Most databases have data arranged in rows and columns, much like a spreadsheet. That is what we usually have in mind when we start to do analysis. However, there is another type of database where the data is arranged as nodes connected to one another. It’s called a graph database. If you think of a social network with people connected to one another, that’s an example of a graph.
The term “graph” is rather unfortunate, since to most people it means a chart; however, in mathematics, a network of nodes is called a graph and the lines connecting the nodes are called edges. It’s a terminology we should get comfortable with since graph databases are a powerful way to organize data.
Many HR professionals are familiar with organizational network analysis (also called social network analysis) where you see who connects to whom in the organization. The people doing the analysis may not have mentioned it, but they were probably using a graph database to do the analysis.
While graphs databases are not new, they are worth paying attention to now for several reasons.
One is that Workday, one of the dominant players in HR tech, is using graph databases as part of their architecture. This gives it credibility. The second reason is that there are more off-the-shelf tools to analyze graph databases. For example, Neo4J has a whole suite of tools that will make analyzing graph databases much easier. Finally, we now have a group of people in most organizations, the HR analytics team, with the skills and curiosity to put graph databases to use.
While we don’t want to get super nerdy, there are programming languages that use what is known as object-oriented programming. Conceptually these have some similarities to graph databases in that they involve connecting “objects” together. Point is, since a graph database is a fairly technical subject it might well be worth an analytics professional’s time to take a course. There is a free one at Coursera and no doubt many other options, as well.
While there may be no urgent need to apply graph databases at your organization, this is a fundamental technology — and analytics pros should have an understanding of it so that they will recognize the opportunity to apply it when the time comes.