The concept of big data and business success is nothing new. Companies with high data adoption rates tend to have greater productivity than their competitors. So, what’s unique about those companies which find success around the application of Big Data?  The answer is data literacy.  

Data alone is not a competitive advantage.  It is the specific ways in which businesses embrace and exploit data which is key.  There have been countless examples of companies over the last 20 years with access to huge amounts of data which still did not succeed —  MySpace being one of the most famous examples! 

Why Data Literacy?

The problem enterprises face today is that as the number of analytical problems grow exponentially a Data Scientist’s capacity grows — at best —  linearly.  In 2020 we analyzed the mid-term supply of data science graduates through new degree programs in Germany  and compared that to the estimated demand for data scientists over the next 5 years.  The results showed an expected shortage of more than 50.000 employees.  

The point is that there will never be enough Data Scientists to solve all of an enterprise’s data-driven problems. Companies with high data adoption rates solve for this by flattening the data literacy curve. 

Data literacy is the ability to read and communicate data in context.  That means   understanding how data is stored, analyzed and applied.   It also means describing data use cases and its value. 

Many companies rely on the economic value of their data assets. Revenue generated through data assets takes on greater significance with each passing quarter.  

As data and analytics take on greater importance in business so, too, does a data-literate workforce.  Building a data-literate workforce, however, is easier said than done. 

Achieving Data Literacy

A data scientist’s expected level of knowledge usually comprises a fully grown portfolio of technical, analytical and mathematical skills. Data literacy is more light weight as a concept. It focuses  on promoting general analytical understanding and defining relevant data skills within the day-to-day context of the user.

There are two basic  approaches to how enterprises apply or construct data literacy programs.  The first is an open approach and the second is a closed approach.

The open approach borrows conceptually from Bertelsmann University’s  “Data Curriculum” which provides all employees with a universal training program in the field of Data. It focuses on six data “roles”  ranging from data scientist to business partner. These roles all tend to be found within most B2B businesses. 

This “roles approach”  is centered around the idea that everybody needs to have some level  of data knowledge. It allows for universal access to a baseline data skill set.  It covers a range of activities across an entire ‘data-to-business’ process, i.e. from data extraction and exploration to analysis and operation. And it covers the communication of results.

The alternative approach is a “closed-shop” program. With this approach every company develops a walled-garden version of their  own data qualification program.  The closed approach tailors programs specifically  to the needs of a given company and/or  the needs of their product universe.  

There is merit in both approaches.  The closed approach, though, tends to have an adverse effect on universal data literacy by negatively  impacting data adoption rates.  This, in turn, depresses data literate workforce numbers.  

Enterprises with high data literacy rates also tend to build programs in small, digestible pieces. People usually shy away from complex scientific matters.  Successful data programs are broken into smaller, digestible pieces and then presented contextually.  This allows employees to put data science skills into real-world  business environments.  Making sure data literacy programs have a close connection to the things people do in their day-to-day business lives (and which are  actually helpful!) results in a far greater degree of  adoption and retention. 

More and more companies are also building out products using a  data driven approach.  This makes the products adopt data science.   Machine learning algorithms enhance legacy tools and processes.  They automatically detect critical business issues instead of relying on manual analyses. In this way big data methodologies make their way into an organization without users even noticing.

With data literacy, though, it always has to start at the top.  A business leader’s  awareness of data capabilities or data science use cases creates top down demand within an organization. You can’t delegate data knowledge.  While not everyone needs to become a data scientist, everyone — even the CEO — needs to do his/her homework and acquire the required level of data knowledge to define the right path for an organization

However a business ultimately gets there, data literacy is vital. When data is truly adopted it can have an amazing impact on productivity and business success.  Adoption requires literacy.  The best way to achieve literacy  is through a universally tailored “roles-based” data program that is digestible, contextual and relevant.