A study on the inadequacies of fitness trackers highlights a couple of essential behavioural truths. 

One, as discussed by Kahneman in Thinking Fast and Slow, is that the most exciting area of behavioural research is now 'spillovers', or what happens next after your first behaviour has been influenced. Let's say you have a fitness tracker, and you hit your target for the day. Does that achievement promote you to then keep up the good work and get an early night? Or does it permit you to have a couple of pints on the way home? We don't know enough about how spillovers work, and devices like these tend to target the first behaviour only. Paul Dolan and Matteo Galizzi at the LSE did a nice little experiment on this, showing that we tend to permit ourselves to over-eat, cancelling out the first behaviour and taking us further from our overall goal (losing weight). http://cep.lse.ac.uk/pubs/download/dp1286.pdf 

Two, and linked to this, is that we're social creatures. Tech is rarely enough in achieving social outcomes. I've seen that in services we've evaluated, where the technological element is important but never sufficient. You need a human, somewhere, to encourage your promoting spillover, whether that's about improving your health, supporting environmentalism or increasing volunteering.

In the field of social innovation, these two points sometimes get lost in the fun of a new gadget!