Meta Ads

Audit Checklist

Instrumentation (GTM)

Has a Meta Pixel firing on All Pages or similar

Platform Settings

Events Manager > Pixel Settings: Configure domain "allow" list to only include production domains.
Events Manager > Pixel Settings: Disable "Track events automatically without code". All event implementation should be done though GTM tags/triggers

1st Party Cookies

All examples below use a placeholder value ([PARAM:X]) for a given "click identifier" (where X would be fbclid). This is done to keep the examples brief and easy to read/reference.

Cookie Name	Value Format
`_fbc`	`fb.1.[TIME:MS].[PARAM:fbclid]`
`_fbp`	`fb.1.[TIME:MS].[RANDOM:#]`

The "real" values of the [PARAM:fbclid] placeholders can be as long as 255 characters.
Make sure that whatever field/column (e.g. CRM or database) can store a value of this length.

Conversions API

Data Collection Checklist

_fbc cookie value is collected and stored with CRM record(s).
_fbp cookie value is collected and stored with CRM record(s).
User Agent is collected and stored with CRM record(s).
IP Address is collected and stored with CRM record(s).
Conversion URL (e.g. Capture URL) is collected and stored with CRM record(s).

It is fairly common practice to collect the fbclid directly from the URL parameter. This technically works, but Meta requires formatting the fbclid in its _fbc format (which includes a timestamp of when the fbclid was encountered) when sending the data to the CAPI. Therefore, it is recommended to grab the full _fbc 1st party cookie value to simplify CAPI integration.

Aside from the data prep above, if you wish to configure a redundant setup (recommended by Meta), we recommend using Server-Side GTM via Stape.io to implement a real-time CAPI event stream in-parallel to the client-side Meta pixel.

Hashed 1st Party Data (Advanced Matching)

Priority Order

Email Address
Phone Number
City, State (Region), Zip (Postal Code), and Country
Date of Birth, and/or Gender
Name (First & Last)

Meta's CAPI works best when leveraging hashed (SHA256) user data such as: email/phone and similar data. In order to obtain the best match rates, you MUST normalize the data before SHA256 hashing using the instructions below.

First/Last Name

Lowercase & trimmed
No special characters
No whitespace

Email

Lowercase & trimmed

Phone

Remove non-numeric characters
Must be prefixed with valid country code (e.g. 1 for US/CA numbers).

City

Lowercase & trimmed
No special characters
No whitespace

State (Region)

2 character ISO 3166-2 code
Lowercased

Zip (Postal Code)

US: Only 1st 5 digits
Non-US: Lowercase, no special characters, no whitespace

Country

2 character ISO 3166-2 code
Lowercased

Date of Birth

Must be in YYYYMMDD format

Gender

m for male, f for female