Supported and Preferred Identifier Combinations
See the sections below for information on which identifiers LiveRamp supports in Connect, which PII combinations work best for matching, and how we treat CTV IDs and CIDs.
Note
This article focuses on Connect Activation and Measurement Enablement workflows. For LiveRamp Clean Room–specific guidance, see "Format Your Clean Room Data".
Identifiers LiveRamp Supports
LiveRamp uses identifiers (touchpoints) such as name and postal address, email address, phone, RampIDs, cookies, mobile device IDs, and custom IDs (CIDs) to match your records to other identifiers in our Identity Graph.
Supported Known (PII-based) Identifiers
Known identifiers are sometimes called offline identifiers or PII identifiers. LiveRamp supports the following known identifiers for matching:
Name and postal address (NAP): Plaintext only. NAP requires the following elements for matching:
First name
Last name
Street address
City
State
ZIP/postal code
Note
If any of the required NAP components are missing, the NAP field will not be used as a matching identifier and may reduce performance.
Email address: Plaintext or hashed (SHA‑256, MD5, or SHA‑1, depending on region and workflow).
Phone number: Plaintext or hashed with SHA‑1 (depending on region and workflow).
Note
Hashed phone numbers are not allowed in EU/UK files.
Supported Pseudonymous Identifiers
Pseudonymous identifiers are not directly PII and are often called online identifiers. LiveRamp supports the following pseudonymous identifiers:
Cookies
Mobile device IDs (MAIDs): Plaintext or hashed with SHA‑1 (AAIDs and IDFAs)
CTV IDs: Plaintext device identifiers for Connected TVs.
Note
CTV IDs are fully supported as a pseudonymous identifier. For best performance, set up a CTV ID sync via event files or pixels before sending CTV IDs, so we can recognize and match more of your devices.
IP addresses: A unique number assigned to each device connected to a network (like the internet).
Note
IP addresses are supported only in specific LiveRamp workflows (mapping files, measurement, and CTV/podcasting/gaming/IoT targeting).
RampIDs: LiveRamp’s people-based, cross-device pseudonymous identifiers.
Custom IDs (CIDs): Platform-specific IDs (e.g., publisher or platform IDs that LiveRamp has a sync with).
You can include only one pseudonymous identifier type per file, so there are no supported pseudonymous identifier combinations.
Note
Do not mix known and pseudonymous identifiers in the same file for ingestion.
Preferred PII Identifier Combinations
For PII-based files, each record must include at least one usable identifier for us to match on (email, phone, or full NAP).
In order of preference, include these identifier combinations when possible to maximize match rates and accuracy:
Email + Phone + Full Name & Postal
Email + Full Name & Postal
Email + Phone + first & last name
Email + first & last name
Phone + first & last name
Note
For combinations that include first and last name only (no postal address), we primarily match on the email and/or phone. First and last name helps with data quality and accuracy but is not required for matching. We cannot match on just first and last name alone.
The following single identifiers are supported but less optimal for match rate and accuracy:
Email
Plaintext email typically yields the strongest single-identifier performance.
Hashed emails are supported, but match rates may be slightly lower than plaintext
Sending plaintext and all three allowed hash types (SHA‑256, MD5, SHA‑1) maximizes coverage.
Phone
Name & Postal
Including CIDs
In Connect, the audience key is the field we use to deduplicate rows and treat them as a single person record. A good audience key has near‑100% fill rate to avoid ingestion pauses and ensure reliable deduplication.
A client customer ID (CID) is usually the best choice for this key for the following reasons:
Stable over time:
A CID is designed to be a long‑lived internal ID for a person or account, even if their email, phone, or postal address changes.
Emails and phones churn frequently; if you use them as the key, you can end up with multiple “different” records for the same person when their contact info changes.
Cleaner deduplication and updates:
During ingestion, Connect uses the audience key to collapse duplicate rows for the same person and to apply updates cleanly across files.
A high‑fill CID column lets you:
Merge multiple source systems into a single view of the customer.
Run incremental updates without creating duplicate people when an identifier changes.
More flexibility across workflows:
In Clean Room and Embedded Identity setups, we commonly map CIDs to RampIDs and reuse that mapping across datasets.
Using CID as the key makes it easier to:
Join CRM, transaction, and exposure data on the same underlying person.
Swap or add identifiers over time without re-keying the dataset.
Note
LiveRamp can also match on CIDs themselves when your CID has been configured as a recognized custom ID and linked to RampIDs (for example, in LiveRamp Clean Room or Embedded Identity workflows).
Other identifiers (such as NAP or emails) can be used as the audience key, but because people often have multiple or changing addresses and emails, doing so can fragment one person into several records and make deduplication and updates less reliable.
For more information, see "Audience Key”.