From counting steps to changing behavior

Four design principles to build more human-centered, sticky, and successful self-tracking services


Four years ago, I bought a Nike+ chip to track my runs. I didn’t buy it because I’m a passionate runner, but because I’m a passionate researcher. I love collecting data; analyzing and learning from it. I hoped that if running was more about research, I might start to love it too.

That was just the beginning. Over the next year, I started monitoring how much water I drank, my mood, sleep, period, heart rate and alcohol intake. And I started running twice a week.

I also learned I’m not the only one tracking personal data. In 2005, info-graphic designer Nicholas Felton created his first Annual Report, visualizing his year’s activities. Gary Wolf and Kevin Kelly of Wired coined the term ‘Quantified Self’ in In 2007. They founded Quantified Self Labs, which organizes international meetings, and connects users and makers of self-tracking tools. Self-tracking seemed to be everywhere.


The promise: know thyself…

Curiosity about ourselves is inherently human. What’s new is the technology to satisfy it. Sensors have become smaller; self tracking devices wearable and affordable. As a result, lots of products and services have appeared over the past few years.

These technologies have huge potential. In recent years, degenerative conditions like type 2 diabetes, cancers or coronary diseases have, for the first time in history, replaced infectious diseases as the number one cause of death in the US. Self-tracking can help people better understand their body and what affects it, to make more informed choices about their health.


… the reality: numbers, without meaning

Despite that promise, self-tracking services are mostly used by geeks or sports enthusiasts. They don’t reach people who’d benefit most, like the elderly or chronic disease sufferers. And they don’t always deliver meaningful data. Users are left with numbers but little insight.

Without impact, self-tracking sooner or later leads users to ask the same question: So what? Many self-trackers stop after an initial period of curiosity. I know: I’m one of them.


Designing better self-tracking services

That’s led me to look at self tracking not from a personal perspective, but as a design researcher trying to create positive impact in people’s lives. And to ask: how can we help these services to fulfill their potential?

Based on our design research process for collecting and using data here at IDEO, my colleague Hannah Peres and I developed hypotheses about people collecting their own data.

After talking to users and designers of these services, we developed our hypotheses into three design principles (and one bonus) to make self-tracking more human-centered, more meaningful, and more successful.


1. Ask the right question.

Self-tracking services should address a human need, and have a clear purpose for how data can address that.

Fertility tracking service Glow, for example, asks: How can we empower women to take control of their reproductive health? It starts by asking users a specific question: Do you want to get pregnant? Or do you want to avoid getting pregnant?

This puts data into a meaningful context: an important personal goal. Based on that, Glow gives users tailored advice, and differs in tone: either advising on the risk of getting pregnant, or the chance of getting pregnant.


2. Move from data to insight.

Most self-trackers are not interested in numbers, they want to better understand themselves. We interviewed the founders of Melon, a headband, that tracks brain activity and teaches users about their cognitive performance. They don’t see Melon as a quantified self device, but an understood self device.

Melon helps users recognize patterns and draw conclusions by asking them to add contextual data such as time of day, or mood. That allows the user to not just quantify focus, but draw insights, such as music helping improve their focus.


3. Move from insight to action.

People’s ultimate purpose in collecting and analyzing data is behavior change. And that’s where it gets tricky: knowing we should exercise doesn’t make us hit the gym.

Mira — a wearable fitness tracker IDEO helped develop — has found a way to address that. Designed for busy women, the device aims to integrate more activity into their daily routines. First, by getting to know what exercise they like — say, evening power walks — which they enter manually.

Then it helps them improve, offering actionable fitness insights about their choices, and suggesting personalized challenges. Called ‘Boosts’, these pieces of content are customized to their progress and preferences, and aim to focus and motivate them.


+1. Tell a human story.

My Nike+ chip helped me to develop a running routine… for a while. I stopped running after a year. Yet, the data doesn’t give much insight about why: I could have stopped because I had a baby, a broken leg, or stress at work. There’s a human story behind these numbers, but it’s not captured.


Four Steps In The Right Direction

So, applying our three, or rather four, design principles: what would have taken Nike+ — and my running career — to the next level?

Asking the right question: knowing my goal — running a marathon, losing weight, or coping with stress, for example — would have enabled Nike+ to better tailor communications.

Moving from data to insight: adding contextual data would have helped me better understand what affects my running, and how running affects my life.

Moving from insight to action: based on these insights, Nike+ could have helped me change by, say, giving me real-time feedback in the moments that count.

Last but not least, Nike+ could have kept me running — and using the service — by doing one thing that makes all the difference: not just showing me data, but telling a very human story about me.