A Semantic Graph Unlocks HyperIntelligence
Analytics, with a few exceptions, derives its data from other sources: this is “used” data. It is used in the sense that it was captured and stored for some other purpose than the business at hand—sales data, human resource data, maintenance data, and more.
Analytics, with a few exceptions, derives its data from other sources: this is “used” data. It is used in the sense that it was captured and stored for some other purpose than the business at hand—sales data, human resource data, maintenance data, and more.
Historically, this “used” data has been mostly hand-me-downs within the organization—something we now call “operational exhaust.” But for reasons everyone is familiar with—digital transformation, cloud, big data, massive computing resources—it is often referred to as “digital exhaust.”
Early data warehouse practitioners learned a painful lesson: even internal and structured data can be challenging to understand and work with based on just technical metadata such as table name, column name, and datatype. Everything is considerably more complicated than that.
The creators of MicroStrategy understood this when no one else did. Finding the latent value in used data was only possible by creating an abstraction layer between the data sources and the MicroStrategy engine. I didn’t discover this on my own; MicroStrategy introduced me to the idea more than 25 years ago. In 1996, MicroStrategy contracted with me to develop a definitive document, “The Case for Relational ROLAP.” If you are interested in the original concept, my old friend and former MicroStrategy CTO Jeff Bedell filed the patent (US701051881) for it in 2001, granted in 2006. You can look it up if you like athttp://patft.uspto.gov/.
In 2019, MicroStrategy took the semantic “layer” to the next level: the semantic graph. I like the way Saurabh Abhyankar described it at MicroStrategy World 2019 as a set of features:
Of particular interest is how the “Last 30 Years” provided not only semantic maps for the data itself, but many other important artifacts of the analytical environment. When you look at the list and get below “derived metrics,” everything else is there to manage the entire stack, and 2019 extends that to dossiers, usage, synonyms, collaboration, and even HyperIntelligence™ Cards. The future looks pretty impressive.
Today, getting the models together relies on at least some of the business users understanding the location, naming conventions, semantics of the data, if not the intricacies of the crafting of the analytical derivatives such as metrics, hierarchies, grants, and workflows. This is a considerable barrier to progress.
Business people need to define their work in their terms. A business modeling environment is necessary for designing and maintaining structures. It is especially important to have business modeling for the inevitable changes in those structures.
