More thoughts on the FitBit One and Personal Data

I've been posting about the FitBit One since I got mine back in January.  I was having lunch with a friend and mentioned that I had a FitBit.  He mentioned that he had one too, a while back, until his son put it in the washer.  But, he wasn't too torn up about it.  He felt that he had learned pretty much what he was going to from the data provided.  This, along with two other incidents got me thinking about the following:

When is Data Most Useful?

I have found in many situations that I care most about my data immediately or soon after it has been created.  This is not always true (e.g., pictures, movies), but for many things it is.  I take notes at work (via Mac Mail, fairly convenient given the built-in synchronization of IMAP) and find that I refer to them most frequently within the next week of writing them.  In addition, I do a fair bit of searching in my email to recall what happened in certain work situations or what decision was made.  That said, it is rare (but still happens) that I want to go beyond the last week or so of email activity to find my answer.

How does this Apply to Personal Data?

I think people need good feedback to change their behavior.  In school, we all got very timely feedback about how well we were learning the material (or at least, to the degree to which testing is an indicator of such).  When we work, we get reviewed at least annually and more importantly get raises that recognize our increased value to our firms.  The physical side of our world has few quick and easy means of providing feedback.  The FitBit should be that feedback mechanism:

• Quick
• Clear
• Actionable (advice?)
• Easy

How does FitBit and its Competitors Stay Relevant?

It seems to me that makers of more race oriented devices will have no problem staying relevant.  Think Garmin with their GPS watches that are serving runners and bikers quite well.  For companies like FitBit that are trying to fit into a more routine, they are going to have to do better in keeping people's attention.  They are trying to do this now, to some degree, by providing weekly summaries of activity and badges.  I think that they'll have to do better if they want to continue to grow and to remain useful for their current customers.

Here are some thoughts on how they can do this:

• Add features/information to the weekly summaries:  Right now my weekly summary only includes the stats for the week and little up or down arrow indicating if a specific statistic increased or decreased over the last week.  Not terribly helpful.  How about links that allow me to see in detail how this compares to other of my weeks / my friends week / all FitBit users' week / all weeks of all FitBit users' / all weeks of FitBit users of my rough characteristics (male, age, location?).  There seems to be a profound lack of imagination on the part of the developers.  
• Compare Variables: Are there any correlations in the data that have been found between certain behavior and weight or other activities?
• Better interpret activities: I've logged the fact that I'm doing weight lifting many mornings before I do some interval running.  Why can't the FitBit suggest that I could have been doing this and let me log it by hitting an "ok" button rather than going through the whole logging experience on the website.  I'm not sure where processing is done on the FitBit (before or after the data is transferred to the website), but in either case, it would seem that it could likely infer what I'm doing and auto-log.
• Send reminders to people: Allow synching of the FitBit to one's iCal or better Google Calendar.  Meeting or other reminders could cause an "alarm" on the FitBit.
• Initiate Long Term Studies of behavior:  I think the risk of losing most of its users is pretty high for FitBit and their competitors.  They'd be in a much better position if most of their new devices were being sold to people upgrading (think the Apple iPhone) than to new customers (churn is a terrible thing).  Studies of the persistence of changes in personal behavior and the potential medical benefits of having long term data about people's movement could be compelling to keep otherwise fence-sitters from dropping it.

Fitbit after 250 miles

Like them or not, when you wear a Fitbit and track your activities, you will get badges.  "Badges", you say, "We don't need no stinkin' badges!"  Maybe not. But in any event, me getting my 250 mile badge was a good trigger for me to sit down and write this blog entry to say what I'm liking and not liking about my Fitbit One.

The Good

• Battery Life: It's been excellent.  I charge for maybe an hour every week or so.  Each time I plug in, it appears as if I'm still halfway charged.  I thought I'd have to manage this more closely and am very glad that I do not need to.
• Data: It collects data in a very seamless way.  Especially when combined with the Aria scale, its a pretty powerful set of information to drive good decision-making.
• Alarm: I no longer wake my wife up in the morning with my alarm clock.  Which is good because I just broke it.  Powerful motivator to wear the Fitbit at night.
• My wife likes it.

The Bad

• Nighttime Braclet: The odd velcro-like closure is already starting to wear and catch on my covers leading to it coming off a number of times while I was tossing and turning.  I expect that I will have to be routinely purchasing these (perhaps at 4 to 6 month intervals).
• Sleep Monitoring: Seemingly uncorrelated with how well I feel when I wake up.  I moved to the sensitive mode because I was routinely sleeping with 99% efficiency.  Now, it claims I'm only sleeping between 4 1/2 and 5 hours a night.  I hope not.
• No feature updates via software (yet): I'd still very much like to be beeped at if I remain stationary for 10 or 15 minutes.  And then more vigorously beeped at if 15 becomes 30 minutes.
• Proprietary Data: Where is it?  Can I get to the raw or treated stuff?  Not clear.
• Dashboard 2.0: I switched over to it not long ago.  Lacks an easy ability to see different date ranges as you could with the original.  Some displays of the data are clunky (while others are notably improved - sleep over the last seven days).
• Changing Logs:  At least for sleep, once a record is created, there doesn't appear to be an easy way of changing it.  Make a mistake and forget to tell it your done sleeping?  You are out of luck.  Seems like an obvious oversight.

One Step Beyond?

• More types of data: I'd love to have my pulse, oxygen levels, sweat levels and blood pressure constantly across the day.  I don't know how to do it, but somebody will figure it out.
• More sensors: I can tell the Fitbit that I did a "light to moderate effort" free-weight workout for 35 minutes.  The moment I go for a run on the treadmill, the Fitbit knows exactly what is up.  It would be much more useful if I didn't have to tell it what I was doing, it should be able to figure whatever it was, out.  I'd be willing to wear ankle sensors, wrist sensors and a waist sensor, almost whatever it would take.  Whoever figures this out, will have me in a heartbeat.
• Waterproofedness: I've been spending a fair bit of time in the pool with the boys (now that we have a basketball hoop and volleyball net).  None of this activity can be logged or reasonably represented from my fitbit, as noted here.  Waterproofing seems like a fairly reasonable upgrade and one that would likely provide some additional robustness to their design (avoid failure due to being overly sweaty). [Added 2013-07-22]