---
title: "Create a Facebook Custom Audience From Your Customer List"
description: "Create a Facebook custom audience from your customer list: prepare the file, upload and map columns in Ads Manager, handle opt-outs, and choose how to use it."
url: "https://webleveling.com/blog/facebook-custom-audience-from-customer-list/"
lang: "en"
published: "2026-10-02"
modified: "2026-10-02"
author: "Cal Hewitt"
tags: ["facebook-ads","social-media-marketing","small-business"]
image: "https://webleveling.com/og/blog-facebook-custom-audience-from-customer-list.jpg"
alternates:
  es: "https://webleveling.com/es/blog/audiencia-personalizada-de-facebook-desde-lista-de-clientes.md"
---

# Create a Facebook Custom Audience From Your Customer List

Create a Facebook custom audience from your customer list: prepare the file, upload and map columns in Ads Manager, handle opt-outs, and choose how to use it.

![A closed laptop beside a stack of blank index cards, a pencil and a sealed envelope on a worn wooden desk.](https://webleveling.com/blog/facebook-custom-audience-from-customer-list/customer-list-featured.jpg)

You typed "facebook ads create custom audience" into Google because you have a spreadsheet of past customers and a feeling that it should be worth something. Maybe you want to send a repeat-purchase offer to people who already bought. Maybe you want to stop paying to show a new-customer ad to people who are already customers. Either way, the list is sitting in a CRM export or a booking system, and Ads Manager looks like it was built for someone else. You also have a quiet worry: will the file get rejected, will only a sliver of it match, and is it even allowed to upload these people at all? The upload itself is a short job that an owner can do alone in one sitting. The parts that deserve your time are the cleanup before it and the permission questions around it. The steps below follow the order you will do them, and the interface names are the current ones, which Meta can change.

**Key Takeaways**

**Do the upload yourself first**

A clean export, a few minutes of spreadsheet cleanup and one pass through Ads Manager is enough for an ordinary campaign. Paid help matters when the data or the rules are messy.

**Clean the file before you upload**

Lowercase emails with no spaces, phone numbers in international format, one person per row, and no opt-outs. Upload trouble starts in the spreadsheet.

**Not every row will match**

Match rates depend on your data and on who uses Meta, and Meta's terms say it does not reveal who matched. Read the audience status in your own account instead of trusting a promised percentage.

**Permission comes first, hashing second**

Meta's terms require you to have the rights, permissions and a lawful basis to use the data. Hashing does not replace any of that.

**Pick one clear use per audience**

Send a retention offer, exclude customers from a new-customer campaign, or use the list as a lookalike seed, and name the audience for that purpose and date.

## You can build a customer list audience in one sitting

A customer list audience is a group you build in Meta Ads Manager from contact details you already own. Meta matches those details against its own accounts so your ads can reach, or skip, the people on your list. It is different from a website audience, which is built from visitors and needs a tracking pixel. If you want that one instead, the post on [retargeting website visitors with Facebook ads](https://webleveling.com/blog/retarget-website-visitors-with-facebook-ads/) covers it. If your list is for Google rather than Meta, see how to [use your customer list in Google Ads](https://webleveling.com/blog/use-your-customer-list-in-google-ads/).

The whole job runs in four moves: export, clean, upload, then wait for processing. Block out an hour, and have your CRM or spreadsheet open next to Ads Manager. Most of the time goes into the first two moves, not the third.

![A closed laptop beside a neat stack of blank index cards and a pencil on a worn wooden desk.](https://webleveling.com/blog/facebook-custom-audience-from-customer-list/customer-list-ready-to-sort.jpg)

The upload is the quick part. Sorting the list before it is where the hour goes.

## Your export should hold only what Meta needs to match

Start in the system where your customers live: your CRM, booking software, store platform or a plain spreadsheet. Export one row per person, and keep only the identifiers that help with matching. Email addresses and phone numbers are the main ones. Meta's upload flow also accepts fields such as first name, surname, city, state, ZIP and country, which you can map in a later step. Leave out anything you do not need to match, and do not export sensitive fields.

Before you go further, count three things in a notes file: total rows, rows with blank or duplicate identifiers, and rows where the person agreed to the marketing use you have in mind. Keep a note of the source system and the export date, and write down what the audience is for. That note will matter if anyone later asks why a person was in the audience.

Save the file as CSV or TXT. Meta also offers its own list template in the upload flow, which can save you from guessing column order.

