Entity Resolution Meets Graph Visualization

At The 2025 Gartner Data & Analytics Summit

At the 2025 Gartner Data & Analytics Summit, Linkurious and Senzing showcased their combined approach to data analysis in their joint presentation, Find the Best Customers & Worst Criminals Hidden in Your Data. Following their presentation, Scott Taylor (aka The Data Whisperer) sat down with Matthieu Besozzi, Head of North America at Linkurious, and Paco Nathan from Senzing to explore the Linkurious Decision Intelligence Platform with Senzing Inside™. The interview provides insights into how this strategic partnership creates a powerful solution to find hidden connections within an organization’s data.

Besozzi and Nathan explore the practical implications of entity resolution. Nathan explains that proper entity resolution connects records that represent the same person despite minor variations, distinguishing between similar but distinct entities. As Nathan puts it, “Senzing specializes in technology that finds the footprint of entities across different data sources.”

Besozzi illustrates the need to address this fundamental challenge in data management with a clear example using his name: “If you have ‘Mat’ with one t and ‘Matt’ with two t’s as different people in your dataset, but it’s actually one person who’s just misspelled, you have a big problem. You cannot have the whole context around the entity, the 360-degree view, which impacts your understanding of the customer.”

The conversation also covers applications from fraud detection to customer relationship management (CRM). In banking, for example, duplicate records prevent institutions from accurately identifying risk and opportunities.

Nathan shares an illuminating example: “I know of cases in banking where someone initially shows up as a red flag but turns out to be a really good customer. When you drill down using these tools, you can discover the real context.”

Besozzi explains that Linkurious selected Senzing after a thorough evaluation of the market. The partnership hinged on a number of key factors differentiating Senzing® entity resolution technology from competitors.

• Scalability emerged as a primary consideration. “Senzing is probably the only solution out there able to do entity resolution on massive datasets with billions of entities,” Besozzi states. “Scalability and speed are very important, and they do it quicker than the rest of the market.”
• Explainable entity resolution results are crucial for business application users. Entity resolution and data matching are most useful and generally only trusted when a user can quickly and easily find out why records match or don’t match, as well as the details that determine how entities come together over time.
• A nondestructive approach to data processing. “When you do entity resolution [with Senzing], you’re not erasing your existing data – you can go back. That is very important,” Besozzi emphasizes. 

The advantages of Senzing combined with Linkurious’ easy-to-use data platform to make entity resolution easily accessible for non-technical users. The integration allows for sophisticated data analysis capabilities where business users can define entity mapping criteria based on their domain requirements.

Understanding entities is foundational to modern data management. Without first resolving who’s who in your data-linking together records that refer to the same person, organization or asset – efforts to detect fraud, personalize customer experiences or apply advanced analytics fall short. The Linkurious-Senzing integration ensures that organizations can accurately identify and unify entities across fragmented, messy or inconsistent data sources.

But only resolving identities isn’t enough. For many use cases, particularly in fraud detection and investigative analytics, additional value comes from understanding not just individual entities but how they’re connected. Creating an entity resolved knowledge graph allows teams to decode complex relationships across data silos – revealing hidden patterns, networks and risks that would otherwise go undetected. This is where the combined strengths of Senzing and Linkurious truly shine.

The Linkurious-Senzing integration turns this foundation into actionable intelligence. Once entity data is accurately resolved and relationships are mapped, organizations can explore that information through intuitive graph visualizations. These visuals make it easier for teams – technical or not – to spot key connections, surface hidden risks and understand the broader context around an entity. Whether you’re identifying your highest-value customers or exposing fraud networks that evade traditional detection methods, the combination of entity resolution and graph analysis brings the full picture into focus.

The interview marked a milestone for the two companies with the first public demonstration of the end-to-end integration between Linkurious and Senzing. A full commercial rollout is scheduled for 2025.

Visit Linkurious to explore the Decision Intelligence Platform with Senzing Inside™ and discover how it can reveal hidden connections in your data.

For those who want to explore specific aspects of the interview, here are some key moments:

00:00 Introduction to Linkurious
Matthieu Besozzi explains the company’s mission to clarify complex data relationships through graph visualization.

02:23 Senzing Entity Resolution Approach
Paco Nathan describes how Senzing technology differentiates between similar entities while connecting records of the same entity.

03:45 Real-World Applications
Matthiew and Paco discuss how the integration supports both fraud detection and customer relationship management.

05:14 Industry-Specific Challenges
Examples are given from banking, supply chain management, healthcare and hospitality.

07:50 The Linkurious-Senzing Partnership Decision
Besozzi outlines how Senzing was the clear choice for scalability, explainability, data preservation and cost transparency.

09:03 Entity Resolution as Foundational for AI
Nathan explains why clean, well-structured data must precede successful AI applications.