Facebook Custom Audience From a Customer List
A Facebook custom audience from a customer list is an audience built by uploading your own customer data - emails, phone numbers, names - which Meta hashes and matches against its user accounts. It is the only audience type built from data you own rather than data Meta collected, which makes it the most durable targeting asset in the account.
It is also the one most people upload once, use badly, and never refresh.
How does a customer list audience work?
You give Meta a file. Meta turns each row into an irreversible hash and compares it against hashes of its own users.
Nothing readable leaves your side if you use the standard upload tools - the browser hashes the values before they are sent. Rows that match become members of the audience. Rows that do not match are discarded.
You never see which individuals matched. You only see an approximate audience size and, for a while after upload, a match rate.
What should the customer list file contain?
A CSV with one identifier per column and a clear header row.
Meta accepts a long list of identifier types. In practice these carry the weight:
| Column | Format | Why it matters |
|---|---|---|
| Lowercase, trimmed | Highest single match rate | |
| Phone | Include country code | Catches people who signed up by phone |
| First / last name | Plain text | Weak alone, strong as a confirmer |
| City / state / country | Plain text | Disambiguates common names |
| Date of birth | YYYYMMDD | Rarely available, helps when it is |
Include every column you actually hold. Each extra identifier gives the matcher another way to confirm a person, so a file with email, phone and name will out-match an email-only file of the same size.
How do you upload the list?
The route is short, but the preparation matters more than the clicks.
- Clean the export. Strip test accounts, internal addresses, and rows that are obviously junk.
- In Ads Manager open Audiences, choose Create audience, then Custom audience, then Customer list.
- Upload the CSV and map each column to the identifier Meta thinks it is. Check this screen properly - a mis-mapped phone column silently kills the match rate.
- Add a value column if you have one, and tell Meta which column it is.
- Name the audience so the date and the segment are both visible, such as
Buyers-90d-2026-09.
Processing usually takes under an hour for a normal list. Very large files take longer.
Should you segment the list before uploading?
Yes, and this is where most of the value sits.
One giant "all customers" audience treats a one-time discount buyer and a repeat customer identically. Split the export instead:
- Recent buyers - last 30 or 60 days. Best for cross-sell, worst for prospecting.
- Lapsed buyers - bought once, nothing in six months. The natural win-back audience.
- High value - top quartile by lifetime spend. The best seed for a lookalike audience.
- Non-buyers - newsletter subscribers who never ordered. A different message entirely.
Each of those wants different creative. Uploading them separately is what lets you write that creative.
If you include a value column, Meta can also build a value-based lookalike, which weights the seed toward your better customers instead of treating every row the same. A value-based lookalike from high-value customers is consistently the strongest cold audience most accounts have access to.
What is the best use of a customer list audience?
Exclusion, more often than targeting.
Every prospecting campaign should exclude existing customers unless you are deliberately selling to them again. Otherwise you spend acquisition budget showing "discover us" creative to people who ordered last week.
Three practical setups:
- Exclude recent buyers from all prospecting. The simplest profit improvement in most accounts.
- Exclude all customers from a first-purchase offer, so the discount only reaches new people.
- Target lapsed buyers with a win-back campaign and a reason to return.
Setting up exclusions correctly is fiddly enough that it deserves its own walkthrough - see excluding a saved audience for the exact placement.
How often should you refresh the list?
Treat it as a scheduled job, not a one-off.
A customer list audience is a snapshot. It does not update itself when someone buys tomorrow. Left alone, the "recent buyers" exclusion slowly stops containing your recent buyers, and prospecting starts paying to reach them again.
Monthly is a sensible default for most stores. Weekly if volume is high or margins are thin.
If your ecommerce platform offers a direct integration, use it - a connected source refreshes on its own and removes the human step entirely.
Refreshing also matters for measuring. A stale exclusion inflates apparent new-customer acquisition, which then distorts your LTV to CAC ratio in the flattering direction.
Does a customer list replace competitor research?
No. It tells you about people who already bought, which says nothing about what is working for the brands competing for the same buyers.
For that, the Meta Ad Library shows every ad a competitor is currently running, free and without an account. Our Meta Ad Library downloader saves those ads in bulk with a searchable index, which makes it practical to study a rival's whole account rather than scrolling one ad at a time.
Use the customer list to decide who to exclude. Use competitor creative to decide what to say to everyone else.
FAQ
What is a good match rate for a Facebook customer list?
It varies with data quality and market, but a rate in the 50-70% range on an email-and-phone file is a reasonable target. Email-only files usually land lower. If you are well under half, check your column mapping and phone formatting first.
Is uploading a customer list to Facebook safe?
The identifiers are hashed in your browser before upload, so Meta does not receive readable emails or phone numbers through the standard tool. You are still responsible for having a lawful basis to use that data for advertising in your market - check your own obligations.
How many customers do I need for a custom audience?
Meta needs roughly a thousand matched people before the audience will deliver reliably. For a value-based lookalike seed, more is better - a few thousand high-value customers produces a noticeably better result than a few hundred.
Why did my customer list audience shrink?
Lists decay. People deactivate accounts, change emails, or opt out of ad targeting. An audience that was usable a year ago can fall below the delivery threshold, which is the main argument for refreshing on a schedule.
Can I target a customer list with retargeting creative?
Yes, and it usually works well because the audience already knows you. Existing retargeting ad examples apply directly, though a win-back message to a lapsed buyer should not read like an abandoned-cart reminder.
Should I exclude customers from lookalike campaigns too?
Yes. A lookalike is modelled on your customers, so it naturally includes people very like them - and often the customers themselves. Layering the customer list as an exclusion on top of the lookalike keeps the spend on genuinely new people.
The Klipio extension adds a download button to every ad in the Meta Ad Library: one click per ad, or bulk-save a whole search as a ZIP with a searchable swipe file inside. Free, no sign-up.
Get the free extensionKlipio reads a competitor's live Meta ads, ranks them by how long they have been running — the honest signal that an ad is profitable — and turns the winning angle into on-brand creative for your own brand. 3-day free trial.
Start free — 3-day trial →
