Why is our Meta custom audience not working or too small to use, and how do we fix it?
Our answer
Usually too few people matched or the list went stale. Read the audience's status in Meta and upload at least 1,000 customers with email, phone and name. Website audiences update themselves, but uploaded lists need an automatic sync.
- Confidence
- Medium
- Strongest proof
- 6 Meta rules
- 4 practitioners, 3 opinions or claims
- Sources
- 7 sources
- 13 statements
- M
KRM
New here? Take the 1-minute tour
Why this is the answer
4 reasons and our judgmentMeta tells you what's wrong. Each audience carries a status: too small to run, low match rate, invalid entries, pixel not firing, still populating, or out of date after 30 days unused (Meta). Audiences named for a health condition or finances get blocked (Meta).
A list reaches only the people Meta can match. More identifiers per customer give a better match rate, and Meta asks for lists of at least 1,000 customers (Meta). Most email-only lists match 40 to 60%, claims Rokt mParticle, a seller of match enrichment.
Uploaded lists don't update on their own. Meta refreshes website audiences automatically, while a customer list keeps the people you sent until you remove or replace them (Meta). Klaviyo's sync pushes list changes to Meta continuously, with a 24 to 48 hour lag.
Your purchase events already mark many buyers. With Acquire new customers selected, Meta excludes people who fired a purchase event, so a list matters most for buyers your events miss, such as offline ones (Jon Loomer). Meta's new-customer setting covers when to use that setting.
Our judgment
Low matching and stale lists are the likeliest causes, given how far below the list size match rates run and how soon Meta flags a list that hasn’t been refreshed. When an audience fails, read its status in Meta first and fix the cause it names, and treat an uploaded list as a snapshot unless an integration syncs it. Where you exclude buyers (Excluding existing customers from prospecting says which), combine a synced customer list with website audiences and Meta’s purchase-based exclusion, since none of them covers every customer alone.
Our judgment, built from Meta, Klaviyo, Ads Uploader, Jon Loomer and Rokt mParticle: our reading of these sources, not a finding from any one of them.
What the sources say
13 statements from 7 sourcesMeta says a customer list should include at least 1,000 customers and one main identifier. A list not updated in 30 days may hurt delivery, and phone numbers need a country code.
“Match score shows a score out of ten for how well identifiers in your customer list have matched profiles on Meta platforms in your last successful upload.”
Meta says customer list audiences are matched by hashing identifiers like email and phone, and more identifiers give a better match rate, especially to supplement mobile advertiser IDs after iOS 14.
“The more information you can provide, the better the match rate (which means our ability to make the matches).”
Klaviyo says its Meta audience sync needs a list of over 100 profiles, matches by email so sizes differ, and custom audiences can take 24 to 48 hours to update in Meta.
“Meta requires a minimum of 100 profiles to create a custom audience.”
GrowthSpree, a B2B SaaS agency, says Meta removed detailed-targeting exclusions, so exclude existing customers, recent converters and unqualified segments with custom audiences, and enrich CRM lists with personal emails.
“work emails match poorly to personal Meta accounts, so enrich CRM lists with personal emails.”
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)”
Jon Loomer says accurate, complete purchase event data matters more than customer list audiences for Meta's automatic customer exclusions, but add a list if some sales happen offline.
“Sending accurate and complete event data is more important than segmenting your customers by custom audience.”
Ads Uploader says customer list match rates commonly run 20% to 70%, so exclusions leak, and suggests refreshing lists, stacking identifiers and layering audience types. It shows no data.
“Match rates on customer-list audiences commonly land somewhere between 20% and 70%, never 100%.”
Ads Uploader, which sells an upload tool, says custom audience exclusions still work in 2026, including inside Advantage+ Audience, where they act as hard constraints. No data.
“Custom audience exclusions still work in 2026, including inside Advantage+ Audience, where they act as hard constraints rather than suggestions”
Rokt mParticle, a match-rate vendor, says in a sponsored article that most brands' email-only lists match 40 to 60% on Meta, with no method shown.
“Suppression lists only suppress the customers a platform recognizes.”
Meta's developer docs say website custom audiences keep people for 1 to 180 days, set by the retention you choose, and update automatically.
“Number of days to retain someone in audience. Between 1 and 180 days.”
Meta says custom audience changes take up to 24 hours, audiences suggesting health conditions or financial status are blocked, and customer lists unused for two years get flagged for deletion.
“You won’t be able to use flagged custom audiences when creating new campaigns.”
Common Thread Collective says Meta extended Purchase custom audience retention from 180 to 730 days in May 2026, and suggests recency segments with a dedicated win-back sequence for 180 to 730 days. No performance data.
“Build layered audiences: buyers from the last 30 days, 30-90 days, 90-180 days, and 180-730 days.”
Meta's Marketing API lists custom audience status codes, such as low match rate, pixel not firing and out-of-date audiences, and says audiences unused for over 2 years begin to expire.
“If an audience hasn't been used in an active ad set for over 2 years, it will begin to expire.”
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