File Requirements for Matched Lists or Lookalike Generation

This document defines file formats that are supported for generating matched lists or lookalike audiences. For the same, files will need to be formatted as .csv (optionally gzipped) with headers designating each column.

Column headers are limited to a maximum of 256 characters
Column values are limited to a maximum of 1024 characters
Column headers are case-insensitive
Max file size: 1GB (drag and drop) and 10GB (SFTP)

For users who are generating matched lists with CSV files and also using Excel to view/edit the CSV files, please follow the instructions linked here.

Single quote is not allowed in the filename.

Supported Column Details

Column	Description	Examples
email* email1 email2 email3 email4	Zeta accepts email addresses in US and international formats. Zeta accepts up to 5 email addresses as separate attributes. Up to 5 clear text email addresses may be considered in a single input record, in the matching waterfall order of: email, email1, email2, email3, email4. E.g., If there is no match for “email”, the next field considered for matching is “email1”, and so on.	username@hotmail.co.uk your.name@gmail.com myname@yahoo.com
email_md5	Zeta accepts email addresses in US and international formats in md5 hash format.	f068d15f6196394d5e12e6bd4dec5879 674c1cc1cb589cde0af7071c3f4a7121 a3e45a16e0e3392daaa774808e685ca2
email_sha256	Zeta accepts email addresses in US and international formats in sha256 hash format.	000074c57969d281ea9f0cfe105d6768ce03bdef21d70b39861014e8a7963148
zync_cookie	Cookie ID for a visitor from Zeta integrated site.	21fa99ff-3f2a-43a3-b362-f87a827fc0dd:1604567862.93
sizmek_cookie	Cookie ID surfaced in the bitstream of our Demand Side Platform (DSP).	1875819619179876701
first_name*	Zeta accepts first name and first name initials, with or without accents. Initials can be provided with or without a period.	John F. Émilie
last_name*	Zeta accepts full last names with or without accents.	Smith Sørensen Jacobs-Anderson
address_line_1*	First line of the postal address	777 1ST AVE N
address_line_2*	Second line of the postal address	APT 7
zip*	Zeta accepts US and international zip and postal codes. US zip codes may include a 4-digit extension as long as they are separated by a hyphen. The extension is not required and will not further improve the match rate.	W11 2BQ 94104-1207 94104
city*	Zeta accepts full city names as they normally appear.	Paris London New York
state*	Zeta accepts full names of US and international states and provinces, as well as the abbreviated versions of US states.	AZ California Normandy
maid	Mobile Advertising Identifier (MAID) is used to target Mobile phones and tablets using Android as GAID (Google Advertising ID) and IOS as IDFA (Identifier for Advertising).	0093f37c-5b37-4217-a7e7-a59629ecf384
phone phone1 phone2 phone3 phone4	Zeta accepts 10-digit phone numbers. Phone numbers must include a US country code to be used for matching; 1 must precede a phone number in the United States. The Matched List won’t accept any non-numerical characters such as spaces, periods, etc. Zeta accepts up to 5 phones as separate attributes. Up to 5 phones may be considered in a single input record, in the matching waterfall order of: phone, phone1, phone2, phone3, phone4. E.g., If there is no match for “phone”, the next field considered for matching is “phone1”, and so on.	12345678910

When uploading a file for Lookalike, it is necessary to have the first header of the file as "Email". However, it is not mandatory to have any values in the field.

The name and postal match can only be performed if all the required/mandatory fields (marked with a *) have values inserted:

email
first_name
last_name
address_line_1
address_line_2
and at least one city, state, or zip

If the aforementioned requirements are not met in the input file, the (name + postal) stages match process will yield 0 results.

The address_line_2 field is mandatory but nullable. If you do not have a value, you can leave it empty.

If the matched list is created using email_md5, the mandatory fields mentioned above are not necessary.

Unless the fields matched upon are (name + postal), only one field needs to be in the header of a matched list seed file. However, a seed file’s header may contain multiple fields. Even so, only one populated field per record is required to yield results, with the exception of the record only containing (name+postal).

For file-based uploads containing hashed email data intended to generate a matched list, the data is not stored in Snowflake. Instead, it is securely uploaded to an S3 location, where it is processed by our servers. The physical location of the S3 storage aligns with our secure data infrastructure protocols.

Matching Logic of Matched Lists

2,000 records are required as a minimum to generate a matched list.

CODE

postal_fullname_address
phone
email_md5 | email*
email_sha256
ifa_ifa
maid_maid
cookie_disqus
cookie_zync
cookie_sizmek
cookie_appnexus
bsin

where postal_fullname_address refers to the following fields:

CODE

postal_fullname_address = [
    "postal_firstname",
    "postal_lastname",
    "postal_addressline1",
    "postal_addressline2",
    "postal_city",
    "postal_state",
    "postal_zip",
]

Example File

2021-10-28 Matched List Seed File.csv