The public sector should be leading on Big Data

  • Vote This Post

    1

If there’s one class of enterprise whose IT folk should be focused on big data pretty much as a given, it’s the public sector.

Here’s why: evidence-based policy formation.

We’re all aware that the public sector is facing a squeeze. Governments in deficit need to return to balanced budgets, perhaps even return to retiring past debt (in Ontario, for instance, the “Ministry of Debt Servicing”, with its single program — pay interest on the province’s accumulated debt — is the third largest ministry by spending, and closing in fast on number two).

That means there’s little money or appetite for redirecting some for a massive project.

Yet, at the same time, it’s the public sector who’s best placed to have some early big data wins.

Municipally, there’s a constant debate about traffic flows and patterns. Should streets be one way, or two? Should a lane be taken away for bicycles, or not?

Instrument the roads involved with traffic counters, measure for a few months, implement the change, measure again.

Now you know what the effects are.

You may be asking yourself, “but what does this have to do with IT?”

As with BYOD, big data heralds a core transition in IT work. It becomes less about business processes and efficient operations (they still matter, but a lot of that work is done) and more about value generation.

You want to partner with your business colleagues in the enterprise, you need to bring ideas to their table. Ideas that don’t assert an outcome, but show how it can be tested relatively quickly and inexpensively.

IT gets to learn about the tools and techniques needed to do this repeatedly. The business area gets hard data they can slice and dice in many different ways to do their job.

In the public sector, that’s policy recommendations.

Imagine a City Council that could be taken through minute by minute ebbs and flows on a change, see waves of traffic move through streets.

Imagine seeing as well whether the usual fear — if I change this street, the next street over gets the overflow — is real or not.

For, with traffic, sometimes that is what happens — and sometimes the traffic just melts away. People find other ways to get around than driving a car.

IT’s contribution has simply been to be able to put facts on the table. The policy work happened where it should, in the business. But they can defend their recommendation, not with opinions from consultants, but with data.

Start small. Pick a target. Keep it cheap and simple. Build your expertise.

Delivering wins will help keep the dollars that are available flowing into IT, because you’ve shown you produce results with them. Isn’t that what you want?

Big data could be the public sector’s time to shine.

Bruce Stewart Bruce Stewart (99 Posts)

Bruce Stewart is a 39 year veteran of IT management and above. He is an executive advisor serving CIOs and senior executives in areas of governance, strategy, complex architectural transitions, portfolio yield and value generation.


  • DonSheppard

    I happen to support getting more opinions from consultants, but have no objection to those opinions being based on more “hard” facts!

    One of the questions is:  How do you get started?  For example, you say “instrument the roads with counters” – how expensive would that be?  Would the data collected that way be significantly bettr than taking samples manually?  Possibly, but it can be hard to convince people in a business case that requires purchasing equipment, installing networks, adding the databases and acquiring analytics software.

    Perhaps the TTC would be a good spot to start – counting riders getting on a subway should be relatively easy, at least in aggregate.  Perhaps even counting people leaving a station might be feasible.  But this is municipal rather than provincial and may run into funding issues.

    All in all, its a great idea but maybe the times aren’t quite right for this to take off quickly?  Finding the way to start small is the key, and that’s where cloud computing might offer some advantages.

    • Bruce Stewart

      Earning my living more from consulting than anything else over the past fifteen years, I agree with your observation!

      What I’m thinking of is creating bursts of activity. Road counters (those grey boxes with hoses attached: passing tire pressure “clicks” the counter) aren’t very sophisticated, but most cities have a few hundred of them. Normally they’re assigned all over the city. But if, for a few months, you brought them all to one location you’re thinking of a major change in, put them on the side streets as well as the main routes, you’d have a dataset built that mapped reactions to the temporary changes you introduced.

      The TTC already has cameras by the driver on its bus and streetcar fleets: easy enough to trap a count of bodies moving in and out (the same technology is used in retail to map traffic patterns in stores). For the subway you’d need a camera at each door to do the same thing rather than one at each end of the carriage (those are security features now). Again, not hard to get a dataset built that shows real ebbs and flows in the system for, say, a month.

      So in both these cases (deliberately municipal in nature, because money there is hard to find) you could bootstrap up an analysis, act on it, and report to show why more investment would make sense…