How should our ads for winning new customers differ from our ads for retaining or reactivating existing customers?
Our answer
Run acquisition and retention as separate campaigns with separate budgets. Judge acquisition on cost per new customer, and keep paid retention small and lift-tested. Low-retention brands and frequent launchers can combine them on manual bids (Andrew Faris).
- Confidence
- Medium
- Strongest proof
- 3 controlled tests
- 2 Meta rules, 2 datasets, 13 practitioners, 5 opinions or claims
- Sources
- 10 sources
- 25 statements
M
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentWithout separate retention campaigns, Meta's delivery leaks budget to past customers. Broad delivery spends about 20 to 25% of budget on past visitors, engagers and customers, says Jon Loomer, a Meta ads educator. Meta made Customer Lifecycle Strategy available to all advertisers in October 2026.
Brand type decides whether to split campaigns. Common Thread Collective, an ecommerce ad agency, runs separate evergreen acquisition and retention campaigns. Andrew Faris, an ecommerce growth consultant, combines new and existing customers for low-retention brands and frequent launchers, and separates them for high-retention single-product brands.
A separate lapsed-customer campaign added sales in one holdout. In Phil Kiel's Haus test on one UK brand, it made 1.93 times incremental ROAS on returning-customer sales against a 1.5 times bar. On all store sales it measured 2.45 times, against Meta's own 6.2 times.
Our judgment
- Keep acquisition and retention apart, except for low-retention brands and frequent launchers, where Faris combines them. This follows Common Thread Collective's practice for 7, 8 and 9-figure brands over Loomer's view that broad delivery already reaches past buyers. Keeping the retention side small and lift-tested is where they meet.
- Keep paid retention small and reach past buyers first by email and SMS. Common Thread Collective and Portland Leather Goods both reach existing customers this way before a sale. That practice covers sale moments only. We extend it to retention generally.
- Keep retention ads only where a lift test shows they add sales. This follows Faris, who spends on existing customers and lift-tests it, and Kiel's holdout. Kiel's test covers one UK brand that has subscribers and also sells on Amazon, so applying it to other models is an extension.
- Judge acquisition on cost per new customer. Common Thread Collective says converting existing customers does not grow your brand and tells brands to track their new-customer ratio after switching the setting on. Wicked Reports and Northbeam both report new-customer cost apart from blended cost.
- When testing Meta's Acquire new customers setting (Meta's new-customer setting), cut its 730-day automatic exclusion to 180 days or shorter, and keep it if the test wins. That draws on the 180-day new-customer definition Common Thread Collective reported at launch, the buyers of about the last 90 days that Faris excludes from retention campaigns and the shorter windows Meta offers as Loomer lists them (Excluding existing customers from prospecting).
Our judgment, built from Common Thread Collective, Jon Loomer, Andrew Faris, Phil Kiel, Wicked Reports, Northbeam and Portland Leather Goods: our reading of these sources, not a finding from any one of them.
What the sources say
25 statements from 10 sourcesTony Chopp of Common Thread Collective suggests weighting the early November forecast toward returning customers reached through your own email and SMS.
“I think it's a case in point for a heavy up on your forecast early in the month of November from your returning customer cohort, specifically from your own channels on email and SMS.”
Meta has removed the existing customer budget cap from Advantage+ sales campaigns, but says you can do the same by hand with an audience exclusion or two ad sets with a spend limit.
“Existing customer budget cap, which allowed Advantage+ sales campaign advertisers to choose whether or not to spend on existing customers, is no longer available.”
In one UK brand's Haus geo test, a Meta campaign to lapsed customers returned 1.93 incremental ROAS on returning-customer sales against Meta's reported 6.2x; Haus counted 40% of the orders Meta claimed.
“a retention campaign showing 6x in Ads Manager is not a 6x campaign.”
Phil Kiel says a Meta retention test aimed only at lapsed customers (no purchase in 60 days) still lifted new customer sales 3% in the test regions, against a Haus holdout.
“New customer sales still rose 3% in the test regions.”
Jon Loomer says Meta's "Acquire new customers" setting now automatically excludes purchasers in your conversion data from the last 730 days, and this can't be turned off. Offline buyers still need a custom audience.