## Clean values match better than raw values

Matching works on exact normalized values, so the cleanup step is where an owner can do the most for the result. Apply the same few rules to every row.

Raw value versus clean value (made-up examples)

| Field | Raw value | Clean value |
| --- | --- | --- |
| Email | Jane.Doe@Example.com (with a trailing space) | jane.doe@example.com |
| Phone | (203) 555-0142 | +12035550142 |
| Phone | 203.555.0142 | +12035550142 |
| Country | USA | US |

Lowercase every email address and remove spaces. Put phone numbers in international format, which means the country code first, so a U.S. number starts with +1. Remove duplicate rows and rows with a blank identifier. The values above are invented for the example, so do not copy them into a real file.

Meta's upload flow can handle unhashed identifiers and hashes customer data locally before it is sent, per Meta's [Customer List Custom Audiences Terms](https://www.facebook.com/legal/terms/customaudience/update), effective April 25, 2023. You do not need to hash the file yourself unless you choose to. Whether you hash it or not, the clean version of each value is what you want in the column.

![A small pile of paper slips with unreadable pencil scribbles and a few crossed out lines, next to a pair of reading glasses.](https://webleveling.com/blog/facebook-custom-audience-from-customer-list/customer-list-cleanup-slips.jpg)

Duplicates and blanks are easier to remove before the upload than after it.

## Remove the people who should not be in the audience

This is the step owners skip, and it is the one with real consequences. Meta's terms say you must have all necessary rights and permissions and a lawful basis to disclose and use the data. People who have opted out of having their data used for targeting audiences must be left out, and a later opt-out must be honored by removing that person.

So, before you upload, take out anyone who unsubscribed, objected or asked not to be marketed to. Then check that the people left in the file agreed to a use that fits what you are about to do. The UK Information Commissioner's Office says businesses should be fair, lawful and transparent, and should explain that customer details may be matched with platform users for advertising. Its [direct marketing guidance](https://ico.org.uk/for-organisations/direct-marketing-and-privacy-and-electronic-communications/direct-marketing-guidance/plan-direct-marketing/) is a reasonable place to read how a regulator frames that expectation. If your privacy notice does not mention advertising matches, read it again before you upload.

Two more cautions apply. Hashing is not a substitute for permission: the ICO says hashing "does not include all the data protection considerations and is not an anonymisation technique." And do not upload health, financial, government ID or children's information unless the rules that apply to you clearly allow it. This is general information, not legal advice. Privacy rules vary by place, data type and purpose, so if your list crosses borders or includes sensitive details, ask a privacy lawyer.

## The upload takes six steps in Ads Manager

With a clean file in hand, the upload is a short walk through Ads Manager. Menu names can change, so treat these as the path as of now.

1.  **Open Audiences**: In Meta Ads Manager, open Audiences from the menu.
2.  **Create the audience**: Choose Create audience, then Custom Audience, then Customer list.
3.  **Add the file**: Upload your CSV or TXT file, or use the list template.
4.  **Map your columns**: Match each column to its identifier, such as email, phone number, first name, surname, city, state, ZIP or country. Review the mapping before you continue, because a mismatched column is a common reason for a poor result.
5.  **Accept the terms**: Read and accept the Customer List Custom Audiences Terms.
6.  **Name it and create it**: Use a name that carries purpose and date, such as "Past customers, retention offer, October 2026."

After you create it, the audience needs time to process. Meta's terms say it uses the data for matching and deletes the hashed data promptly after matching. Do not judge the audience until processing finishes, and do not build a campaign around a number you saw in the first few minutes.

![A wooden desk with a closed notebook, a sealed envelope and a small clock lying face down.](https://webleveling.com/blog/facebook-custom-audience-from-customer-list/customer-list-waiting-to-process.jpg)

Let the audience finish processing before you decide whether it is big enough.

## Match rates depend on your data, not on a promise

It is natural to ask how many of your customers will show up. The limit is that the answer depends on facts you control and facts you do not. The number of people matched depends on which identifiers you supplied, how accurate they are, whether those people use Meta, and whether they are eligible to be included. Meta's terms also say it does not reveal who is in a customer list audience, so you will see an estimated size, not a list of names.

That is why a claim like "80% match rate" or "ready in 24 hours" from a vendor is a sales claim and not a Meta guarantee. A small audience can show as too small for delivery or reporting, and the exact threshold can vary by product, campaign setup and current platform rules. Confirm the current threshold, the status wording and any processing estimate inside your own Ads Manager upload flow.

