Seven core UX principles to guide the development of quantified self services
We’ve been involved in the development of a couple of quantified self services recently. Listed below are the UX principles that we use to guide our work in this space.
1. You catch more bees with honey. Always reward positive behaviour don’t tell people off for not achieving things.
There are plenty of examples of this in the form of structured rewards and badges. Nike FuelBand is a great example of this. Badges are rewarded for hitting daily goals, hitting three or four daily goals in a row, cumulative achievements etc. Check out the rewards here.
2. You can’t break bad habits you can only replace them with better ones.
We often think in terms of breaking bad habits; eating too much, drinking too much but it turns out it’s not as simple as that. Habits are so ingrained you can’t simply get rid of them, you have to replace them with something else. That often means developing alternatives that are appealing for the end user. If you want to stop people going to the pub after work you’ll need to offer a viable alternative rather than simply telling them it’s a bad idea.
3. Data capture should be frictionless and ideally automatic.
Quantified self works best when the data is captured automatically and the user doesn’t have to actively input anything, Sleep monitors are a perfect example. Once the device is activated and laid on the bed all you have to do is go to sleep. If you do have to input data then it has to be as simple as possible. NetDiary a food monitor, for example, allows you to scan the bar code of thousands of common food products speeding up and simplifying the input process.
4. Surface data in simple useable interfaces.
Simply surfacing data can cause behaviour change in it’s own right, understanding how much you eat, drink or how active you are can lead people to adjust things for the better. It is even more powerful if the interfaces that play the data back are context specific. Nike FuelBand, for example, has a super simple display that shows how well you are doing against your daily goal moving from red to green as you progress towards it. Download the data from the FuelBand to your phone or computer and you can view more data broken out by week and month.
5. Comparative data is most motivating when it’s specific to the user.
Behavioural science tells us that people are heavily influenced by what people who are close to them either geographically or demographically. If I’m monitoring how much energy I use in my home it is more important to me to see how well (or badly) I’m doing compared to my immediate neighbours. If I’m 42 and monitoring my running it’s better to be benchmarked against other people of the same sex and age. The more specific the comparative data the more motivating it is to the user.
6. A public pledge to achieve.
People who pledge publicly to achieve a certain level of attainment are far more likely to achieve it . Make sure that people can set goals and update them on progress towards them.
7. A support community is better than a support function.
Access to a group of peers who can support and motivate someone is more motivating than company support. Peers can ask the silly questions, say the things that companies can’t and spend the time supporting each other. Where possible test the role of community.
What do you think? Is there anything we’ve missed? Interested in working with us on a quantified self project? Say firstname.lastname@example.org