Value-Based Customer List Builder for Meta Lookalike Audiences

How to build the list for a value based lookalike audience

A value based lookalike audience starts from a customer list with one number per person. Meta's guide to creating a customer value custom audience calls this the source audience. Store exports have one row per order, refunds mixed in and totals like “$1,249.00”. This builder turns that export into the source list in one pass, on your own computer.

1. Export your orders. A Shopify orders export works as it is, and so does any order or customer report with an email and a money column. A longer date range gives Meta more buyers to learn from.
2. Check the matched columns. Email, phone, order total, order number, payment status, refunds, currency and marketing consent are matched for you.
3. Pick what “value” means. Total spent is the usual choice. Average order value is one of Meta's own examples. Number of orders works like Meta's “number of items ordered” example, counted per order instead of per item.
4. Choose plain or hashed. Ads Manager hashes a plain file for you. Pick hashed if no raw emails may leave your company.
5. Read the report, then download. You see which orders were skipped and why, and how much revenue sits with your top 10% of buyers.

How the value column is worked out

• One row per customer. Orders are grouped by email address, or by phone number for orders without an email.
• Each order counted once. Shopify writes extra rows for every extra product in an order, with the total only on the first row. Only that first row is used.
• Paid orders only. Refunded, voided, canceled and failed orders are skipped, and so are unpaid ones — Shopify counts Authorized, Pending, Expired and Partially paid as unpaid. A refunded amount is subtracted.
• One currency. The most common currency in the file is kept. Orders in other currencies are listed, not added up.
• Numbers only. Meta's formatting guidelines ask for a numeric value with no currency characters, so “$1,249.00” is written as 1249.

What to do next in Meta Ads Manager

1. Start a customer list. Go to Audiences, then Create audience › Custom audience › Customer list.
2. Answer the customer value prompt. Prepare the list, answer the Customer value question and select Next. Add the file you downloaded here and give the audience a name.
3. Point Meta to the value column. Select Yes, customer value is included in my customer list, open the Select value column drop-down and choose value.
4. Check the mapping and import. Fix anything marked with a yellow exclamation mark, then select Import and create.
5. Create the lookalike. Use the prompt Meta shows after the upload, or go to Audiences, open Create audience, choose Lookalike audience and select this list as the source.

Common problems with value-based lookalikes

Value based lookalike audience not available

Meta says the source “needs to be a custom audience that includes customer value as the identifier type”. Meta's API notes say a list without values can't later be replaced by one with values, so upload a new list instead. A list created for Only exclusions can never be a lookalike source. Brand-new ad accounts may also have to wait a few weeks before they can create customer list audiences.

The value column has currency symbols or commas

Meta wants a numeric value only, without currency characters. Values like USD 49, 49,00 € or $1,249.00 don't fit that rule. Files from this tool always write 49, 49.5 or 1249, whatever the format in your export.

The audience is too small

Meta asks for at least 1,000 customers in a customer list. Export a longer period of orders, or add older exports from a previous platform in the same currency.

Value based lookalike vs regular lookalike

A regular lookalike treats every customer the same; a value-based one weighs people by your number, so a few big spenders can steer it. Test both side by side.