A vendor says its AI-augmented CDP resolves nearly all of your customers. Ninety percent of what, exactly? Identity resolution is the most requested CDP capability, and it is also the easiest to oversell. Before you sign, you need a test that measures errors, not a slide that reports a single match rate. Here is how to run one.
Why does everyone ask for identity resolution?
Demand is not in question. A Forrester report for Bluecore, as summarized by the CDP Institute, found that identity resolution (35%) topped the list of expected CDP features. The same write-up noted that respondents also expected it from CRMs and marketing clouds, which are notoriously weak at it.
That tells you something about the buying dynamic. Buyers want the capability badly and understand it poorly. Vendors know this. “AI-powered” now sits in front of almost every identity pitch, and the label rarely says what the AI actually does.
For influencer programs, the stakes are concrete. If a creator’s code redemption, a landing-page visit, and a later in-app purchase land in three separate profiles, your creator ROI looks worse than it is. If two different shoppers get merged, you credit the wrong creator and suppress the wrong audience. Both errors distort budget decisions.
Match rate is the wrong headline number
A match rate measures how many records got linked. It says nothing about whether the links are right. A system that merges aggressively will post a spectacular number and quietly corrupt your profiles.
The CDP Institute’s guidance is blunt on this point: measure both false merges and missed matches using representative data, rather than relying on one vendor-reported figure. Treat that as your evaluation standard.
Ask every vendor for two numbers, not one: how often it merges people who are not the same, and how often it fails to connect people who are. A match rate alone hides both.
We have covered the same trap on the measurement side in our piece on match rate claims. The logic carries over to CDPs. Whatever the denominator is, make the vendor define it in writing.
Deterministic, probabilistic, or hybrid?
Start by understanding what is under the hood. The CDP Institute defines the split this way: deterministic matching uses exact identifiers or explicit rules, while probabilistic matching estimates the likelihood that records belong together from several signals. A hybrid approach combines both.
- Deterministic: transparent and reproducible. Your team can inspect the rule and see why two records joined. It misses people who use different emails, check out as guests, or change names.
- Probabilistic: recovers relationships exact rules miss, but thresholds, training data, and explainability decide whether it is safe to use.
- Hybrid: uses high-confidence anchors (a loyalty ID, an authenticated user ID) and scoring for the messy remainder.
Where does AI fit? Usually in the probabilistic layer: fuzzy name matching, scoring, or suggested merges. That is useful. It is also where opacity creeps in. If a vendor cannot explain why two profiles were merged, you cannot audit a bad attribution result, and you cannot defend it to a finance partner.
A vendor test you can run in two weeks
Do not accept a demo on the vendor’s sample data. Run a bake-off on yours.
- Build a ground-truth set. Pull a few thousand records where you know the truth: logged-in customers with stable IDs, plus known duplicates from past CRM cleanups. Include messy ones: guest checkouts, shared household emails, privacy-relay addresses.
- Seed creator-driven traffic. Include records from a past influencer campaign: promo code redemptions, UTM-tagged sessions, and post-click purchases. This is the use case that will break first.
- Hide the answers. Give the vendor the data without the labels. Have them resolve it.
- Score the output. Count false merges and missed matches separately. Break results out by channel and by identifier type.
- Interrogate the misses. Ask the vendor to explain a sample of wrong merges. Can they show the rule or score behind each one?
Insist on a tunable policy, too. A conservative threshold suits a customer-service view, where a wrong merge is embarrassing. A looser one may be fine for paid-media suppression. The CDP Institute notes that match policies should be tuned to the risk and purpose of each use case. If a platform offers one global setting, that is a limitation to price in.
Where does the identity data come from?
This is the question sales teams hope you skip. Some vendors boost coverage with third-party identity graphs, partner data, or broker-sourced attributes. That raises a provenance and compliance question your legal team will care about.
Regulators are paying attention to data brokers. In early 2026, the FTC sent letters to 13 data brokers about complying with PADFAA, warning that violations could carry civil penalties of up to $53,088 per violation. The act targets data flows tied to foreign adversaries, so it will not touch every vendor. Still, it is a signal. If a CDP’s match lift depends on purchased data, ask who the sources are, what consent covers them, and who carries the liability if a source is challenged.
Get these answers in the contract, not on a call:
- A list of third-party data sources feeding the graph, and which are optional.
- Warranties on consent and lawful collection.
- Indemnification terms if a data source is later found non-compliant.
- Data deletion and portability rights, including the resolved ID map.
If a CDP’s coverage gain comes from data you cannot trace, you have not bought accuracy. You have bought an unpriced compliance risk.
CDP, identity graph, or something lighter?
