Every Q4, brands pour budget into digital advertising at the most expensive moment of the year. CPMs climb, competition for every impression sharpens and the pressure to prove return peaks. Yet many of those same brands are sitting on the one asset that would make that spend work harder: their own customer data.
Most brands don't have a first-party data problem. They have an activation problem. The lists exist. They just never make it into the DSP. With peak season having started and Black Friday only weeks away, there's still time to fix that. But not much.
Almost any first-party list can make a real difference to a programmatic campaign, as long as it matches at a high enough rate to use. Four get left on the shelf more than any others.
Used well, these lists do three jobs. They cut waste through suppression, bring previous buyers back through remarketing and model new audiences that look like your best customers.
There are three segments I’d activate first for peak.
First, recent purchasers as a suppression list. It's the quickest efficiency win available. At peak CPMs, every impression served to someone who has already converted is wasted. And refresh that list during peak, not once beforehand.
Second, lapsed and past purchasers for remarketing. They know you, and peak offers give them a reason to come back. This group usually delivers the best CPA of any audience.
Third, lookalikes built on high-value customers and YouTube engagers. When competition is fiercest, prospecting has to work harder. Seeding models with your best customers points spend at the people most likely to behave like them.
First-party data isn't a switch you flip in November. Its job changes as peak unfolds.
Before peak, prospect. Run lookalikes while CPMs are still lower, so a warm pool is ready for the big weeks.
During peak, retarget and suppress. Hit site visitors and past purchasers with your offers, and exclude anyone who has just converted. Refresh suppression lists as often as your setup allows, ideally daily through Black Friday week.
After peak, retain. Don't let new customers become one-off buyers. Re-engage them with cross-sell and loyalty messaging, then add them to your CRM base and lookalike seeds for the new year. January's prospecting starts with December's buyers.
The biggest blocker is usually getting the data into the DSP at all. Plenty of brands have strong lists sitting in a CRM or email platform with no route into programmatic. Check now that your data can be exported or connected, whether through a direct upload, an existing integration or an onboarding partner. Then run a test upload well before peak, so problems surface while there's still time to solve them.
The second blocker is match rate. Expect a big drop between your list size and the audience you can actually reach, because only a portion of any list matches to an identifiable user. Plan reach and budget around the matched audience, not the headline CRM number. Where you have consent, add more identifiers, such as phone number or postcode alongside email, to lift the match. Then use the list as a lookalike seed to rebuild scale.
Realistically, allow two to four weeks. Hashed emails uploaded directly to a DSP can be live within days. Routing through an onboarding partner, or connecting a CRM to a clean room, can take several weeks once legal signoff, data formatting and match testing are factored in.
So the work that decides your peak performance needs to happen this week:
Peak seasons reward the brands that are prepared. The data is already yours. The only question is whether it'll be switched on in time.
Most brands don't have a first-party data problem. They have an activation problem. The lists exist. They just never make it into the DSP. With peak season having started and Black Friday only weeks away, there's still time to fix that. But not much.
The data you already have
Almost any first-party list can make a real difference to a programmatic campaign, as long as it matches at a high enough rate to use. Four get left on the shelf more than any others.
- Past purchasers: Rarely used for suppression, so brands keep paying to reach people who have just bought
- Lapsed customers: People who know the brand but haven't bought in six to 12 months. Often the cheapest group to win back
- CRM and newsletter sign-ups: Collected diligently, then not used for remarketing or as seeds for lookalike modelling
- YouTube engagers: People who have watched or subscribed through a brand's organic or paid YouTube activity. A warm prospect pool that's easy to overlook
Used well, these lists do three jobs. They cut waste through suppression, bring previous buyers back through remarketing and model new audiences that look like your best customers.
Where to start
There are three segments I’d activate first for peak.
First, recent purchasers as a suppression list. It's the quickest efficiency win available. At peak CPMs, every impression served to someone who has already converted is wasted. And refresh that list during peak, not once beforehand.
Second, lapsed and past purchasers for remarketing. They know you, and peak offers give them a reason to come back. This group usually delivers the best CPA of any audience.
Third, lookalikes built on high-value customers and YouTube engagers. When competition is fiercest, prospecting has to work harder. Seeding models with your best customers points spend at the people most likely to behave like them.
Before, during and after
First-party data isn't a switch you flip in November. Its job changes as peak unfolds.
Before peak, prospect. Run lookalikes while CPMs are still lower, so a warm pool is ready for the big weeks.
During peak, retarget and suppress. Hit site visitors and past purchasers with your offers, and exclude anyone who has just converted. Refresh suppression lists as often as your setup allows, ideally daily through Black Friday week.
After peak, retain. Don't let new customers become one-off buyers. Re-engage them with cross-sell and loyalty messaging, then add them to your CRM base and lookalike seeds for the new year. January's prospecting starts with December's buyers.
What gets in the way
The biggest blocker is usually getting the data into the DSP at all. Plenty of brands have strong lists sitting in a CRM or email platform with no route into programmatic. Check now that your data can be exported or connected, whether through a direct upload, an existing integration or an onboarding partner. Then run a test upload well before peak, so problems surface while there's still time to solve them.
The second blocker is match rate. Expect a big drop between your list size and the audience you can actually reach, because only a portion of any list matches to an identifiable user. Plan reach and budget around the matched audience, not the headline CRM number. Where you have consent, add more identifiers, such as phone number or postcode alongside email, to lift the match. Then use the list as a lookalike seed to rebuild scale.
The clock is running
Realistically, allow two to four weeks. Hashed emails uploaded directly to a DSP can be live within days. Routing through an onboarding partner, or connecting a CRM to a clean room, can take several weeks once legal signoff, data formatting and match testing are factored in.
So the work that decides your peak performance needs to happen this week:
- Pull and clean your lists: current, consented, hashed emails
- Confirm legal and consent sign-off for advertising use
- Upload a test list and check the match rate. If it's low, there's still time to fix it
- Agree a refresh schedule, so suppression lists stay current through peak
Peak seasons reward the brands that are prepared. The data is already yours. The only question is whether it'll be switched on in time.
Written by
Related Insights
The GTA effect: What's your move?
You’ve got AI budget. Now what?
Is your brand underestimating TikTok?
Your brand is visible in AI. But is it being recommended?
Is your brand showing up in ChatGPT? Here's why it matters
