How to choose an identity resolution platform

Identity resolution platforms all promise the same outcome: turn fragmented signals about the same person — an email, a phone number, a loyalty ID, a device ID — into one stable identifier you can target, measure and enrich against. What they actually differ on is where the underlying data asset comes from, and that question matters more here than in almost any other analytics category, because it determines both what the vendor can legally do with the data and how much scrutiny a partnership will draw from your own legal and privacy teams.

This category sits close to, and sometimes overlaps with, customer data platforms — several CDPs build their own identity resolution in-house. The tools here are used when you need identity matching as a standalone capability, often to connect a brand's own customer records to advertising and measurement partners it doesn't control.

Where the data asset actually comes from

This is the single most important thing to understand about each vendor before anything else.

• Credit bureaus. Experian Marketing Services and TransUnion TruAudience both draw on consumer databases built from the same corporate lineage as their credit-bureau businesses, restricted to marketing-permissible attributes rather than credit-file data. Both are explicit that credit data and marketing data are governed by different regulatory regimes even though they trace back to a related asset — ask each vendor directly what is and isn't permitted to cross that line.
• Agency and media relationships. Merkle Merkury links a client's first-party CRM data with offline records and behavioral signals gathered through Merkle's (Dentsu's) own agency and media operations. Because the graph is assembled through media buying, ask specifically what sources feed it and under what consent basis, since this differs meaningfully from a bureau-based or purely first-party approach.
• API-first, privacy-by-design matching. FullContact resolves fragments an organization already holds — hashed emails, phone numbers, device IDs — into a single identifier without requiring or returning raw PII, and offers a zero-data-retention option for regulated customers.
• A partner network built around one identifier. LiveRamp resolves your own customer records into RampID, a pseudonymous identifier matched against the same ID held by publishers and data partners, without either side exchanging raw personal data directly.
• An open, free standard. Unified ID 2.0 isn't a vendor product at all — it's a hashed, salted email/phone identifier originated by The Trade Desk and now governed by the independent Prebid.org under the IAB Tech Lab, typically adopted alongside a commercial vendor rather than instead of one.

Deterministic versus network-based matching

The records here point to two different matching philosophies. Bureau-backed vendors (Experian, TransUnion) and LiveRamp work by matching your records against a large, existing identity graph the vendor already maintains — accuracy depends on how comprehensive and current that graph is, which you largely have to take on faith or ask for match-rate evidence. FullContact instead resolves fragments you already hold into one identifier via API, which is a narrower job — it doesn't claim to expand your reach into new audiences the way a graph-matching vendor does. Understanding which job you actually need — expanding reach to new prospects, or cleaning up duplicate records of customers you already have — should come before comparing any vendor's claimed match rate.

Reach versus control

There's a real tradeoff between how much of the open advertising ecosystem a vendor connects to and how much control you have over your own data in the process. LiveRamp's core value is its partner network — DSPs, SSPs and walled-garden platforms — which gives reach at the cost of depending on LiveRamp's own infrastructure for interoperability. Unified ID 2.0 goes the opposite direction: it's open-source and vendor-neutral by governance, so no single company controls it, but as a standard rather than a resolution engine, it still needs a commercial identity vendor or your own infrastructure to do the actual matching.

What replaces the third-party cookie

Every tool in this category exists, in part, because third-party cookies are an increasingly unreliable way to recognize the same person across sites. The approaches differ: LiveRamp and UID2 both use variants of pseudonymous, hashed identifiers built for the open web; FullContact does the same at the record-matching level with hashed inputs and outputs. Bureau-backed vendors (Experian, TransUnion) and Merkury instead lean on offline data assets that were never cookie-dependent in the first place. If your primary goal is specifically replacing cookie-based targeting on the open web, UID2 and LiveRamp are the more directly relevant starting points; if it's appending demographic or offline data to records you already have, the bureau-backed vendors are closer to the job.

Pricing and how deals actually work

None of the six vendors here publish pricing. All are quoted per engagement, and several — Experian, TransUnion, Merkury — are commonly bundled with a broader agency, media or bureau relationship rather than sold as a pure standalone contract. Unified ID 2.0 is the one exception: it's free, since it's operated as shared industry infrastructure rather than a commercial product, though you'll still typically pay a commercial identity vendor for the matching work around it.

A shortlist by situation

• If you need to append demographic and lifestyle attributes to a CRM file from a large, established consumer database, look at Experian Marketing Services or TransUnion TruAudience.
• If you want privacy-preserving matching delivered as an API, with a zero-retention option for regulated use cases, look at FullContact.
• If you need one consistent identifier connecting your CRM to a broad network of advertising and measurement partners, look at LiveRamp.
• If you're already working with Merkle or Dentsu on media or customer experience, Merkury bundles identity resolution into that relationship.
• If you want a free, vendor-neutral identifier for the open web and already have — or plan to build — the matching infrastructure around it, evaluate Unified ID 2.0.

Questions to ask vendors

• Exactly which data sources feed the identity graph, and under what consent basis was each one collected?
• What is a realistic match rate against our own customer file, demonstrated on a sample rather than quoted as an average?
• Is any output raw PII, or only a pseudonymized or hashed identifier?
• If the vendor is bureau-affiliated, what specific firewall exists between marketing data and credit-file data?
• Can matched data be used inside a data clean room for collaborative analysis without either party exposing raw records?

Common mistakes

Assuming "identity resolution" means the same underlying data and method across vendors is the most common error — a bureau-backed graph, an agency-assembled graph, and a fragment-matching API answer genuinely different questions. Skipping the consent-basis question because a vendor's marketing emphasizes scale over provenance is a close second, and one that a privacy or legal review will eventually force anyway — better to ask it before signing than after. Treating Unified ID 2.0 as a drop-in replacement for a commercial vendor, rather than infrastructure that still needs one, is a specific and avoidable misunderstanding of what it actually is.

For two direct comparisons inside this category, see LiveRamp vs TransUnion TruAudience and Experian Marketing Services vs TransUnion TruAudience. The full list of tools in this category is at every tool in this category.