What’s so ‘Modern’ about Modern Data Management?
As someone who has been in the master data management (MDM) space since 2002, I thought I had seen it all, understood all the customer requirements that could or could not be met with an MDM solution.
But I was wrong. Reltio’s Modern Data Management Platform solves an entirely new class of operational problems, as well as solves the existing operational challenges in a faster, flexible and more scalable manner.
I spent a number of years implementing operational MDM solutions. There were several areas of the implementation that slowed everything down. Fast time to value? Only if you didn’t have to meet any business requirements as different Lines of Business (LOBs) had different requirements about how to curate and use the data. But traditional MDM solutions force you to have one common view. What happens if your LOB needs different - a more personalized and contextual view? Get in a room and hash it out? Good luck. And while the project snowballs, you can kiss your twelve week implementation goodbye!
Another area that slowed the implementation down was the data load. This was caused by two problems. First, you had to map the data to the data model. Even if you had an ‘industry best practice’ data model, this was a difficult task. No one knew the data model well enough. Oh! Don’t forget to map the foreign keys and intersection tables! The data validation rules that were not built into the database, remember the ones at the application layer? I’m sure you know all those by heart. Because they will cause your records to fail, too. Second was the quality of the incoming data. Data quality tools can help you, particularly profilers. But you still need an analyst to sort through the problem, determine solution and implement it. And you might find thousands of types of data issues that affect millions of records. Good luck. And again, your time to value becomes years, not weeks!
A Modern Data Management platform handles relationships in a modern way. Relational databases handle relationships by way of foreign keys, joins and intersection tables. Hopefully, the application server logic correctly commits the data so that referential integrity of the database does not get compromised. That can be pretty tricky if you try that across applications (which is why most companies don’t). This approach is fine until something changes. Oh, your business requirements changed? You need to relate entities in ways that had not been thought of in the initial implementation? Sure. Step right up to consulting hell as someone tries to figure out how to undo what has been done, get the new relationships in place and put together a plan to get from the existing state to the new one. Oh, you wanted that next week? Or yesterday? Hmmmm…. We’ll get back to you on that.
One of the key advantages of the Reltio platform is that instead of using one or two technologies to solve the operational master data management problem, it uses a number of them. Highly scalable database? Check. Search available immediately when data is loaded? Check. Reltio Graph to manage all kinds of relationships dynamically? Check. Data Quality and Survivorship? Check. Am I forgetting something? It’s all there. Without the need to put together the pieces yourself and worrying about messy installs or upgrades, all running on the scalable and security of the AWS platform. In addition, Reltio platform leverages machine learning to improve data quality and enrich data with relevant insights.
Hadoop, HBase, Cassandra, Graph Databases, MapReduce, Spark and the continuous stream of new technologies have changed the way data can and should be managed. However, stitching together all of the pieces required to have a complete end-to-end offering and support a wide variety of business needs across an enterprise is a complex undertaking. A Modern Data Management Platform helps you keep up with this evolving technology landscape and allow you to solve all your operational (as well as analytical) challenges faster and in a more agile manner.