“You Gotta Move”: Fitbit Flex And Behaviour Change

However, while they may create healthier decision-making in the short-term, there has been a general lack of evidence for longer-term behavioural change in fitness trackers [2].

“In the midst of this frenzy of anticipation, the dirty secret of wearables remains: most of these devices fail to drive long-term sustained engagement for a majority of users.” – Endeavour Partners [3]

We were keen to test whether the availability of fitness and healthcare data would actually translate into healthier and more active behaviour. We selected several popular activity trackers popular on the market, such as the Fitbit Flex, Jawbone UP and Nike Fuelband. I opted for the Fitbit Flex as I knew several people who used one and thought it would be fun to compare activity stats.

I was intrigued to see whether the Fitbit would stay with me for the long run (pun intended) considering the primary failing identified for activity trackers has been the lack of sustained user engagement over time. A recent report found that more than half of American consumers who have owned a fitness tracker no longer use it – a third of consumers stopped using the device within 6 months [3].

My general user profile was one of a novice starting out: I had never calorie-counted or used any sort of sensing device to track fitness before. I am definitely not a self-identified “quantified selfer” that eagerly tracks my own data, and may be one of those users that needs an extra push in order to engage with this [4, 9]. My main requirement for this device was for it to be simple to use and become a part of my daily routine without too much effort on my part.

‍

‍

Fitbit Flex: the low-down

The Fitbit Flex has a simple and elegant design, is reasonably durable and has a good battery life (it lasted 6-8 days when I used it, though the figure might have been lower if it was syncing with my smartphone). The black band was unobtrusive enough, and bar the occasional hiccup with getting caught in blouse sleeves, comfortable to wear.

Fitbit uses visual and haptic feedback to let you know when you’ve completed a goal – slight vibrations and twinkling LED indicators indicate when you have reached one of your daily goals (e.g. target number of steps). Goal progress can be viewed at any point by tapping on the band, which is quite nice for a quick check of how you’re doing that day.

The design appealed to me in its simplicity, and the simple representation of goal completion through an increasing number of lights made for very easy and immediate feedback. I could forget about wearing the band until it slightly buzzed to indicate a goal had been reached. So far so good.

When using the sleep tracking feature, you have to rapidly tap the Fitbit for a couple seconds in order to toggle between sleep and normal modes. The sensitivity was a bit off for me: at night I would have to manically tap the Fitbit before it activated sleep mode, while occasionally any slight pressure would send it into sleep mode during the day.

While trialling my Fitbit, I repeatedly forgot to switch between the two modes. This was problematic as I would frequently forget to tap out of sleep mode until midday, and thus would lose out on tracking any of my activity levels in the interim. The device lost points with me on this one as I frequently forgot to toggle between modes and so felt like it wasn’t accurately tracking my activity.

I expected Fitbit to be able to automatically switch between these modes, and indeed other trackers such as Jawbone UP do this. If it’s logging more than a couple dozen steps during sleep mode that isn’t the sort of thing that can be put down to tossing and turning, and it should be able to sense that and switch to awake mode (bar potential sleep-walking– but hey you’d still want to track steps taken when sleepwalking!).

‍

‍

Fitbit app

Fitbit didn’t have an official Windows phone app at the time of testing the device or writing this [5], so unfortunately I missed out on all of the smartphone-based notifications and user experience. I used the desktop app instead, which meant I had to consciously start syncing with the app by running the desktop app from my work computer or laptop at home.

In order to sync, you need to use the Fitbit dongle which conveniently uses bluetooth so this guarantees compatibility with any laptop. No surprises here, but I also frequently left the dongle in either my place of work or at home. This was frustrating as I wouldn’t have up-to-date information on my activity levels when trying to check. While I liked the UI of the desktop app, and could readily understand what graphs were referring to, I didn’t actually end up using it that often as I felt the data was not representative of my latest activity levels.

Another limitation of fitness trackers such as Fitbit is their reliance on motion rather than exertion to track activity levels [4]. I discovered this when realising my weekly trips to the gym barely registered with the Fitbit, as it isn’t able to sense strength-training activities such as Pilates or weight training. This largely contributed to my lack of reliance on the data.

I tried to use the calorie-counting feature of the app for the first week. Because I didn’t have the mobile app, I had to rely on remembering to log meals on the desktop app when at home or at work. What’s more the food database was heavily US-centric: it ended up being too much effort to try and figure out whether a Campbell’s tomato soup is equivalent to a Sainsbury’s tomato soup, so that did not last very long.

‍

‍

Behavioural change – going the distance?

Although I did wear the Fitbit for about 6 weeks, I stopped engaging with any of the data after the first few weeks due to the limitations described previously. During the time I was wearing it, I did not consciously notice a change to my activity routine, as my weekday fitness levels stayed pretty consistent and weekends were variable depending on whether they were spent on city breaks or binge-watching shows on Netflix.

It’s difficult to ascertain where exactly fitness trackers such as the Fitbit Flex are missing the mark. It could be that the technology limitations in themselves are deterring all but the most motivated users [6]. The simple fact that these devices aren’t that sophisticated yet in tracking different types of activity and automatically sensing varying contexts of use.

Although companies are trying to appeal to users by adding on other features to track general “wellness”, such as sleep or caloric intake, these may not be honed enough to actually deliver value for users at present [8].

“It may be that they are presently so primitive that it’s no surprise that people give them up: they’re too big, haven’t discovered the killer app that we want out of them.” – Charles Arthur – The Guardian [7]

What’s more, forming new habits is difficult at the best of times. Integrating fitness tracking into a daily routine may currently be hampered by the amount of intermediate steps required to get representative and accurate data (e.g. manually logging time spent at the gym). In order to effectively create new habits, the apps should provide effective and relevant external triggers to prompt behaviour, as well as a reward-based system to recompense desirable behaviour.

External triggers could for instance involve sensing less than optimal behaviour patterns and prompting users to change these in the moment (e.g. suggesting an alternative activity when you’re about to plop down on the couch for a lazy Sunday). In terms of rewarding behaviour, people have natural competitive tendencies and benefit from social motivation to a certain extent, which could be further exploited in fitness trackers.

“Products and services that provide utility but fail to have a meaningful impact on users’ behaviours and habits—such as an activity tracker that provides data but doesn’t inspire action—end up failing in the market.” – Inside Wearables, Endeavour Report [3]
Ultimately, even if fitness trackers manage to deliver useful and reliable data, engagement with the device needs to become habitual and meaningful if they are to create long-term behavioural change in users. The key in long-term behaviour change has more to do with social theories of human behaviour, and these gadgets may need to go back to the drawing board to see how exactly they can create and sustain motivation over time.