If the audience looks small, check these first

| What you see | What to check | What to do |
| --- | --- | --- |
| Far fewer people than rows | Duplicate, blank or badly formatted values | Fix the spreadsheet and upload a fresh file |
| Audience marked too small | The size of your eligible list | Add an identifier or combine lists you have permission to use |
| Odd or mismatched results | The column mapping | Reopen the upload and assign each column again |
| Still processing | Time | Wait before judging the result |

A small list is not a failure. A business with a few hundred past customers may still run a useful retention or exclusion campaign, as long as the audience status in the account says it can be used. A bigger raw file does not guarantee a bigger usable audience, because accuracy, eligibility, consent, account settings and campaign restrictions matter more than an impressive row count.

## Three uses give the audience a job

An audience with no job is just a file. Pick one use for each audience and name it that way.

**A retention or replenishment offer.** Show existing customers a message that fits what they already bought: a reorder reminder, an upgrade, a loyalty offer or a related product. This is the plainest use, and it is the one where a customer list fits best, because you already have a reason to talk to these people.

**An exclusion from new-customer campaigns.** If an offer is meant only for new customers, use the customer audience as an exclusion, where the campaign and current Meta controls allow it. That stops you paying to show a first-order discount to people who already bought. It also avoids annoying loyal customers with an offer they cannot use.

**A lookalike seed.** An eligible customer audience can be used as the starting point for a lookalike audience, unless the campaign falls into a category where Meta restricts lookalikes. The separate post on [lookalike audiences on Facebook](https://webleveling.com/blog/facebook-lookalike-audiences/) goes into that one. A lookalike is not the same as targeting your list directly, since it reaches new people who resemble your customers.

![A set of three plain paper envelopes in a row on a wooden counter, one open, one sealed and one with its flap tucked in.](https://webleveling.com/blog/facebook-custom-audience-from-customer-list/customer-list-three-uses.jpg)

One audience, one job. Name it for the use and the month.

## Housing, jobs and credit ads change the rules

Some campaigns are not ordinary. Meta's Special Ad Category rules apply to ads for housing, employment, and financial products or services, including credit. In those categories the rules can restrict audience selection and prohibit lookalike-style targeting. Meta's January 2023 paper, [Toward fairness in personalized ads](https://about.fb.com/wp-content/uploads/2023/01/Toward_fairness_in_personalized_ads.pdf), describes the background to those limits.

Controls differ by country, campaign type and account. If your campaign falls in one of those categories, confirm which targeting, exclusion, audience-sharing and lookalike controls are available during campaign setup, and expect the lookalike option to be limited or missing. Do not assume the audience you built for a standard promotion can be used the same way.

Also remember that customer list audiences cannot be sold or transferred, except through Meta tools and agreements that permit sharing. Do not hand the list to another company to upload on your behalf unless your agreement and your permissions cover it.

## Keep the audience fresh and your records tidy

A customer list goes stale. People opt out, new customers arrive and old addresses stop working. Set a regular refresh: export again, clean again, remove opt-outs and replace the audience. Name each version by purpose and month so you can tell them apart.

Keep a short record for each audience: where the data came from, when you exported it, who was excluded and why, what the audience is for and what business action you will count as success, such as a repeat purchase or a qualified inquiry. The business that handles this well has a clean source, a documented permission, accurate email and phone fields, and a clear purpose. The business that does not tends to get stuck on upload errors, reaches too few people, and cannot explain why a person was targeted.

If you want to know whether the campaign worked, measure it against that business action. Meta's reporting can tell you who saw and clicked. Your own sales or booking records tell you who came back. The post on [how much to spend on Facebook ads](https://webleveling.com/blog/how-much-to-spend-on-facebook-ads/) can help you set a budget that fits the size of the audience you ended up with.

## Hire help only for the parts that are hard

You can do an ordinary upload yourself. If your CRM export is clean, your permissions are documented and the campaign is a standard one, paying someone only to click through Ads Manager adds little.

Help becomes worth the money when the work around the upload is the hard part:

- **Messy data**: Customer records are scattered across several systems and nobody can say which are current.
- **Unclear opt-outs**: You cannot tell who asked to be left out.
- **Special Ad Categories**: The campaign covers housing, employment or credit.
- **Recurring refreshes**: The list should update on a schedule instead of by hand.
- **Measurement**: You want audiences tied to repeat purchases or qualified leads, not clicks.

A good provider shows you how the audience was built, what data went in, what was left out and how success will be measured. Nobody can guarantee Meta's match rate or your results, so walk away from anyone who does.

