Join top executives in San Francisco on July 11-12, to hear how leaders are integrating and optimizing AI investments for success. Learn More
In an era where data is more distributed than ever before across different applications and modalities, the need for master data management (MDM) is critically important to ensure data accuracy and quality. MDM is a foundational type of data technology, providing a “golden master,” or uniform data master record for every entry.
Among the leading vendors in the space is Reltio, which has developed a cloud-native MDM platform. Reltio raised $120 million in funding in November 2021 and has steadily expanded in the months since then, including adding data quality measurement capabilities in June 2022. The company has increasingly added artificial intelligence (AI) and machine learning (ML) powered enhancements to its MDM platform as well, helping organizations better manage and master their data.
Today, Reltio is further expanding its MDM capabilities with the launch of what it calls “velocity packs” that package up configuration and best practices for MDM, designed for specific industry verticals. The first pair of verticals are life sciences and healthcare, with more to come in the upcoming months. The velocity packs are in part a response to the current economic climate where organizations are under pressure to demonstrate value from data faster.
“There is a larger focus now on what is the total cost of ownership and how things can be tied back to ROI type of outcomes,” Manish Sood, founder and CEO of Reltio told VentureBeat. “Where customers may have taken any data investment very easily before, now they’re focused on how it will get them to value in a shorter time frame.”
Join us in San Francisco on July 11-12, where top executives will share how they have integrated and optimized AI investments for success and avoided common pitfalls.
The intersection of AI/ML and MDM
According to Sood, data in the modern world is an asset that is being continuously updated and it needs to be continuously curated as well.
A core part of keeping accurate and updated data, for any industry, is to have some form of data quality rules that help to ensure the freshness, lineage and accuracy of the data. Being able to automatically apply smart rules to MDM data quality is one area where Reltio has been investing its time and money. A significant amount of the effort falls into the realm of AI and ML.
Sood said his firm has been focusing on applied AI for managing the data. There are several areas in particular where Reltio is using AI today, including entity resolution. He noted that the MDM system is taking data from multiple sources, and there can be slight variations across the sources in some of the field entries; for example, with how a name is spelled or how a product is categorized. Entity resolution helps to solve that issue.
“We’re using AI algorithms to bridge that gap and understand that these are the same type of details that are being provided for the same person, product or supplier type of information, and we can create a unified record,” Sood said.
The AI capabilities developed by Reltio can also go a step further and help with data cleansing for deduplication, as well as enrichment to add context.
With the new velocity packs, the AI capabilities have been specifically tuned to enable the right data formats and information for a given industry vertical. Sood explained that the velocity packs also integrate canonical schemas defined for the specific data domains.
The graph model at the core of Reltio MDM
While AI is important, underlying the Reltio MDM platform is a graph data model that is the foundation for the platform.
Sood said that the graph data model that his firm has built is actually a hybrid approach that integrates both columnar and graph database capabilities.
“The simple way of thinking about it is that the entities are the key constructs that are defined with the infinite attribution, which comes from the columnar capabilities, as well as any kind of relationships or linkages between different entities,” Sood explained.
He added that Reltio uses a polyglot persistence model for storage, where older data can be stored in secondary and tertiary storage to help reduce latency for the most active data, while still enabling analytical queries for historical data.
Overall, the goal for Reltio is to use its graph data model and data science expertise to help organizations recognize the value of data faster, Sood said. He emphasized that the new velocity packs are a core part of achieving that goal.
“Everybody so far to this point in time has sort of assumed that if we have better data we will have better outcomes,” Sood said. “But now it’s time to put that to test and make sure that we can deliver those tangible outcomes.”
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.