Short answer: A lift test shows how much of your spend is incremental. The changes come after. Exclude warm audiences (past buyers, site visitors, engagers) as well as Meta allows, move budget toward ads that reach new people at low frequency, cut or shrink campaigns that mostly re-reach buyers, reset your targets using cost per incremental conversion, and retest to confirm. The test is the diagnosis. The media buyer is the treatment.
Why the test alone isn't enough
Most brands that run a lift test do one of two things with the result: nothing, or a blunt budget cut. Neither is the point. A low lift result tells you your ads are spending too much on people who would have bought anyway. The job is to find where that's happening and redirect it.
Step 1: Find where the waste is
Before changing anything, look at who your spend reaches:
- New vs returning delivery: use the Audience segments breakdown to see how much spend goes to existing customers and engaged audiences. See new vs returning delivery.
- Net new reach: is the account still finding new people, or recycling? See the Meta reach report guide.
- Frequency by ad: high daily frequency ads are usually talking to people who already know you. See daily frequency as a funnel signal.
- Campaign by campaign: retargeting and "warm" campaigns are the usual suspects. See is remarketing incremental.
Step 2: Exclude warm audiences as well as you can
The most direct fix. In prospecting and scaling campaigns, exclude:
- Past purchasers (a customer list upload, refreshed regularly)
- Recent site visitors and add to carts
- Email and SMS subscribers
- Instagram and Facebook engagers, where it makes sense
Meta's automated campaigns don't always respect every exclusion, and some placements leak. So exclude to the best of your ability, then check the Audience segments breakdown to see whether warm delivery actually dropped.
Step 3: Favor ads with low frequency
Ads that keep a low daily frequency are usually reaching new people. Ads with high frequency are usually showing the same people your ad again and again, often people who were already going to buy. Shift budget toward the low frequency ads that still convert, and let the high frequency ones run off.
A personal view, not a proven rule: in the accounts we run, we lean toward video and creator content over static ads for incremental reach. We haven't proven that statics are less incremental, and it varies by account. If you want to know, test it with a multi cell lift study. See conversion lift vs brand lift, and single vs multi cell.
Step 4: Cut or shrink what isn't incremental
If a lift test on a retargeting or warm campaign shows little or no lift, cut its budget hard or turn it off. If the whole account shows low lift at your current spend, lower total spend until cost per incremental conversion is back under your target. For big, tapped out accounts, this is where the 25% to 30% savings usually comes from. See why big accounts waste 25%.
Step 5: Reset your targets
Stop managing to Ads Manager CPA alone. Use the incrementality factor from your test to translate reported results into incremental ones:
True CPA estimate = Ads Manager CPA ÷ incrementality factor
If Ads Manager shows a $30 CPA and your factor is 0.5, your true cost per incremental purchase is about $60. Set budgets on that.
Step 6: Retest
Changes need proof. Run another lift test after the changes have had a few weeks to settle. If lift went up and cost per incremental conversion went down, the changes worked. Then keep testing every quarter, because incrementality moves as you scale.
Key takeaways
- The lift test is the diagnosis; the account changes are the treatment.
- Find the waste first: new vs returning delivery, net new reach, frequency.
- Exclude warm audiences as well as Meta allows, then verify.
- Favor low frequency ads that reach new people.
- Reset targets with the incrementality factor, then retest.
FAQ
Should I turn off retargeting completely? Only if a lift test on it shows no meaningful lift. Some brands keep a small, capped retargeting budget for reminders.
Will exclusions hurt my reported ROAS? Often, yes. Reported ROAS falls because you stop getting credit for sales that would have happened anyway. Incremental results are what matter.
How long until I can retest? Give changes a few weeks to settle, then retest.
Do Advantage+ campaigns respect exclusions? Not always fully. Use the customer settings Meta provides and check the Audience segments breakdown to see what actually happens.
Want a media buyer to act on your lift results? We do a free Meta audit: a full account audit plus net new reach and cumulative reach reports. Reach out on our contact page.
Sources
- Meta, "Measure What Matters with Incrementality" practitioner deck
- Seer Interactive, "We tested Meta's new Incremental Attribution setting on $1M in ad spend"
- Haus, "Understanding Meta incrementality testing."
Check what you learned
Tap an answer to see if you got it right.
1. What is the lift test, compared with the account changes after it?
2. Where do you look first for the waste?
3. What do you do after making changes?
Ready for the full test? Take the Measurement exam and get certified. Want this run on your account? Get my free audit.
Module 3: Measurement · article 11 of 15
Previous: Conversion Lift vs Brand Lift on Meta, Plus Single Cell vs Multi Cell Tests
Next: Meta Incremental Attribution: What the Setting Does and Whether to Trust It
About the author
Jake Mehani, founder of Creative Analytics
Jake has spent 15 years running Meta and paid social ads for consumer brands. Meta published his RBX Active Advantage+ case study, he taught the DMAnc Meta and TikTok ads masterclass, and he spoke at Motion's 2024 Creative Strategy Summit. His senior team runs media buying, creative and measurement like an in house team.