Big Data failure

Blogs
(Image: StockFresh)

17 November 2014

Paul HearnsA colleague drew my attention this month to a somewhat hysterical piece by a fairly respectable academic asking if the hype around Big Data is about to implode as one its poster children runs into major market woes.

Michael Schrage is a research fellow at MIT Sloan School’s Centre for Digital Business, the author of several books and a commentator on digital business in general.

In a blog post for the Harvard Business Review (tinyurl.com/ka9podl), Schrage wrote about Tesco’s current travails and asks if the analytics advantage from its club card scheme was actually an advantage at all, and if the retailer’s failures were due in part to over reliance on data that wasn’t all that valuable.

Schrage suggests, as opposed to argues, that “Tesco’s decline present [sic] a clear and unambiguous warning that even rich and data-rich loyalty programs and analytics capabilities can’t stave off the competitive advantage of slightly lower prices and a simpler shopping experience. Better insights, loyalty and promotion may not be worthless, but they are demonstrably worth less in this retail environment.”

He started that paragraph with the rather Helen Lovejoy stance of: “How damning; how daunting; how disturbing for any and every serious data-driven enterprise and marketer.”

This was in reference to Tesco’s failures, despite access to all of this rich data for which it has been held up as the best practice example for other data-driven enterprises.

Schrage does acknowledge the possibility that a lack of creative thinking, competitive drive and further investment may actually have meant that Tesco failed to capitalise on the data and insights from its loyalty card scheme, but he also goes on to say “But when we’re talking about customer data, insight, loyalty and all the ingredients that — supposedly — go into giving digital enterprises their information edge, then it’s time to get nervous and ask hard questions.”

I have to say that after reading the piece a few times, because my initial reactions were incredulity mixed with no small amount of confusion, I was moving toward bewilderment. Schrage seems to have spectacularly missed the point.

The overwhelming import from the entire thing, even as Schrage outlines it, is that all the data in the world will not save you if you haven’t a bloody clue what to do with it.

Far from saying that the Tesco debacle is a warning to data-driven enterprise everywhere, the impression is that if you don’t know what you are doing, it doesn’t matter what information is at your disposal

Far from saying that the Tesco debacle is a warning to data-driven enterprise everywhere, the impression is that if you don’t know what you are doing, it doesn’t matter what information is at your disposal.

I’m reminded of the words of Kaimar Karu of Axelos (see the News section) who warned of a culture within organisations that falls to HIPPO Syndrome — the Highest Paid Person’s Opinion. Looking into the wider context, Tesco’s woes seem to have come from the top down, as successive CEOs have felt the need to comment on previous regimes, and not in a good way.

When added to all of this is an outline of a reliance on loyalty rewards before pricing as the key driver for custom, Schrage again fails to make a case for the data-driven strategy as a source of Tesco’s woes.

He goes on to cite various sources making much of Tesco’s falling market share, the rise of the discounters in the space and a customer backlash as generous benefits for the loyalty scheme were reduced, but again fails to actually make any argument for the data failing.

The lesson I drew from all of this, is that even a market innovator, with massive resources and a cash rich environment, can take their eye off the ball and lose it all if they are not careful.

Forgetting the particular pressures of the supermarket retailers for a moment, what happened here is a fairly familiar story in the tech world in particular: a spectacular market advantage gained through the very clever user of emerging technology is lost as levels of development, investment and management are reduced. The list of companies that have fallen into this trap is not only long, but also reads like the sign-posts  of the road of acquisitions.

So first of all, let us dispel Schrage’s rather hysterical suggestion: there is no great threat to the data-driven enterprise idea. Having spoken recently to HP, Oracle, Sumo Logic, IBM, EMC and others, all of whom tout various customers that have successfully implemented data-driven and analytics initiatives to derive intelligence from operational and other data, the value of such programmes is unquestionable.

The fact that Tesco has made a complete mess of it is despite, not as result of, its access to all that data.

The tools to source, aggregate, analyse and visualise data have never been better, more accessible, affordable or easier to implement. The intelligence derived is unequivocal. However, the cautionary tale is still the same — if you don’t do something smart with it, it’s all for nought.

There is an old, somewhat vulgar, variation of the leading a horse to water proverb, usually attributed to Dorothy Parker, which we’ll adapt slightly here, as it is no less illuminating. Let’s say you can lead to a horse to culture but you can’t make it think.

A data-driven enterprise can have all of the analytics insights in the world presented to it, but if it fails to understand how those insights can inform decision making to drive competitive advantage then it makes little difference.

Data-driven analytics is just another tool to help decision making, but if the people, processes and structures are failing around those tool sets, what hope is there? None would appear to be the answer from the Tesco experience. But that does not stop Amazon, Netflix, Avis, Roku and a whole host of other businesses from using insights every day from analytics to give them competitive advantage in their respective markets.

So, don’t panic: there’s no need for someone to please think of the children, and there is especially no need to reconsider existing or future plans for a Big Data project. You just might want to remove Tesco as a supporting argument for your big pitch, lest your executives fail, as Schrage has done, to see the wood for the trees.

 

 

Read More:


Back to Top ↑

TechCentral.ie