Customer Information Management
Modern Data Management Puts Relationships First To Meet Customer Expectations
by Aaron Wallace, Principal Product Manager, Customer Information Management, Pitney Bowes
Relational databases have been the go-to organizational structure for customer data since the early 1970s. They mirror the standard methods of data storage used well before the advent of business computing. This legacy model, which features neatly-stored information in relatively easy to navigate tables, made sense for most businesses up until the very recent past, given the easily comprehensible nature of this design.
Businesses today, however, are collecting greater quantities of customer data than ever before. Not only is there a higher volume of this information, but it is far more varied, diverse and complex.
In theory, this is great for the enterprise and the consumer, since an influx of complex data should empower businesses to better tailor services to meet the needs of a much more familiar customer base. Traditional relational databases, however, don’t have the ability to “think” like people do. Rather, the rows and columns in this database do little more than list data sets without mapping and interpreting the relationships between different data types.
With graph databases, relationships are everything. Rather than forcing the business to infer connections between different data sets, graph databases assemble simple abstractions of nodes and relationships into connected structures that facilitate the creation of sophisticated models and maps. Because this is how people process and understand data in their own minds, services therefore can more intuitively determine customer preference or intent.
For instance, this is how Netflix is able to pick out what movies users should watch next. The company’s databases are constantly collecting customer data, possessing the ability to interpret and remember user activity. The database organizes this information based upon learned indicators that determine what information is related, enabling Netflix to mirror and predict a user’s wants or intents.
This personalized user experience is now commonplace across the digital landscape, with companies like Amazon, Facebook, Google and LinkedIn all using graph databases to not only help collect massive volumes of data, but also put it to good use. As a result, customers across industries now expect an intuitive and intelligent digital experience, placing a great deal of pressure on businesses to deliver.
Compared to relational databases, the graph design gives businesses the tools to:
- Explore existing data for new insight and answers: Businesses can ask questions about data relationships that they might not even know they have.
- Gain a full view of master data, in context: They can retrieve data across systems, hierarchies and databases (e.g., product database to customer database) without having to change them.
- Conduct faster and more useful data queries: Companies receive better support for questions about data relationships with fast access to answers.
- Master data deduplication and unification: Potential duplicates are more easily identifiable by inferring relationships between instances of master data for true unification of data in a single system.
Pitney Bowes was participating in a webcast on March 3 that explored the merits of graph-based MDM. To learn more, you can also visit Pitney Bowes at the Gartner Enterprise Information & Master Data Management Summit from March 12-18 in Grapevine, Texas to dive deeper into the benefits of intelligent databases.