Often, much less than it looks. Remarketing shows ads to people who already visited your site or engaged with your brand, and many of them were going to buy anyway, so attribution gives the ads credit for sales they did not cause. To find out whether yours is incremental, compare its results across attribution windows, check how much of your budget goes to warm audiences, look at frequency, and confirm with a Conversion Lift test.
Remarketing feels like the safest money in an ad account. The audience already knows you. The ROAS is high. The cost per purchase is low. It is the campaign everyone points to when they want to show the account is working.
That is the problem. The same things that make remarketing look great on a dashboard are the things that make it hard to trust.
What does incremental mean for remarketing?
A remarketing sale is incremental only if it would not have happened without the ad. If the shopper was coming back to buy regardless, the ad did not create that sale; it just stood next to it.
Think about who is in a remarketing audience: people who visited your site recently, added something to cart, watched your videos or follow your page. Many of them are already close to buying. They will come back through an email, a Google search for your brand name, a bookmark or a reminder from a friend.
When those people also see a remarketing ad, attribution gives the ad credit. Over time, the campaign reports a beautiful ROAS built partly on sales that were going to happen anyway. You are paying for a sale you already had.
Why does remarketing always report such high ROAS?
Because attribution rewards being near a purchase, and remarketing is designed to be near purchases. It targets the people with the highest chance of buying, so it collects the most credit, whether or not it changed anyone's mind.
This is not a flaw in remarketing specifically. It is how attribution works. Attribution answers "was this ad in the path?" not "did this ad cause the sale?" Prospecting ads have to earn credit from people who did not know you. Remarketing ads collect credit from people who were already walking toward the register.
That is why we say remarketing ROAS is often the most inflated number in an account. Not because anyone is lying, but because the measurement is tilted in its favor. For a full breakdown of how attribution windows hand out credit, see 1 day click, 7 day click and view.
How can you tell if your remarketing is incremental?
You cannot tell from a single ROAS number. You can get a strong read by looking at several signals together, and a definitive one with a lift test. Here is the order we work in.
1. Compare attribution windows
Pull remarketing results on 1 day click, 7 day click and Meta's incremental attribution, side by side. Then pull view through on its own.
What to look for: if remarketing looks excellent on 7 day click and views but much weaker on 1 day click and incremental attribution, a lot of its credit is coming from the generous rules. That is a strong hint that many of those buyers were on their way anyway.
2. Check how much of your budget goes to warm audiences
Look at the share of spend going to remarketing, and more importantly the share of delivery actually reaching people who already know you. Campaign names can mislead. A campaign called "prospecting" can still spend heavily on warm audiences.
Where Meta offers its audience segment breakdown (new audience, engaged audience and existing customers), use it. It shows how delivery splits between people new to you and people who already know you. A healthy, growing account spends most of its money reaching new people. When warm delivery takes a large share, it is usually a setup problem worth fixing. Some agencies lean on remarketing because it makes reported results look strong, so a heavy warm share should always prompt a closer look.
3. Look at frequency
Frequency is how many times, on average, each person saw your ads. Remarketing audiences are small, so frequency climbs fast.
What to look for: high frequency on a small warm audience means you are showing the same people the same ads over and over. Past a point, more impressions are not changing anyone's decision. They are just collecting credit when the person buys.
4. Watch net new reach
Net new reach measures how many people saw your ads for the first time in a period. When remarketing takes a growing slice of budget, net new reach often flattens. That is the account trading growth for easy credit.
5. Run a Conversion Lift test
This is the only way to measure incrementality directly. A Conversion Lift test holds back a random group of your remarketing audience from seeing ads and compares their purchases to the group that did see them. The difference is what remarketing actually caused.
If you only do one thing from this list, do this. Lift tests on remarketing are some of the most eye opening results we see, because the gap between reported and real impact is often largest there.
Is a dedicated remarketing campaign a red flag?
It is a flag to investigate, not an automatic condemnation. When we audit an account and find a separate remarketing campaign holding a meaningful budget, we dig in before drawing any conclusion.
Here is why it gets a closer look. In the structure we use, there are two main campaigns. Creative testing runs to cold audiences, with past buyers and recent visitors excluded, so we can see which ads truly win new people (we explain why in why creative testing must run to cold audiences). The scaling campaign excludes past customers but deliberately leaves warm non-buyers in the pool. That means the scaling campaign already reaches warm visitors, alongside cold audiences, and Meta's budget system decides how to split the money based on where results come from.
Carving a separate, fixed budget out for remarketing does two things. It pays for the least incremental conversions in the account, and it takes budget decisions away from the system that would otherwise allocate it more efficiently. So the absence of a dedicated remarketing campaign is not a problem in our book. Its presence is a question.
When one exists, we look at its share of spend, its frequency, how far back its audience windows reach, and what it shows across attribution windows. Then we recommend a lift test.
