Rodrigo Garcia
Ask any Fortune 500 executive in the retail industry of the problem that keeps them up at night, and their answers will likely strike the same chord: How to extract value from the increasing volume of data their company collects.
When the Boston Consulting Group asked senior executives which areas of innovation would have the most impact in their industry in the short-term future, Big Data came in second, only behind the speed of adopting new technologies.
An Accenture survey of executives demonstrated their belief in the transformational power of Big Data, with 89% of them agreeing or strongly agreeing that it will revolutionize the way they do business much like the advent of the Internet has since the early 90s.
These data-related struggles are not unique to the retail industry and only promise to continue as businesses undergo the digital transformation that will create even greater volumes of data and more need for efficient analysis.
Another survey by Vanson Bourne for CA Technologies of IT and business executives highlights the challenges posed by digital transformation. In addition to the obvious volume of data, the data quality, siloed data, technical infrastructure, governance and security, and in-house expertise are all areas of concern.
While all this may sound familiar, it is cold comfort for those needing to implement Big Data solutions within their enterprise. It can all feel overwhelming, so rather than addressing Big Data as a problem that must be solved, think in terms of business problems that can be solved with Big Data.
Here are four things to consider:
1) Identify Your Big Data Goals
Have clear goals in mind about what business problems you need to solve and then identify the data you need to address that problem. This will cut your scope down from a massive, “Where To Start?” challenge to a manageable, doable project.
2) Crash The Silos
Don’t create a separate Big Data division or operating unit.
Think about how major retailers addressed online shopping by creating ecommerce operations that were separate and distinct from overall business operations. They are dealing with the consequences to this day in the form of unnecessary barriers to CRM systems and logistics, to cite merely two examples.
Instead, plan to incorporate and integrate Big Data solutions into your existing operations at the outset so you avoid potential future digital gaps.
3) Set Expectations
Big Data appears to have fallen into the Trough of Disillusionment and is entering the Slope of Enlightenment in Gartner’s Hype Cycle.
But the good news is that people across the organizations are starting to realize that Big Data is not going to be cheap, easy and fast. It is necessary, but hard because Big Data is here to stay.
4) Choose A Pilot Project
Selling a Big Data project can be tough. The very phrase emphasizes the word BIG, which sounds to many ears like complicated and expensive.
Start small by identifying a project with which you can prove the value of Big Data analytics. You’ll want a manageable project that doesn’t require a massive dataset (nor a massive investment) in order to prove your concept.
Chose a project for which you have a high confidence in demonstrating value that you can then extrapolate that value across the enterprise and for an entire year.
Frame your project as the floor of potential from which you can make the argument that you can do much more sophisticated analysis. And given the experience gained from the pilot project, you would have a greater understanding of the data and therefore a higher confidence in your overall estimate.
