Thursday, 17 September 2015

Keeping track, losing track: when data and human endeavour collide

"Three quarters of companies are investing in big data, but the ROI remains elusive" is the headline of Colin Barker's ZDNet article yesterday.  Writing about Survey Analysis: Practical Challenges Mount as Big Data Moves to Mainstream, covering a June 2015 Gartner survey of 437 large organisations, Barker relates of ROI ("return on investment") that
When asked about the ROI expected from big data, the majority of those companies with plans to invest in big data and those that have invested expect an unspecified, positive ROI. However, a large proportion of companies (43 per cent of those planning to invest and 38 per cent of those that have already invested) don't know if their ROI will be positive or negative.

"This uncertainty highlights the challenges in determining the value of big data projects," the report said.
While there are many possible explanations as to why ROI should remain elusive or uncertain -- and maybe even a different explanation in each instance -- and without prejudice to the fact that Big Data means different things to different people, there is one possibility that suggests itself. 

Like Bigfoot, ROI can be
elusive -- but something
can be done about it.
Identification of the sort of data one wants is relatively easy. Assuming that the data sought is actually out there, can be harvested and is capable of being brought home in an intelligible format, acquiring it shouldn't be too much of a problem either. But once it is accessible, is the organisation that has invested in it geared up to be able to make effective use of it? Does the acquisition of such data involve different gearing of corporate decision-making processes, different analytic routines, or new procedures for signing off of expenditure arising from its use? How is it curated and distributed within the organisation and are steps in place to deal with its obsolescence and subsequent refreshment or replacement where necessary?  The absence of new routines for taking advantage of Big Data, or unfamiliarity with them, can easily lead to a diminution of its value.

Thinking about it, an analogous issue arises with regard to the very opposite of Big Data: small and individualised items of information of the sort that, when amassed and commoditised, collectively form databases from which Big Data may be collated.  This was apparent from a discussion that took place in a workshop today in Vienna at the annual conference of MARQUES, the European trade mark organisation. 

The workshop, "M&As and restructuring: brand challenges", led by the organisation's Intellectual Asset Management Team. took a fascinating turn when Dieuwerke van der Schalk (Legal Manager, Global Brands, Jacobs Douwe Egberts) and Nunzia Varricchio (Senior Trade Mark Counsel, DSM) spoke of problems that can occur during the M&A process when decisions taken at the top of the corporate tree cause havoc with day-to-day management of IP portfolios because no system, or only an inadequate one, is put in place for distributing, processing and acting upon those decisions. In result of this, data from individual patent, trade mark or design files can end up floating freely around, being effectively lost.

The workshop threw up plenty of examples of things that can go wrong when there is no internal scheme and structure for dealing with information. Obligations of confidentiality that are absolutely necessary during negotiations where an IP portfolio is at the heart of a transaction may only get in the way once knowledge of the deal becomes public and it has to be put into effect; uncertainty as to which side does what when it comes to due diligence, compliance with record-keeping formalities and the like may cause problems too; records of licences and assignments may disappear into data rooms or even be archived just when it is imperative to gain access to them for routine IP right maintenance; and in-house IP departments may be under impossible pressure of time and resources when required to become involved in executing at ground level the decision made up at the top of the tree.  

To put it another, there is all the difference between deciding to buy a portfolio of 10,000 registered IP rights and sitting down to sort out all the details of 10,000 separate items of property, with their respective renewal dates, territorial limitations, caveats, litigation track records and so on.

So the moral of the story is this: whether it's Big Data or little data, your ROI will be enhanced if you can set up a realistic system to keep an eye on how your information is processed and used.

1 comment:

  1. There's another point to consider here: a business that is on the receiving end of an unexpected takeover or merger bid is unlikely to have a Big Data or information management policy in place that is geared to addressing it.