In this article:

Crossbeam identifies overlaps between company data sets, which raises a key question: when are two records a match? That’s where our matching algorithm comes in.

The algorithm is guided by confidence. Because matches often trigger data sharing, Crossbeam requires very high confidence before declaring a match. False positives (incorrect matches) are far worse than false negatives, and the algorithm is designed accordingly.

We compare multiple record properties to calculate a confidence score, giving extra weight to unique attributes. Customers cannot modify the matching algorithm.

Here are the key data points used in matching.

Domain names are a source of high-confidence company matches, as no two companies can have the same domain. We run domain names through a standardization process to ensure that inconsistencies in formatting don't create false positives. We also maintain a growing awareness of cases where multiple domains are owned by the same company so that indirect matches can be made. Things to note about domain names:

Email addresses are a source of high confidence person matches, as no two people can have the same email address. These addresses are run through a similar cleansing and standardization process as domains. Emails also have a bonus benefit of helping with company match resolution, as we can often determine that companies match based on them having matching people. When certain quality conditions are met, we can also use the domain name of contacts as a matching property for companies. 

DUNS (Data Universal Numbering System) number is a globally recognized identifier that strengthens our confidence in matches, improving accuracy and consistency for our customers.

How DUNS Impacts Matching:

Phone number matching identifies overlaps between accounts that share the same valid phone number—even when their domains don’t match. This ensures you capture connections that would otherwise be missed in traditional domain-based matching. Crossbeam automatically normalizes, validates, and deduplicates phone numbers using international standards, removing invalid or test data to deliver accurate, high-confidence matches.

Brand matching enhances the Crossbeam Matching Engine by identifying overlaps between accounts that share the same brand name—even when their domains differ. This helps uncover connections across franchise networks, marketplace listings, and multi-brand organizations that traditional domain-based matching might miss. Crossbeam automatically parses, normalizes, and validates brand data from multiple sources—such as social profiles or marketplace URLs—to deliver accurate, high-confidence matches between related entities.

While individual properties like domain, email, DUNS, phone, and brand provide the foundation for high-confidence matches, real-world names alone are low-confidence. Crossbeam’s algorithm combines multiple dimensions to validate matches, ensuring accuracy while minimizing false positives.

The matching engine is continuously evolving, adding new fields and data points to improve match quality. Occasional minor shifts in match rates are normal and reflect improvements in the methodology.

How does DUNS matching work?

Read more about it here.

Can you match using other dimensions?

Not currently, no, but our matching engine is always getting more intelligent as we add more matching dimensions.