When we started thinking about user's interaction with Breeze we considered it from two points of view.  One was the "real time" point of view -things happening to me right now that I can effect, the other was the progress point of view -a collection of things happening over time that may or may not influence how I act today.

When we set out to design the realtime experience we had several goals:

  1. Give a clear picture of where I am right now and where I should be so I can catch up or relax.
  2. Make goal setting frictionless, dynamic, and challenging (but not impossible).
  3. Get folks to begin to build a single day into a life long habit.
  4. Don't drive users into the app for no reason, push information to them when it matters.
  5. Steps are boring.  Give users something fun and rewarding every single day.

To accomplish these goals we approached the experience in a number of ways.

  1. In addition to a simple visualization of your progress towards a goal, we included a little indicator we internally called "the rabbit" (like from dog racing).  When you look at the rabbit on the dial it shows you where you ought to be, and not linearly, but personalized based on when you are most active during 24 hours for that day of the week.  That way users could, at a glance, say "Sure I'm only halfway done...but normally I'd only be a quarter of the way.  Go me!"
  2. 10,000 steps has been the daily goal ever since some Japanese health campaign in the 80's "said so".  Problem is, if you're 400 lbs., barely walking, and decided to turn your life around 10,000 steps is only a painful reminder of how far you have to go.  Couple that with the fact that no user in our interviews could tell us why 10,000 really mattered, and we knew there was a better way.  Instead we automatically set users goals for them.  We looked at their previous 7 days (or later 4 weeks) of data by day of week throwing out high/lo anomalies and forecasted what should be their goal for the day.  In essence, all we were asking of users was to keep doing what they were doing; if they did better then the goal pushed them farther, if they did worse it adjusted.
  3. While we knew we wanted to focus singularly on "today's activity", we also knew we needed to turn one successful today in many successful tomorrows.  To do that we relied on celebrating and reinforcing "streaks" across a week within that same real time view.  We encouraged users to string them together in multiple views, and gave them special content when they kept a streak alive.
  4. There is very little reason to force a user to open a passive tracking app several times day.  At best its annoying.  Instead we used push notifications to do 3 things.  1)  Automatically congratulate someone when they did a "long" activity and tell them how far it was 2) Automatically congratulate someone when they beat their goal 3) Let users know early enough when they were close to their goal, but not on track to hit it, giving them enough time to change course and be successful.  None of these things really require an app open, so we communicated all through pushes.
  5. Lastly, we wanted to keep things fresh and fun.  We used a different blurred photo background for the app every single day. In addition to looking nice, when users did hit their goal, the picture un-blurred.  Surprise!  Was this a killer feature?  No.  Did it make the app fun, unique, and act as another tool to help users build habits?  Absolutely.