How to be Happy: Experiments in tracking mood and energy

Last year, Pebble introduced built-in activity tracking as the foundation of our new Health app. But health is so much more than steps and sleep. For the past several months, we’ve been brainstorming about what’s next for Pebble Health. I’d like to share with you the results of our first internal experiment: The Happiness App.

Our motivation behind this project was to understand how our mood and energy levels fluctuated throughout the day. Was there a natural pattern or weekly cadence? Could we visualize our personalities based on these fluctuations?

We also wanted to know how different environments, interactions, and activities impacted how we felt. Were we happiest with friends, colleagues, or family? Did we feel run-down during our daily commute? Did drinking coffee, water, or alcohol have a noticeable impact on our emotional stability?

To answer these questions, we built a Pebble app that pops up throughout the day and asks you a few questions. First you rate your energy level and mood. Then you tell the app where you are, who you’re with, and what activities you’ve done in the last hour. These activities cover food and drink intake, exercise, commuting, meditation, meetings, socializing, and work. You can also add voice input over the microphone to add custom responses, for more detailed feedback. (Side note: if you want to join the fun, you can download the Happiness App in the Pebble App Store now!)

The magic of color watchfaces.

The app runs for a week and prompts you every hour or so (during waking hours only, thankfully). We expected that for some testers, these frequent interventions would be too annoying. While that was certainly true in a few cases, the majority were invested enough to stick it out the full week. We even had a couple testers who got so attached to the app that they re-enrolled for another weeklong term!

We found that the app delivered value to testers simply by providing them with a moment to check in on how they were feeling and what they were doing. In essence, the app supplied a call to action — it asked testers to pay attention to their emotional landscapes, and to stay aware of the various activities that can impact their mood.

We also noticed that the app’s prompts subtly encouraged testers to adopt new behaviors. One tester realized that he was always saying “no” when the prompt asked if he drank water in the last hour, so he resolved to drink more water. Another took up yoga thanks to a recurring yoga-related question. Since the study only ran for a week, these changes are not likely to stick for a lifetime, but it is encouraging to note that subtle suggestions really can inspire healthier habits.

Finally, we gleaned some fascinating insights from the summary data we churned out at the end of the week. Remember our question about graphing people’s personalities? Take a look at the shape of my energy graph versus my colleague Heather’s:

Who would you trust with the nuclear codes?

On the left you have me: buzzing with energy in the morning, eager to face the day. Then I roll into my afternoon meetings and have to start dealing with people. This isn’t all bad! But I bounce back and forth between euphoria and crashing-and-burning until I completely exhaust myself (the 6–7 pm low point) and go home. This plot is fairly reflective of my reality: I’m revved up and excited about something one minute, down and out the next. I’m now in a much better position to deal with those down cycles — my plot shows me they may be frequent, but they are usually fleeting.

On the right you have Heather: the quintessential morning person (and, we must note, much more even-keeled). She starts out strong and gets a boost of energy as she arrives at work and greets her colleagues. She tends to get a small boost around lunchtime, but from there it’s a slow decline until she gets home. Getting home re-energizes her, but only briefly — by 9pm she’s crashing and ready to go to sleep. This plot supplies a useful reminder that for Heather, ending the workday at 5pm is best — her best hours are behind her at that point, so it’s best she go home and recharge.

On the mood front, we observed another tester who we knew tended to get hangry near mealtimes. Check out his mood graph:

Hanger is no joke.

Yes, he sees a complete crash in mood before lunch! His pre-dinner slump is not as bad, but eating dinner gets him back to gleeful levels of good feeling. If only more of the world’s problems could be solved with a granola bar.

Early in the development of the app, we decided to distinguish between energy and mood. I’d wondered how this decision would play out — if I’m feeling energetic, am I necessarily happy? It turns out that for me, energy and mood don’t always align. For example, my results showed that I tend to be fired up and high energy during commuting hours. (Am I frustrated with transit? Eager to get home? Buzzing from caffeine intake? The jury’s still out on this one.) However, commuting hours are also my unhappiest time:

Road rage in graph form.

I also got deeper insight into what drives my mood from my activities analysis:

Disclaimer: I was really annoyed leaving the gym one day during the test — I’m pretty sure I don’t just hate exercise?

I was surprised that social activities tended to precede my good mood: whether that was talking on the phone, chatting with a colleague, or even attending a meeting (especially since, in the aggregated data, attending meetings had the most negative impact on mood for our testers!). I’m now much more inclined to give myself a few minutes to say hi to folks around the office, since I can see it makes my day better.

This is just a taste of the kinds of insights the app delivered — for the full story, go try it for yourself! Overall, our testers found the experience interesting and insightful (and some of the results are utterly hilarious, especially if you can compare with friends). We hope you’ll try out the app and let us know what you think!