How To Define The “R” In ROI

Last week, I outlined the steps to building a big data strategy. Included in Step One, Chart Your Vision, is the notion that you need to define your “ideal” customer experience and then identify which business questions must be answered so you can deliver on that ideal. Chief among those business questions, of course, should be a discussion about ROI.

Before proceeding with any big data strategy, you need to know:

What will be the return on investment (ROI) from this initiative?

Interestingly enough, determining the “I,” the investment, is usually not too onerous. (Well, I’ll admit, it may be considered onerous by those paying the bills. But what I mean here is that the calculation of required investment is typically rather straightforward.)

No, it’s not the “I” that’s difficult. What’s difficult – and what seems to be getting more and more complex by the day – is calculating the “R.” How should you assess the “return”? And even before you get there, how do you determine what kind of “return” you should be assessing?

After all, when it comes to big data strategies, the returns can take multiple forms, and knowing which one demands your attention can be somewhat tricky. Let me explain.

For starters, I like to group the potential returns into two broad categories: quantitative and qualitative.

Quantitative returns. These returns are the hard data, the facts that lead to analysis and insights. For example, when Ace Hardware implemented its Teradata Ace Data Warehouse (ADW) system, traditional constraints on new data sources and reporting workloads were immediately relieved. 2,500 stores now transmit daily POS transaction detail to the warehouse, and the query execution times fell from days to hours or minutes, even as hundreds of new users began accessing the warehouse. The ability to manage and report on that data has given Ace’s merchandising, marketing and operations departments a vastly better understanding of what’s working at the retail level, and how Ace retailers as a whole can benefit. As a result, the average value of retail transactions has increased, the number of same-customer store visits is up, and all nine Ace product departments have recorded retail profit margin increases.

Qualitative returns. These returns cannot be easily measured – but they can have a BIG impact nonetheless. As I mentioned back in November, big data analytics can help mitigate the impacts of natural disasters and other emergencies, like hurricane Sandy. It is also playing an increasingly important role in medical research and the delivery of healthcare. As I like to say, big data is an opportunity to learn, to innovate and to create a better world – and not all of those benefits are quantifiable. (Watch me address some of these issues in this short video clip.

Yes, big data has many, many complexities, and each strategy must be evaluated individually, with a balanced approach that includes both quantitative and qualitative aspects. You need to start with your vision and the “R” when creating a big data plan and building a data-driven enterprise.