What is identity resolution?
Identity resolution is the process of matching and linking data points across systems to create a unified view of a real person or organization. In B2B, this means connecting a form fill, a CRM record, a LinkedIn profile, and a support ticket into one accurate, deduplicated identity — even when names, emails, and titles differ across sources.
Identity resolution is related to entity resolution (matching records within a database) and data enrichment (adding new fields to existing records). It operates at a higher level — stitching together a complete identity from fragments across multiple systems and time periods.
How identity resolution works
- Data collection — Identity fragments are gathered from multiple touchpoints: CRM records, marketing automation, website visits, form fills, support tickets, social profiles, and external data sources.
- Deterministic matching — Records are linked using exact identifiers like email address, phone number, or social profile URL. High precision but limited recall — people have multiple emails and change phone numbers.
- Probabilistic matching — When exact identifiers don't match, algorithms assess the likelihood that two records represent the same person based on name similarity, company affiliation, job history, location, and other signals.
- Identity graph construction — Confirmed links form a graph where each node is a data point and each edge is a verified connection. The graph grows as new touchpoints are observed.
- Profile unification — The identity graph resolves into a single canonical profile per person or organization, with field-level confidence scores and source attribution.
Identity resolution vs entity resolution
Entity resolution operates within a single database or system. It answers the question: "Do these two CRM records refer to the same person?" The scope is bounded — you're comparing records you already have.
Identity resolution operates across multiple systems, channels, and time periods. It answers a broader question: "Is the person who filled out this form the same person in our CRM, who visited our website last month, and who our sales rep connected with on LinkedIn?" The scope is unbounded — you're stitching together fragments from everywhere.
Identity resolution is the superset. Entity resolution is one technique used within identity resolution, but identity resolution also incorporates cross-system linking, temporal matching (tracking identity across job changes), and external data integration.
How Salmon resolves identities
Salmon resolves identities across your CRM, marketing, and compliance systems. AI agents match across name variants, email addresses, social profiles, and company affiliations to build a single verified identity — with confidence scores and source attribution on every link.
Why identity resolution matters
Without identity resolution, organizations operate with a fragmented view of their customers and prospects:
- Fragmented customer view — The same person exists as multiple records across CRM, marketing automation, and support systems. No one sees the full picture.
- Duplicate outreach — Sales and marketing contact the same person from different records, creating a disjointed and unprofessional experience.
- Inconsistent compliance records — When identity fragments aren't linked, KYC/KYB verification has gaps and audit trails are incomplete.
- Unreliable attribution — Marketing can't accurately attribute conversions when the same person has multiple identities across systems.
With identity resolution, teams get a unified 360-degree view, accurate targeting, complete audit trails, and reliable attribution — all built on a foundation of verified, linked identity data.
Related concepts
See real-time enrichment on your data.
Send us a sample from your CRM. We'll show you what Salmon enriches, verifies, and fixes — live, in 30 minutes.