**Quiz: Check your customer list before you upload**

1. Which email format should go in the file?
   - Jane.Doe@Example.com with a space on the end
   - jane.doe@example.com, lowercase with no spaces **(correct answer)**
   - Jane Doe at example dot com
   - Any format, because Meta fixes it
2. A customer unsubscribed from your marketing last year. What should you do before uploading?
   - Leave them in, because hashing makes the upload anonymous
   - Leave them in, because they are still a customer
   - Remove them from the file **(correct answer)**
   - Move them to the last row
3. A vendor promises an 80% match rate on your list. How should you treat that promise?
   - As a Meta guarantee
   - As a sales claim, since the match depends on your data and on who uses Meta **(correct answer)**
   - As the minimum you should accept
   - As proof your file is clean

## Frequently Asked Questions About facebook ads create custom audience

**Can I use customer emails for [Facebook ads](https://webleveling.com/services/facebook-ads/)?**

Yes, if you have the rights, permissions and lawful basis Meta's terms require. Remove anyone who opted out, and make sure the use fits what customers were told.

**Can I use phone numbers instead of emails?**

Yes. Use accurate numbers in a consistent international format, with the country code first. Supplying both an email and a phone number for each person gives the file more to match on.

**Does Meta hash the list?**

Meta's terms say customer data is hashed locally before upload in the matching process, and that Meta deletes the hashed data promptly after matching. Hashing does not remove your privacy obligations.

**Will every customer on my list match?**

No. The result depends on your identifiers, their accuracy, whether people use Meta and whether they are eligible. Check the audience status in your own Ads Manager account.

**Can I exclude customers from a campaign?**

Yes, where the campaign and current Meta controls permit it. Use a customer audience as an exclusion in a new-customer campaign so existing customers do not see that offer.

**Can I make a lookalike audience from my customers?**

An eligible customer audience can be used as a seed, subject to current campaign restrictions. Housing, employment and credit ads fall under Special Ad Category rules that can restrict it.

## The Bottom Line

A customer list audience comes down to four moves: export one clean row per person, normalize emails and phone numbers, upload and map the file in Ads Manager, and give the audience time to finish processing. Permission matters as much as formatting. Remove opt-outs, keep the file free of sensitive data, and remember that hashing does not replace a lawful basis. Not every row will match, and your own account shows the status that counts.

Done well, the audience pays off in plain ways. You stop showing new-customer offers to people who already bought, you reach past customers with something relevant, and you give any later lookalike a better starting point. A short record of where the list came from and what it is for keeps the whole thing easy to defend and easy to refresh.

If you would rather hand off the data cleanup, the opt-out checks and the measurement, [Web Leveling](https://webleveling.com/) can set that up with you, and our [Facebook Ads](https://webleveling.com/services/facebook-ads/) work covers audience setup and reporting. If the upload is the only thing in your way, do it yourself first, because it is a job you can finish this afternoon. We work with small and medium businesses across the country and overseas, so [talk to us about your customer list](https://webleveling.com/contact/) whenever you want a second set of eyes on it.

## Terms

Customer list audience words in this post

Tap a term to see what it means.

**Custom Audience.** A group of people you build in Meta Ads Manager so your ads can reach them or skip them.

**Customer list.** A file of contact details you already own, such as emails and phone numbers, used to build a custom audience.

**Hashing.** A way of scrambling values so they are not sent in plain text; it is not the same as anonymizing data.

**Lookalike audience.** A group of new people Meta finds who resemble the people in a seed audience.

**Special Ad Category.** A Meta classification for housing, employment and credit ads that limits some targeting options.

**Opt-out.** A person's request not to have their data used for marketing or targeting, which you must honor.

- [Boost Instagram Live: Is Paying for a Live Worth It?](https://webleveling.com/blog/instagram-boost-live-video/): Instagram now lets accounts promote a live while it is on. See who can use it, how it is paid for, and when a live boost beats boosting a post or Reel.
- [Is Threads Worth It for a Small Business? Test It Small](https://webleveling.com/blog/is-threads-worth-it-for-a-small-business/): Is Threads worth it for business? Meta reports 500 million monthly users, but Pew found 8% of U.S. adults. Run a ten-minute check and a small 12-week test.
- [Is Meta Verified for business worth it? Price and fit](https://webleveling.com/blog/is-meta-verified-for-business-worth-it/): Is Meta Verified for business worth it? See Meta’s dated US prices from $14.99 to $499.99, what each tier includes, and who it fits before you pay.