The Importance of Data Analytics

problem-449364_640

The current obsession with predictive analytics and big data has many organizations fixated on “bottom-up” analysis. Hype has resulted in business leaders thinking there are huge nuggets to be mined from their data lakes—if only they can find the right tools.

Time and time again, I see this mindset result in disappointment and disillusionment with the whole idea of data analytics and data-driven decision-making.

Some of the myths about big data analytics are:

  • existing data is a source of rich insights
  • Artificial Intelligence can make sense of any data
  • size equals quality
  • data analytics can replace strategy

The Big Data Analytics, Business Objective Correlation

Vendors and pundits have vested interests in making data analytics seem like a panacea for all modern organizational woes. But, in the long term, such overselling will turn people off. And that’s a pity as predictive analytics have much to offer.

Part of the problem, is that many of the showpiece big data case studies are presented as bottom-up analyses that produced profound insights. Stories coming from organizations like Amazon, Facebook, Netflix and Google are indeed compelling. However, these are companies that are largely built on data. There is going to be a tighter correlation between data and business objectives in companies like these. Analysis of raw data is closer to business strategy in these companies, than in their more traditional counterparts.

As the data collected in digital-native companies is strongly connected to their core business objectives, a direct analysis of that data is going to almost trivially result in actionable insights.

The data collected by most companies tends to be fairly haphazard. Apart from a few lovingly-crafted relational databases, most of the data held by companies is digital exhaust—a by-product of doing business. To think that there’s a wealth of strategic insights hiding in stored data is wishful thinking.

Data analytics is not a replacement for good old-fashioned management science. We still need to have strategy, an understanding of where we want to go, and some idea of how we might get there. Data-analytics can significantly magnify and improve strategic planning and analysis.

We no longer have to rely on gut-instinct. With data analytics, we can test our strategic theories. Real-time analysis lets us monitor our strategy, so we can quickly pivot when we are shown to be wrong by the facts, or when the assumptions underpinning our strategy have changed.

An effective data analytics approach will be strategy-led. We need to find and collect data that can guide our decision-making—not assume that running on a few tools over any old data can replace quality thinking.

Don’t think I’m arguing for a return to the annual planning retreat and personal biases masquerading as “bold vision.” Far from it. Data-driven decision-making and strategic hypothesis testing are powerful approaches. Any organization that fails to adopt them will rapidly fall behind.

What I am advocating is that we make sure we avoid overselling the potential of predictive analytics, so that we don’t end up throwing the baby out with the bathwater.

If you are interested in data analytics, Learning Tree has a number of courses you may wish to consider, including:

 

Andrew Tait

Type to search blog.learningtree.com

Do you mean "" ?

Sorry, no results were found for your query.

Please check your spelling and try your search again.