Google Ads is giving advertisers two new ways to improve Customer Match audience matching: IP addresses and interaction timestamps.
The change adds more first-party signals to a system already built around advertiser-owned data such as email addresses, phone numbers and mailing information. But it also creates a more sensitive operating requirement because Google says the new IP and timestamp fields must be sent unhashed.
Key Takeaways
- Customer Match now accepts IP addresses and interaction timestamps as additional first-party matching signals.
- Google requires those two fields to be uploaded unhashed and excludes end users in the EEA, UK and Switzerland from IP matching.
- The update makes privacy controls and regional eligibility part of the paid-media data pipeline, not just a legal review outside the campaign workflow.
Table of contents
Jump to each section:
- Customer Match gets two more identity signals
- Unhashed data changes the operating risk
- Regional eligibility has to be enforced before upload
- First-party data quality is becoming a media-performance lever
Customer Match gets two more identity signals
Google’s Customer Match upload documentation now includes fields for a user’s IP address and the timestamp associated with that interaction.
The IP field can contain either IPv4 or IPv6 addresses. Google says advertisers should trim whitespace and pass the value as a plain, unhashed string. Timestamps can record the first and most recent observed interaction from the corresponding IP address.
The timestamp is not designed as a standalone identifier. Google’s documentation says it cannot be sent without an accompanying IP address. If an advertiser uploads an IP address without a timestamp, Google may use the latest known user associated with that IP.
That means the practical value is not simply “one more identifier.” The platform is giving advertisers a way to add context about when an IP address was actually observed, which can help make a volatile signal more useful for matching.
Unhashed data changes the operating risk
Customer Match already requires advertisers to think carefully about what customer information leaves their own systems. This update makes that governance question more concrete.
For traditional identifiers such as email addresses and phone numbers, advertisers can normalize and hash data using SHA256 before upload. Google explicitly says IP addresses and interaction timestamps should remain unhashed.
That difference matters operationally. An export containing raw IP addresses now has to move through whatever customer data platform, cloud storage, agency workflow or API connection prepares the audience.
A campaign manager may never see the raw file, but the organization still needs controls around who can generate it, where it can be stored, whether it appears in debug logs, and how quickly temporary files are deleted.
ContentGrip has previously covered why first-party data becomes more important as privacy rules and third-party identifiers change. Google’s Customer Match update shows the other half of that transition: better first-party activation can also increase the burden on the systems handling the data.
Regional eligibility has to be enforced before upload
Google’s documentation says IP matching is not supported for end users in the European Economic Area, the United Kingdom or Switzerland.
That turns geography into a technical eligibility check. An advertiser operating globally cannot safely treat the new fields as a universal addition to every Customer Match export.
The upload pipeline needs to distinguish records that are eligible from records that are not. If the source system cannot reliably establish where a user was located when the IP address was collected, the safer implementation is to leave the new signal out rather than send it by default.
Google also requires advertisers to confirm that Customer Match data was collected and shared in compliance with its policies. The platform’s Customer Match data-use documentation says uploaded files are used to create Customer Match audiences and for policy compliance, rather than to build or enhance broader customer profiles.
For teams using agencies or external data processors, this is where the implementation can become more complicated. The media brief may say “improve match rate,” while the actual work spans consent language, collection logic, regional filtering, access control and audience upload tooling.
First-party data quality is becoming a media-performance lever
Google’s Data Manager API documentation says advertisers are encouraged to provide timestamps alongside IP addresses and that match-rate improvements from these signals are expected to begin in October 2026.
That gives performance teams a reason to test the feature, but it should not be treated as a guaranteed lift. Match rate is only useful if the resulting audience is still accurate, eligible and aligned with how the advertiser collected the data.
The larger direction is clear. Paid-media platforms are asking brands to bring richer first-party signals into their systems as third-party tracking becomes less dependable. The quality of the campaign increasingly depends on the quality of the advertiser’s own data infrastructure.
That changes the role of audience operations. Customer data hygiene, consent status, timestamps and regional eligibility can directly affect how much of a customer list becomes usable media inventory.
For marketers, the useful question is therefore not simply whether IP matching raises a Customer Match percentage. It is whether the organization can produce the additional signal in a controlled, auditable way without turning a media optimization into a data-governance problem.
