How to create sustained behavior change with your product


FitBit One. Image from article in MedGadget

Every single year, the vast majority of fitness clubs lose between a third and half of their members.[1]  By and large, that’s because the members wanted to exercise, thought they would exercise, but failed to actually continue doing it. And so they canceled. Heck, I’ve been one of those people.  I’m sure many of you have as well. Sustained behavior change, even when we want to do it, can be bloody difficult.

The same thing happens with most products I’ve seen that try to change user routines: from apps built to support new habits (like running), to apps that help people stop smoking.  People are really interested in the beginning, and most quickly drop off.

Two weeks ago I talked about what changes for a person between the first time they try a product, and the second time.  Now, I’d like to focus on what a product can do to sustain behavior change over time.  I’ll use my wife’s FitBit One, an exercise tracker, as an example.


1)    Build a habit.

Setup a clear cue for action (like 7pm at night).  Have the user practice a consistent, unvarying routine each time the cue occurs (jog around the neighborhood).  Reward the user (show how many steps were taken and calories burned). Repeat over many weeks or months.  Our brains are hard wired to build habits, precisely because they save us (mental) effort on repeated tasks.

Challenges: If something disrupts the routine, like traveling to see family over the holidays, the habit can be broken.  The 7pm=”time to jog” association is weakened.


2)    Make the behavior so compelling on its own, that users will consciously seek it out even without being prompted to. 

If you were to pay me $1000 each time I borrowed my wife’s FitBit and ran around the block, I would. No problem.  I’d do it every day.

Challenges: Most products just don’t have that kind of mojo.  Some things that are inherently really fun or relaxing do.  Unfortunately, we get bored doing the same thing again and again – even if we really loved it at the beginning (Oh, the hedonic treadmill).  So, conscious decisions to seek out the opportunity to  something  are hard to sustain. 


3)    Change the person’s environment so the new behavior is a normal, regularly triggered action.

Help the person take the action a few times.  Remove friction (get good shoes, clothes to run).  Build up a community of friends who also do it.  Setup regular group runs.  The new environment (friends, invitations to run, running being “normal”, having all of the necessary gear at hand), means the person is constantly reminded to act, and need not internalize it or use willpower to act each time.

Challenges: Most products don’t have that much influence on a person’s surrounding “environment” – social networks do, as do some products essential for one’s daily job.  With social networks, though, the person’s sustained behavior is subject to the changing interests and fads of their friends. 


4)    Change the person so the new behavior is a logical extension of who they are.

Help the person take the action a few times, at increasing levels of difficulty (jog a bit further, a bit longer, each time).  Help them overcome obstacles, and reflect upon their successes.  Help the person see herself increasingly as a runner.  Make it a matter of self-identity. Even in new environments, they’ll still look for the gym to do some running.  Because that’s who they are.

Challenges: Marketers love this. It can arise out of a deep brand loyalty, for example. But, we have many facets of self (schemas) that we activate as needed in our daily lives, so even our “identity” is environment-dependent. As far as sustained behavior change goes though, this is a pretty powerful route. 


5)    Shift the burden of work from the user, to the product.

The easiest way to sustain a behavior?  Don’t require it at all.  Cheat. Payroll deductions for retirement, or automatic bill payment are two good examples.  Where you can’t fully automate the action, you can automate tracking and attention – so that the product infrequently grabs the user’s attention, asks for a quick decision, and then takes care of the rest.  A heart rate monitor does this.  Usually, no problem. You don’t have to worry. When there’s a problem though, it’ll tell you!

Challenges: Only certain behaviors can be automated in this way. Running can’t, for example.


There’s no silver bullet, clearly. But, by adjusting the environment, changing how people sees themselves, building habits, or removing the burden of work, it’s possible to sustain behavior change.   What’s your strategy?