When we’re wrestling with complex data management problems, prioritization is key.  But durability is paramount.  The time to recover from a mistake or failure can strike a blow and sometimes even derail an otherwise well structured, executing program.  The two factors aren’t mutually exclusive, so why do we get caught in the chasm of indecision and what can we do to avoid choosing one to the detriment of the other?  Likely the best mitigation strategy here would be to promote re-usability and ensure that what is built is only built once and shared many times.

This is enabled through semantic construction and sharing, which are design ingredients of NexJ Customer Data Management (CDM).  For example, customer profitability requires mapping key customer attributes, costs and revenues to the sources in the organization, such as distributing a cost from the cost center to participating customers.  Next we define computation elements that are evaluated and stored for you. The calculation is stored in the model as it would be in Excel.  No extra work.  No manual labor.  No formulas to write.

Once you’ve defined your data requirements, mapped them to the appropriate data source, and defined calculations, you can use the built-in services to re-use that content, expose the metrics to your portal or even build some self-service reporting capabilities from it.  This would traditionally be a net new project but, with NexJ CDM, the build work is already done for you. For example, some managers may be interested to know the top 10 most profitable customers in their territory or region.  That knowledge can now be served to them from the same model via the corporate portal, embedded into an existing application, exposed to a self-service analysis tool, or distributed by other enterprise reporting methods.  And it will be the same content for all managers, the same cost calculation, and the same view.  No extra work.  No manual labor.  No formulas to write.

Advanced analytics is a very popular topic these days, for all the right reasons!  Historical data is a key ingredient in that process.  Great news: snapshot content is also captured and readily available from NexJ CDM.  So any changes that occur over time to the semantic model attributes are detected.  Think of this as seeing and preserving changes in an Excel spreadsheet: we save the current version, refresh with the new content, then save the new version. This detection method stores the historical version and the new version of those data elements. And these snapshots can be sent to a database, to a hadoop data store or exported to a flat file.  Now you have historical content ready for advanced analytics, predictive modeling or other mathematical algorithms. We can use this to detect patterns of changes in customer and profitability, which is one small step in understanding our customers’ behaviors.  No extra work.  No manual labor.  No formulas to write.

So we defined one semantic data model, like listing columns in an Excel spreadsheet. Next we defined our computations, like creating formula columns in an Excel spreadsheet. With one construction effort, we served the managers’ needs to see their top 10 profitable customers, supported their deeper questioning needs with a self-service report, and defined necessary data for the analytic team to predict changes in profitability.

So, what are you going to do with the extra labor, resources and time?