“Context” is the new buzz-word for data.  Jeffery Hammond talks about it in Systems Of Automation Will Enrich Customer Engagement, Robert Scoble and Shel Israel talk about it in their book “Age of Context”, and you can’t ignore it when it comes to a discussion for Cognitive Computing and the Internet of Things.  We’ve live in a world where data was rationalized, structured, and put into standardized single definition models.  The world was logical.  Today, we live in a world where the digital revolution has introduced context, the semantic language of data, and it has disrupted how we manage data. 

Big data technologies were created not because of volume and cost.  They were created to manage the multi-faceted model that data takes on when you have to link it to how regular consumers and business people see the world.  Performance and cost are only factors that had to be considered to scale in order to support the objective.  Search, recommendations, personalized web experiences, and next best action could not be possible in a structured single definition environment.  Why we know this is that the sculpted purpose built environments that supporting business applications collapsed when analytics to discover causation in relationships and correlations at scale was applied.

That is the tipping point for data architects.

Data architects can’t only concern themselves with understanding the technologies that are available in terms of volume, velocity, variability and variety.  Big data technology is not only there to lower cost to capture and store vast amounts of data.  Simply cataloging what data comes in, the data elements, where it is located in a Hadoop cluster, and the logical model required is not enough. The reason, you need to understand context when working with data. 

How you obtain context is by actually working with the data to divine the insights.  This is where the logical work is translated to the semantic world.  If you don’t use the data, if you don’t think about the data from the perspective of the questions or views you need, you can’t build an architecture to support it.  Even with the best efforts to collect business requirements, your data architecture will fail if context is not only understood but infused in execution.

Data architects need to walk in the shoes of an analyst.  Data architecture is no longer only about the technology you implement, it is about creating solutions for analysts and consumers of data.  If you can’t think like an analyst or business user, you can’t know what they need.  It is time to get educated on using data vs. educating the business on data technologies.  The best way to do this is to experience what it takes to be a data scientist, business or operations intelligence analyst, or customer analyst.

Without the ability to cross easily between logical and semantic models, data architects remain in the realm of developers and not strategic resources for the business.