Should you ever run remarketing?
Sometimes, deliberately, with the right measurement. There are advanced setups where a carefully built remarketing play makes sense, usually for brands that are serious about measuring incrementality.
If you run one, a few principles help:
- Keep audience windows tight. Recent site visitors and recent cart activity are closer to the decision than people who engaged with you many months ago. Very long engagement windows overlap heavily with cold audiences and blur everything.
- Cap frequency. Showing a warm audience the same ad many times a week rarely adds value.
- Do not judge it on ROAS. By design, you are targeting people already in the buying window, so reported ROAS will look good no matter what. It is a false positive by definition.
- Measure it with a lift test. That is the only fair scoreboard.
Can exclusions stop warm audiences from leaking into prospecting?
Partly. Exclusions help, but they are never perfect, so some warm delivery will always slip through.
Privacy changes mean Meta cannot match every visitor or customer to a profile at the moment it decides who sees an ad. People who opted out of tracking, browse without being logged in or simply were not matched become invisible to your exclusion lists. Some of them will see your prospecting ads anyway.
Two practical lessons follow. First, never assume exclusions guarantee zero warm waste; they reduce it. Second, size any leftover warm delivery in dollars, not just percent. A small percentage can be noise on a small account and a large annual cost on a big one. At high spend, reach data, not exclusion settings, is the real referee of whether you are buying new people.
What should you do if remarketing turns out not to be incremental?
Move the money to where it creates new demand. Low incrementality in remarketing is a finding, not a failure, and it usually points to a better use of budget.
Typical steps:
- Fold remarketing budget into a broader scaling campaign that can reach both new and warm non-buyers, and let the budget system allocate.
- Invest in creative that wins new people. The fastest way to grow is ads that reach and convert people who have never heard of you.
- Tighten prospecting exclusions so creative tests stay clean.
- Re-measure. Watch net new reach and run another lift test after the change.
Many brands find that cutting back remarketing barely dents total sales, because those buyers were coming anyway. Meanwhile, the budget moved to prospecting starts finding customers who would not have shown up on their own.
Key takeaways
- A remarketing sale is incremental only if it would not have happened without the ad; many remarketing buyers were already on their way.
- Remarketing usually reports the highest ROAS in an account because attribution rewards being near a purchase.
- Compare 1 day click, 7 day click, view through and incremental attribution to see how much remarketing credit depends on generous rules.
- Check warm audience share of delivery, frequency and net new reach for signs that budget is recycling the same people.
- A dedicated remarketing campaign is a reason to investigate, not an automatic problem.
- A Conversion Lift test is the only way to measure remarketing incrementality directly.
FAQ
Is remarketing a waste of money?Not always, but it is often less valuable than it looks. Some remarketing does nudge people who would not have come back. The only way to know how much is to test it with a holdout group.
Why does my remarketing campaign have better ROAS than prospecting?Because it targets people who were already likely to buy. Attribution gives it credit for those purchases even when the ad did not change the outcome. Prospecting has to win over people who did not know you, which is harder to get credit for.
How much of my budget should go to warm audiences?There is no universal number, because it depends on your account and how your campaigns are built. As a principle, a growing account should spend most of its money reaching new people. If warm delivery takes a large share, look closely at why.
Can I just exclude past customers and visitors from everything?You can exclude them from prospecting and creative testing, and you should. But exclusions never catch everyone because of privacy limits, and fully cutting off warm non-buyers from scaling can hurt efficiency. Judge your structure by its results: incremental sales and reach into new audiences.
What is the fastest way to check remarketing incrementality?Compare its results on 1 day click and incremental attribution against 7 day click and views. A big gap is a quick warning. Then confirm with a lift test.
Find out what your remarketing is really worth
If remarketing is carrying your reported results, it is worth knowing how much of that is real. Our free Meta account audit includes a full account audit plus net new reach and cumulative reach reports, and we will show you how much of your budget is reaching people who already know you. Request yours here.
Keep reading
More on measurement:
1 Day Click vs 7 Day Click vs View Through: Which Meta Attribution Window Should You Trust?
What Is a Meta Conversion Lift Test and When Should You Run One?
What Is the CTSG System? Creative Testing, Scaling and Graveyard for Meta Ads
See every article on measurement
Ready for your free Meta audit?
We will review your account and run net new reach and cumulative reach reports, free.
Check what you learned
Tap an answer to see if you got it right.
1. When is a remarketing sale incremental?
2. Why does remarketing usually show the highest ROAS?
3. What is the only way to measure remarketing incrementality directly?
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 4 of 15
Next: Remarketing Is a Reminder, Not a Sale: Reach vs Conversion Objective on Meta
Previous: What Is a Meta Conversion Lift Test and When Should You Run One?
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.