Not every brand needs a full CDP to fix identity. Our breakdown of CDP versus identity graph explains how the two differ as an attribution foundation. If your core problem is stitching creator touchpoints to purchases, a lightweight identity resolution layer may do the job at lower cost. Large enterprises standardizing on a suite should also read our look at enterprise identity stacks before assuming a bundle solves it.
Agent-based matching is the newest wrinkle. If a vendor pitches autonomous matching, hold it to the same standard we outline in MCP identity agents: show the error rates, show the audit trail.
Red flags in the pitch
- A single match rate with no definition of the denominator.
- “Proprietary AI” with no description of inputs, thresholds, or review workflow.
- No way to unmerge profiles or roll back a bad rule.
- Case studies from industries or data volumes unlike yours.
- Reluctance to run a blind test on your data.
Any one of these can be negotiated. A vendor who refuses the blind test is telling you what the answer would be.
FAQs
What is identity resolution in a CDP?
It is the process of linking records from different systems and channels (web sessions, CRM entries, purchases, app activity) to one customer profile. Methods range from exact-match rules to probabilistic scoring.
What is a good match rate for a CDP?
There is no universal benchmark. A high rate can mean strong coverage or reckless merging. Evaluate false merges and missed matches on your own data, as the CDP Institute recommends, instead of comparing headline percentages across vendors.
Is deterministic or probabilistic matching better?
Neither wins everywhere. Deterministic is more explainable and precise when identifiers are reliable. Probabilistic recovers more relationships but needs governance. Most brands benefit from a hybrid with policies tuned by use case.
How does identity resolution affect influencer ROI?
Fragmented profiles split one customer’s journey into several, which can hide the creator’s contribution. Wrong merges can credit the wrong creator or suppress the wrong audience. Both skew budget allocation.
What should I ask a vendor about data sources?
Ask which third-party or broker data feeds the graph, what consent covers it, whether it is optional, and who is liable if a source is challenged. Put the answers in the contract.
Do I need a full CDP for identity resolution?
Not always. If your goal is narrow, such as connecting creator touchpoints to sales, a lighter identity layer or an existing warehouse may be enough. Match the tool to the use case before buying breadth you will not use.
Next step: before your next vendor call, assemble a labeled sample of your own records, including one past creator campaign, and make a blind test on it a condition of moving to contract.
Top Influencer Marketing Agencies
The leading agencies shaping influencer marketing in 2026
Agencies ranked by campaign performance, client diversity, platform expertise, proven ROI, industry recognition, and client satisfaction. Assessed through verified case studies, reviews, and industry consultations.
Moburst
-
2
The Shelf
Boutique Beauty & Lifestyle Influencer AgencyA data-driven boutique agency specializing exclusively in beauty, wellness, and lifestyle influencer campaigns on Instagram and TikTok. Best for brands already focused on the beauty/personal care space that need curated, aesthetic-driven content.Clients: Pepsi, The Honest Company, Hims, Elf Cosmetics, Pure LeafVisit The Shelf → -
3
Audiencly
Niche Gaming & Esports Influencer AgencyA specialized agency focused exclusively on gaming and esports creators on YouTube, Twitch, and TikTok. Ideal if your campaign is 100% gaming-focused — from game launches to hardware and esports events.Clients: Epic Games, NordVPN, Ubisoft, Wargaming, Tencent GamesVisit Audiencly → -
4
Viral Nation
Global Influencer Marketing & Talent AgencyA dual talent management and marketing agency with proprietary brand safety tools and a global creator network spanning nano-influencers to celebrities across all major platforms.Clients: Meta, Activision Blizzard, Energizer, Aston Martin, WalmartVisit Viral Nation → -
5
The Influencer Marketing Factory
TikTok, Instagram & YouTube CampaignsA full-service agency with strong TikTok expertise, offering end-to-end campaign management from influencer discovery through performance reporting with a focus on platform-native content.Clients: Google, Snapchat, Universal Music, Bumble, YelpVisit TIMF → -
6
NeoReach
Enterprise Analytics & Influencer CampaignsAn enterprise-focused agency combining managed campaigns with a powerful self-service data platform for influencer search, audience analytics, and attribution modeling.Clients: Amazon, Airbnb, Netflix, Honda, The New York TimesVisit NeoReach → -
7
Ubiquitous
Creator-First Marketing PlatformA tech-driven platform combining self-service tools with managed campaign options, emphasizing speed and scalability for brands managing multiple influencer relationships.Clients: Lyft, Disney, Target, American Eagle, NetflixVisit Ubiquitous → -
8
Obviously
Scalable Enterprise Influencer CampaignsA tech-enabled agency built for high-volume campaigns, coordinating hundreds of creators simultaneously with end-to-end logistics, content rights management, and product seeding.Clients: Google, Ulta Beauty, Converse, AmazonVisit Obviously →