Customer Match lets eligible advertisers use first-party customer information to create audience segments across supported Google properties and campaign types, including Demand Gen inventory that can appear in Gmail. Uploading a file is only one step. The team must define a permitted use, prepare fields correctly, protect the data, monitor processing, and avoid assuming every record will match or be available everywhere.
VISUAL LESSON
What you will learn
- 01Define an approved Customer Match use case.
- 02Prepare and validate identifiers.
- 03Interpret upload and activation status.

ILLUSTRATIVE WORKED EXAMPLE
Follow an illustrative audience file through processing
PRACTICAL INTERFACE MAP
Move from audience brief to Demand Gen activation
Record purpose, collection source, retention rule, identifiers, exclusions, owner, and approval.
Select the supported customer-list path, map fields, verify hashing guidance, and review errors.
Check processing, size, eligibility, inclusion or exclusion logic, and Gmail-capable Demand Gen settings.
STEP-BY-STEP LESSON
Purpose → normalize → protect → upload → activate
THE LEAD ATLAS METHOD
Lead Atlas Data can research public business contacts for a campaign's categories, markets, and locations, but the advertiser remains responsible for deciding whether and how any contact data may be used in Customer Match under current Google rules and applicable law.See how custom list research works ↗Approve the audience purpose
Write who is in the file, how the business obtained the information, the relationship or permission supporting the intended advertising use, the campaign purpose, regions, retention period, owner, and deletion process. Check current Google Customer Match policy, account eligibility, and applicable law before uploading.
Separate customers, active opportunities, former customers, and unrelated prospects when their permitted uses differ. Build exclusions for people who opted out or should not receive the campaign. Do not represent a researched public contact as a consenting customer.
Prepare supported identifiers
Use Google's current template and field requirements. Normalize names, countries, postal codes, phone numbers, and email addresses without inventing missing values. Remove malformed rows and duplicates while preserving a source record ID outside the upload file for audit and suppression purposes.
Run local checks for required headers, encoding, whitespace, casing where relevant, country information, phone format, duplicate identifiers, and blank rows. Keep raw source data separate from the working export and restrict access to the people who need it.
Protect and upload the data
Follow the current Google instructions for supported upload methods and hashing. Some workflows can hash normalized customer data; do not improvise cryptography or expose a plaintext file in tickets and chat. Record the audience name, account, uploader, date, row count, and source version.
Upload a small validated file first when practical. Map every field deliberately, read warnings, and capture processing status. If the interface rejects rows, fix the documented format issue rather than padding or fabricating fields.
Interpret matching and eligibility
A processed file is not a promise that every record matched, and a matched audience can still be too small or ineligible for a specific use. Privacy thresholds, account access, region, campaign type, consent settings, and policy can affect activation. Do not publish a match rate as data quality without context.
Record accepted rows, errors, processing time, displayed audience status, available campaign uses, and any current size ranges. Keep the broader Demand Gen audience and creative plan viable so delivery does not depend on one narrow list.
Monitor, refresh, and delete responsibly
After activation, check segment status, campaign inclusion or exclusion, reach, frequency, creative suitability, landing experience, conversions, customer complaints, and suppression updates. Refresh only from the approved source and expire segments when the purpose or retention period ends.
Deliverable: Customer Match purpose brief, source and permission record, normalized schema, validation report, protected upload log, diagnostics capture, activation matrix, suppression and deletion rules, campaign monitoring plan, and responsible proceed, repair, narrow, or do-not-upload decision.
THE TAKEAWAY
Use Customer Match only with an approved first-party purpose, normalize and protect the data, read upload diagnostics honestly, and plan a campaign that can work without treating the matched segment as complete.OFFICIAL REFERENCES