Friday, 16 December 2011

Can data boost growth?

Francis Maude has today been quoted on Twitter that the Cabinet Office is busy talking to 250 companies in a bid to "identify data sets that will boost growth". Leaving aside how much what I know to be a rather over-pressured and under-resourced CO can do and how much knowledge it has, it rather begs the question, not of who they're talking to (though that should be open too) or even the data sets that might be identified (or the methodology by which that is arrived at - that ought to be open too) but rather whether indeed the sought after, mythologised even, 'growth' and the jobs and tax revenues that will allegedly follow can be acheived.

Even if you believe the Euro 40bn (once you've read the report you won't) much of any 'value' to be gleaned comes from savings to government from being the beneficiary of improved knowledge and decision making - valuable yes, growth no.

The big win to be had from big data is analytics; while sales and marketing are traditionally thought of as the areas where enterprises can leverage their customer data (with or without third party data sources such as data.gov.uk) it is frequently the case that such activities are about gaining competitive advantage from fresh insights that help deliver market share, better targeting, improved productivity and greater efficiency. That's also the 'excitement' over FB - being able to deliver specific messages to very specific market niches - doing more with less (though at a premium to untargeted ads).

And so it is with analytics across the board - doing more (or even the same) with less. In current economic times many are having to come to terms with doing less with less such are the level of cuts. And big data sure can help, but is it growth?

On the one hand we have a cry for government to up its game in the analytics field so it can do better itself. On the other hand we are told that "the best thing that will be done with your data will be done by somebody else". Further there is the ongoing potential of an entitlement culture undermining the freemium model often touted as the best chance for leveraging revenue from free or open data. Something like 81% of all apps downloaded are for games; $3000 is the average revenue per developer of apps. That circle doesn't square.

So we have the prospect of a bun-fight as the open data economy struggles to emerge. Take market presence - in the real world we don't take it for granted but however good or useful or politically connected a web presence might be does not mean that others shouldn't be encouraged from improving on that in entering the market. Or take lobbying - dirty word de nos jours perhaps but one only has to look at the membership of various open data entities - panels etc - to see how successful an approach this can be. Or the continued pedalling of ever bigger and unsubstantiated numbers - the growth argument has become an act of faith (Michael Cross) only. And yet there was fuss earlier this year over spotlightonspend, SpikesCavell's data analytics service adopted by some local authorities to both provide easily digestable representation of local authority expenditure to citizens and allow under pressure executives to drill down into how that spend can be rationalised/optimised.

The fuss purported to be indignation that these useful citizen facing services were a smokescreen behind which the source data was still buried; and indeed the raw data is now available. But one wonders also whether there wasn't some moral outrage that a commercial enterprise was providing paid for services on top of open data when some bedroom hacker could do have done it for nowt. But no one ever says this. Government is an organisation like any other and needs surety of service and everything else, things that start-ups, bedroom hackers and many SMEs struggle with, so outsourcing your analytics under a commercial arrangement makes sense and will become commonplace. And as the G-cloud tender documentation illustrates just the hurdles to become an approved supplier can be daunting for an SME.

The future of the freemium model for open data is around smart analytics being used to provide consulting and other services back to...government (either directly or via their third party service providers). Government can't afford to recruit, train and retain a cadre of data scientists when the commercial shortfall of these skills remains acute. It might want to but it just won't be able to and so the best thing will be done by other people. But the sting in the tail is that those people will then be selling both the resulting analytics and the services back - to deliver the change necessary to meet spending and other targets/measures.

This is a game already joined at enterprise engagement level and it seems likely that a kind of PPP position might be arrived at around free and open data in which the public role is to give data away and the private role is to sell it back in order to drive out costs, improve productivity, do more with less. So far so predictable, but growth it aint.

And bigpharma extending its wealth of personal (you dont believe de-anonymisation do you?) data to produce new drugs/remedies as other revenue generating ones run out of patent aint growth either but should at least hold up their billion dollar businesses. They'll be pleased no doubt that the data will be available and won't even notice the fraction of a fraction of a percentage point that the fact that it is free and open affords them but then that's a different story.....

There are those of course who will benefit and will grow out of the back of open data - there are self evidently smart things that can be done and smart positions and relationships to be in in a 'from data to insight' ecosystem - and that is great but I don't think the evidence supports a growth agenda at scale.


And I've just realised that that is only my second post this year - been busy!

No comments:

Post a Comment

Thank you for taking the time to ponder my musings and for any contribution you make. Although comments appear immediately (i.e. unmoderated) I will remove (or if possible) edit offensive comments.