“Customers from your conversion data (730 days)”
CTC says for Black Friday 2026 to turn on new customers only in every prospecting ad set, after checking pixel and Conversions API coverage and uploading about two years of buyers, with a separate retention campaign.
“Without lifecycle controls, Meta's optimization engine defaults to finding the easiest conversions.”
CTC says Meta's Customer Lifecycle Strategy toggle is live in manual Sales campaigns and can limit delivery to new customers (no purchase in 180 days); Advantage+ campaigns used a separate budget setting.
CTC says Meta's Customer Lifecycle Strategy, open to all advertisers from Oct 7, 2026, lets manual Sales ad sets target only new customers, defined at launch as no purchase in the last 180 days.
Meta says Customer Lifecycle Strategy is open to all advertisers: mark a campaign as acquisition and Meta suggests which existing customers to exclude. It is testing one tag that excludes them all.
“already available to all advertisers”
Portland Leather Goods' CMO plans a 55-day holiday season from November 1, reaching existing customers by SMS and email rather than Meta, and turns ads on for the Nov 22 to 28 sitewide sale.
“We now plan a 55-day selling season. November 1st all the way through Christmas.”
Klaviyo's recap of Tatcha and Far Niente marketers: give subscribers or VIPs early Black Friday access with a perk like a gift or free shipping instead of a bigger discount. No figures given.
“Tatcha's early-access-only merch item gave shoppers a reason to buy right away, and the brand didn't need to add a discount.”
Across hundreds of Wicked Reports ecommerce accounts in one August 2026 week, 87% of Meta's click-tracked buyers were new (blended CAC $77 vs $88 per new buyer), against 66% on Google and 39% on YouTube.
“verified at the order level against first party order IDs”
Northbeam, grading its own Apex product, reports A/B tests against Meta's default controls: +12% new-customer ROAS and -18% new-customer cost per order. No sample, duration or significance is given.
“This chart shows the aggregate improvements from A/B testing Apex campaigns against Meta’s default controls.”
In a Northbeam case study, Ka'Chava had Meta optimize to first-time-customer revenue against standard attribution: new-customer ROAS rose 43.2%, CAC fell 25.2%. Spend, length and significance not given.
Andrew Faris says not to blanket-exclude existing customers, since many brands should spend on them and lift-test it. Stronger exclusions, like active subscribers and 90-day buyers, belong on retention campaigns.
Andrew Faris says that for brands with low natural retention, his default is not to separate retention campaigns from prospecting campaigns.
“We do not separate out retention campaigns from prospecting campaigns.”
Andrew Faris says brands with frequent product releases or sales usually shouldn't split new and existing customers into separate campaigns.
“you actually shouldn't even separate out new and existing customers into separate campaigns.”
Andrew Faris says single-product supplement brands with high repeat rates should not spend on existing customers at the same level as on new ones.
“you shouldn't be targeting existing customers with your ad spend at the same level that you're doing new customers.”
Andrew Faris says manual bids reduce the risk of overspending on existing customers in mixed-audience ad sets, as long as you give Meta a target across the board.
“And so if you're running manual bids, it is no problem to include existing”
Andrew Faris feels less need for tight existing-customer exclusions under Meta's Incremental Attribution, and has merged new and existing customers for some wide-catalog clients.
CTC runs evergreen acquisition and retention campaigns, time-bound campaigns for marketing moments, and a recovery campaign for ads that spent less than 3 times the CPA target.
“Populate it with ads from evergreen where spend is less than 3x the CPA target.”
Tony Chopp describes agency practice with no data: run evergreen campaigns split between acquisition and retention, and use broad targeting with exclusions.
“must be correctly configured at the account level”
Jon Loomer says broad or Advantage+ delivery already spends roughly 20% to 25% of budget on past visitors, engagers and customers, so a separate retargeting campaign reaches them twice.
“just reaches them twice”
CTC reports that on May 18, 2026 Meta extended website purchase audiences from 180 to 730 days, a two-year window for retargeting past buyers. Other events keep their prior limits.
“The maximum retention window for purchase-based custom audiences has been extended from 180 days to 730 days.”
CTC's Connor McLaughlin says the agency runs separate prospecting, retention and remarketing campaigns on Meta, and on AppLovin adds about 10% of spend in a prospecting campaign. No Meta data.
“on Google we have brand non-brand, meta we have prospecting, retention, remarketing”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer