Prepare your customer data
This article contains four sections (see below for links to each). The first section is a set of general best practices for providing data from a customer file, pixel or offline event set that can help increase your match rate. The second and third sections provide guidelines and examples for formatting data about your customers and events. The fourth section is about customer lifetime value, a type of data only required for the creation of value-based Lookalike Audiences.
General best practices
Below, you'll find tables listing all the data types we can use for a Custom Audience or offline event upload. To get the highest match rate possible from your data, include as many data types as you can. You can use a single-column file to create your audience, but you may not see as many matches as you would if you used additional data types.
If you're using multiple data types, make sure that your data is organised into separate columns that correspond to the types we accept. For example, don't include a single column for full names. Instead, include two columns – one for first names, one for surnames.
Use the column headers in the table below. Doing so helps us auto-detect data types. You may use your own column headers if you prefer, but you'll then have to map the data to its type manually before uploading your customer list.
The two most important tips are:
- Always include the country code as part of your customer's phone numbers, even if all of your data is from the same country.
- Always include your customers' countries in their own column in your file, even if all of your data is from the same country. Because we match on a global scale, this simple step helps us match as many people as possible from your customer list.
Your file can be either CSV or TXT format. Download example files:
Learn how to create and edit a Custom Audience from a customer file.
Formatting customer data
| Data type | Column header | Description and formatting guidelines | Examples |
|---|---|---|---|
| Email address | We accept one separate email address column in US and international formats. |
| |
| Phone number | phone | Phone numbers must include a country code to be used for matching. For example, a 1 must precede a phone number in the United States. We accept up to 3 phone numbers as separate columns, with or without punctuation. Important: Always include the country code as part of your customer's phone numbers, even if all your data is from the same country. |
|
| First name | fn | We accept first name and first name initial, with or without accents. Initials can be provided with or without a full stop. |
|
| Surname | ln | We accept full last names, with or without accents. |
|
| City | ct | We accept full city names as they normally appear. |
|
| State/Province | st | We accept full names of US and international states and provinces, as well as the abbreviated versions of US states. |
|
| Country | country | Country must be provided as an ISO two-letter country code. Important: Always include your customers' countries in their own column in your file, even if all of your data is from the same country. Because we match on a global scale, this simple step helps us match as many people as possible from your customer list. |
|
| Date of birth | dob | We accept 18 different date formats to accommodate a range of month, day and year combinations, with or without punctuation. |
|
| Year of Birth | doby | We accept year of birth as a 4-digit number, YYYY. |
|
| Age | age | We accept age as a numerical value. |
|
| Zip/Postalcode | zip | We accept US and international zip and postcodes. US zip codes may include a 4-digit extension as long as they're separated by a hyphen. The extension is not required and will not improve the match rate. |
|
| Gender | gen | We accept an initial to indicate gender. |
|
| Mobile Advertiser ID | madid | We accept 2 types of mobile advertiser IDs: Android's Advertising ID (AAID), which Google provides as part of Android advertising, and Apple's Advertising Identifier (IDFA), which Apple provides as part of iOS in its ads framework. |
|
| Facebook App User ID | uid | An ID corresponding to someone who uses an app that can be retrieved through the Facebook SDK. We accept numerical user IDs associated with your Facebook application. |
|
| Facebook Page user ID | An ID corresponding to someone who has interacted with your Facebook Page. We accept numerical user IDs associated with your Facebook Page. |
|
Formatting event data
| Data type | Column header | Description and formatting guidelines | Examples |
|---|---|---|---|
| Event time | event_time | For offline event data, precise timestamps that include the minutes and seconds are recommended so that all unique events are processed, such as multiple purchases made by the same person in one day. See the examples for how to format the date and time combined – please note that there is a capital "T" in between the date and time. |
|
| Event name | event_name | For offline event data, see the examples for the standard conversion events you can track. If you have additional types of conversions you want to track, you can create custom conversions. |
|
| Value | value | The value field must contain a decimal number greater than or equal to zero, and may not include letters, special characters, currency symbols or commas. |
|
| Currency | currency | The currency field must contain a standard 3-letter ISO currency code as per the ISO 4217 standard. |
|
| Order ID | order_id | The order ID field may contain an alphanumeric value with no special characters, currency symbols or commas. For offline event data, use the Order ID field to provide a unique identifier for each transaction. Each Order ID should have a unique event time and customer associated with it. |
|
| Item number | item_number | The item number field may contain an alphanumeric value with no special characters, currency symbols or commas. For offline event data, you can split a transaction into multiple events by using the Item number field to provide unique item numbers for each event that is associated with the same Order ID. |
|
Customer lifetime value
What is customer lifetime value?
Customer lifetime value is a representation of the net profit you think will be attributable to a given customer over the duration of your relationship with them. You can break that value down into a few major factors:
- How often a customer makes a purchase within a typical purchase cycle
- How much a customer spends each time they make a purchase
- How much you project a customer will spend over the duration of your relationship with them
- The potential length of a customer's relationship with you
What can I use customer lifetime value for?
You can send us your customer lifetime value data as part of a customer file Custom Audience. You can then use that Custom Audience as the source for a value-based Lookalike Audience, which is a group of people you can target with ads who are most similar to the high-value customers you already know.
What should I avoid when calculating customer lifetime value?
Different people calculate customer lifetime value in different ways. Here are some ways that you should avoid if you're going to send us the data for use in a value-based Lookalike Audience:
- Rating your customers. Say you have three customers worth £100, £10 and £1, respectively, and you use a rating system of 1-5. Don't send us data where they're rated as a 5, 2 and 1, respectively. This doesn't work because the value isn't proportional to the ranking. In other words, the £100 customer was factored by 20, but the £10 customer was only factored by 5 and the £1 customer wasn't factored at all.
- Ranking your customers. Say you have 100 customers and each one is worth between £200 and £1200. You rank them from 1 to 100. This doesn't work because value isn't proportional here either. It tells us if one customer is more valuable than another, but doesn't account for a scenario where the number 5 customer is worth double what the number 6 customer is, whereas the number 20 customer might only be worth 1% more than the number 21 customer.
Once I've calculated my customer lifetime value data, what should I do?
Send us the data as another column in a customer file you use to create a Custom Audience.
Below are tips for formatting the data, similar to the tips for other identifiers in the first section of this article.
| Data type | Column header | Description and formatting guidelines | Examples |
|---|---|---|---|
| Customer lifetime value | value | We accept customer value as a positive number. Important:
|
|
Once your customer file Custom Audience is created, use it as a source for a value-based Lookalike Audience. Use that Lookalike Audience as the target audience for your ads to reach people who are most similar to your high-value customers.
* Nguồn: Facebook