By Mark Ramsey, PhD
SVP, R&D Chief Data and Analytics Officer
GlaxoSmithKline

 

 

For years, organizations have discussed their evolution to become data-driven.  Some organizations, such as Google, Facebook or LinkedIn, were founded on information and data-driven decisions.  For the rest, the transformation has been painfully slow or even unachievable.  According to the recent NewVantage Partners’ 2019 Big Data and AI Executive Survey, the percentage of firms identifying themselves as being data-driven has decreased during each of the past 3 years – from 37.1% in 2017 to 32.4% in 2018 to 31.0% this year.  Oddly, over 90% of the companies felt that the pace of their big data and AI efforts is accelerating.  How can this be the case?

In assessing the dilemma, Randy Bean & Tom Davenport determined that one of the key drivers in the decrease is business adoption.  The big data and AI efforts are accelerating, but with the lack of business adoption, leaders are not moving from traditional to the desired data-driven approach.  This then leads to the question of why the struggle for business adoption?

To transform the key decisions of an organization to be data-driven, means that the big data and AI programs must directly align with the overall business strategy.  Unfortunately, many of the big data and AI programs are focused on small subsets of the business that, in the end, do not have a significant impact.  In many cases, the driving factor for the programs has been the availability of data, the ease of rationalization, and an eye toward – “Start small and grow over time”.  For example, had the founders of Google started with a focus on implementing a common data model that had to be adopted by all websites in the world to simplify data rationalization, it is doubtful the company would be generating 150+ billion in only 15 years.  Instead, Google crawled all of the websites, ingested the data, and used advanced analytics to rationalize and drive insights across billions of data sources.  Organizations tend to shy away from large transformation programs, but this results in a failure to deliver the business impact.

Becoming a data-driven organization starts with ALL of the data.  It does not start with merely a small slice of data that is easy to extract from an operational system, load into a big data platform and run AI against it.  It means creating a modern data management platform that supports rationalizing across the internal data of the organization and integrating powerful external data.  Treating all data as a strategic asset for the organization is key.  If the NewVantage Partners’ survey would ask organizations how they were approaching the data readiness for their programs, I expect an alarming number are applying data approaches launched in the mid-1980s – ETL, master data, enterprise models, etc.  To be successful, organizations need to apply current technologies, such as AI, machine learning, pipelines, bots, etc. in the creation of a modern data management platform.

Having a solid, modern data management platform, loaded with the internal and external data is the foundation.  Then, data-driven programs are built in direct alignment with the key strategic initiatives of the organization.  Many factors contribute to the success or failure of a big data and AI program; however, those organizations that consider themselves data-driven will most likely hover around 30% until all data is treated as a strategic asset.