Persuasive Technology + Systems Thinking for Gov2.0

Government agencies across the world are organizing data mashup contests as part of their Gov2.0 initiatives. This is great but certainly not enough. Some analysts think it’s a futile exercise. I wouldn’t go that far but I certainly believe that the true audience for Gov2.0 are regular citizens and not just the programmers and data analysts who participate in these competitions. The true value of Gov2.0 lies in energising the masses to participate in policy creation and service delivery processes.

So, what can government agencies do?

Two things:
  • First of all, understand what it takes to influence digital (or otherwise) behaviour of masses. 
  • And secondly, use/design a suitable behavioural modelling tool to gain deeper understanding of the outcomes from your Gov2.0 initiative.

Alright, so what does it take to influence behaviour of masses?

First of all, do you even have a framework for modelling digital  behaviour of citizens for gov2.0 related initiatives? No, so build one – build a model for exhorting regular citizens to participate in policy creation and service delivery. A good place to start is Prof. BJ Fogg’s Behavior Grid (Stanford Persuasive Technology Lab). What kind of behaviour do you want people to exhibit? You want them to do something once (click on vote button) or you want them to do something for a period of time (keep donating money till everyone’s better off) or maybe you want them to do something for the rest of their lives (quit smoking and bring down insurance premium for everyone). All of these require different approaches. Select what you want to do from behaviour grid and then design your digital Gov2.0 programs accordingly.

It’s important to do one small thing that can influence behaviour of a million people than do one exceptionally great thing that can influence behaviour of one guy.

Secondly, once you’ve decided the behaviour that you want your citizens to exhibit, you need a software model to run if-else scenarios. Where would you get such models? Regular statistical models? Sorry that won’t work. Digital behaviour is too dynamic and emergent to be modelled using your regular stats tool…when the line between cause and effect blurs…you have a chicken-and-egg problem… A combination of agent-based and systems thinking model is your best bet if you want to model the emergent nature of digital behaviour. “Systems thinking” is not new for government agencies (see but using it to model digital behaviour of citizens is perhaps new. If you are a Gov2.0 leader, summon your Systems thinking expert first thing tomorrow morning J

There’s also this let-go-of-control philosophy that’s being evangelised by many gov2.0 experts. No complaints. This is certainly what it should be like…after all isn’t gov2.0 all about shifting power from governments to citizens?…but this is just the beginning…we really do need governments to nurture gov2.0 initiatives for a while…once things falls in place, it’ll become a self-sustaining system…and that’s where we are headed…you really don’t have to ask govt. to let go of control. The very design of web 2.0 will take care of that!!

No comments:

Post a Comment