whether this is available to you.
If you qualify, pick your least defensible campaign and start there. If you do not qualify yet, the useful takeaway is knowing the question exists, because attribution has never been able to answer it.
0 comments