One customer record across every device, channel, and location
The same customer on desktop Tuesday, in-app Wednesday, in-store Friday, and on the phone Saturday gets unified into one record — so cohort, LTV, and decisioning all see one person, not five.
The problem
The same customer hits your website on desktop Tuesday, your mobile app Wednesday, your Houston store Friday, calls your support line Saturday, and emails sales Monday. Mixpanel sees three different visitor IDs. Klaviyo has two email addresses for them. Your POS records the in-store visit as anonymous. Your call tracking logs the phone number. None of these systems know they are looking at the same person.
The categories of tools that touch this each handle a slice. Identity resolution specialists (LiveRamp, TransUnion Neustar, Acxiom, Merkle Sonar, Throtle, Tapad) are built for ad-tech at enterprise scale — $50,000 to $500,000+/year, 6 to 9 month integrations, uneconomic for operators your size. CDP-bundled identity (Segment, Tealium, mParticle, Klaviyo CDP, Bloomreach) handles deterministic rule-based stitching; probabilistic matching needs custom ML work. Enterprise customer 360 platforms (Salesforce, Adobe Real-Time CDP, Oracle Unity, Treasure Data) include identity resolution but at enterprise pricing tied to their platform. MDM platforms (Informatica, Reltio, Stibo) run enterprise customer-and-product MDM with 6 to 12 month implementations. Identity-graph data brokers sell access to their graphs, with mixed privacy compatibility.
The gap is identity resolution that combines deterministic-first (email, phone, login, loyalty ID, POS key) with probabilistic-fallback (device fingerprint, behavioral signature) at multi-location operator scale, with privacy rules per state and per vertical built in.
What success looks like
Deterministic matches resolve first with high confidence — the same email or phone number or loyalty card or POS payment identifier across systems means the same customer. Probabilistic matches fill the remaining gaps with a calibrated confidence score: a device fingerprint, a browser cookie, or a behavioral signature that strongly suggests the same person. You set the thresholds for which matches merge automatically; borderline ones route to review.
Multi-location operators get cross-location resolution with LTV rolled up across the chain. Multi-brand portfolios get a corporate base record with per-brand sub-records; cross-brand resolution is gated by privacy and consent. An in-store anonymous visit links to a known customer at the loyalty scan, the payment, or the email capture moment.
State-by-state and vertical privacy rules apply automatically. A CCPA, CPRA, or GDPR data subject access request surfaces every identifier linked to the customer. A right-to-deletion request propagates across the unified record.
LiveRamp, TransUnion, and the other ad-tech graphs stay useful for off-platform addressability. Your in-house decisioning, cohort math, LTV math, churn prediction, and save-flow scoring all run on the resolved record.
How most operators solve this today
A few categories of tools touch this problem, but none of them combine deterministic and probabilistic matching at multi-location operator scale with privacy rules built in:
Identity resolution specialists (LiveRamp, TransUnion Neustar, Acxiom, Merkle Sonar, Throtle, Tapad)
$50,000 to $500,000+/year
Built for ad-tech at enterprise scale. Expensive, with 6 to 9 month integrations. Uneconomic for operators your size.
CDP-bundled identity (Segment, Tealium, mParticle, Klaviyo CDP, Bloomreach)
$120 to $50,000+/year
Deterministic rule-based stitching bundled. Probabilistic matching needs custom ML configuration.
Enterprise customer 360 (Salesforce Customer 360, Adobe Real-Time CDP, Oracle Unity, Treasure Data)
$30,000 to $500,000+/year
Identity resolution included but tied to their platform. Over-built for the operator scale you live at.
MDM platforms (Informatica MDM, Reltio, Stibo Systems, Profisee, TIBCO EBX)
$50,000 to $500,000+/year
Enterprise customer-and-product MDM. 6 to 12 month implementations. Not optimized for marketing identity.
Identity-graph data brokers (LiveRamp IdentityLink, Tapad Device Graph, Experian Mosaic, Oracle Data Cloud)
$30,000 to $300,000+/year
Broker-maintained graphs. You bring your data and get back results resolved to their graph. Privacy compatibility varies.
Build it in-house
Data engineer + ML engineer + ongoing tuning
Probabilistic matching requires ML expertise. API drift across sources eats about a third of an engineer's time.
What changes when this is an agent skill
Deterministic matches resolve first at high confidence. The same email, phone, login ID, loyalty card, POS transaction key, or payment method across systems means the same customer. Probabilistic matches — device fingerprint, browser ID, behavioral signature — fill the remaining gaps with a calibrated confidence score. You set the thresholds for which matches merge automatically; borderline matches route to a review queue.
Multi-brand portfolios get a corporate base record with per-brand sub-records; cross-brand resolution is gated by your privacy policies and consent settings. Multi-location operators get cross-location resolution with LTV rolled up across the chain.
State-by-state and vertical privacy rules apply automatically. A CCPA, CPRA, or GDPR data subject access request surfaces every identifier linked to the customer. A right-to-deletion request propagates across the unified record.
Cohort math, customer change events, LTV math, churn prediction, and save-flow scoring all consume the resolved record as their canonical input. LiveRamp and the other ad-tech graphs stay useful for off-platform addressability.
Agents that include this skill
Skills live inside agent rentals. To get this skill in production, hire any of the agents below — context-tuning at onboarding is included in the first month.
Customer Data Graph Foundation Agent
Resolves DTC subscriber identity, computes LTV math, and emits the canonical customer-data-graph downstream subscription agents consume.
FAQ
- What does identity resolution actually do?
- It resolves the same customer's identifiers across web, mobile, in-store, and phone into one unified record. The same email, phone, login, or loyalty card on multiple devices ties together with high confidence. Where those identifiers are missing, the system uses signals like device fingerprint and behavioral patterns to suggest a match — at a confidence score you can tune.
- How is this different from LiveRamp or TransUnion Neustar?
- Those are built for ad-tech identity at enterprise scale — $50,000 to $500,000+/year with 6 to 9 month integrations. This is purpose-built for the multi-location operator scale, targeting the $2,000 to $10,000/month band.
- How is this different from CDP-bundled identity in Segment, Tealium, or mParticle?
- CDPs bundle deterministic rule-based stitching. Probabilistic matching needs custom ML configuration. This combines deterministic and probabilistic with calibrated confidence scoring out of the box.
- How is this different from Salesforce Customer 360 or Adobe Real-Time CDP?
- Enterprise customer 360 platforms include identity resolution at $30,000 to $500,000+/year. This is purpose-built for multi-location operator scale.
- What identifiers does it resolve across?
- Email, phone, login ID, loyalty ID, POS transaction key, browser cookie, device ID, behavioral fingerprint, payment method, and address (with privacy controls).
- How are confident and uncertain matches handled?
- Deterministic matches resolve at high confidence. Probabilistic matches get a calibrated score from 0 to 1. You set the threshold above which matches merge automatically; borderline matches route to a review queue.
- How does this handle CCPA, CPRA, GDPR, and privacy rules?
- Privacy rules per state and per vertical apply automatically. A data subject access request surfaces every identifier linked to the customer. A right-to-deletion request propagates across the unified record.