Can an app make us behave better?

Everyone’s talking about a new iPhone app, Mailbox. It uses simple swiped gestures to archive email, or postpone it, or send it to a “to do” list. Mailbox users can quickly highlight their entire inbox and instruct it to come back “tomorrow”. If you have an iPhone, you’re probably already in the queue for the app.

What’s intriguing about Mailbox is that it is basically a redesigned front-end for Gmail; it adds very little actual functionality to Gmail but it strongly nudges us in particular directions, making it easy for us to handle our email in the way we should have been handling it anyway. The word “should” is intriguing: I’ve strong opinions about how to handle email. Mailbox also has strong opinions, unlike the bland “do with me as you will” of most email clients.

Modern software is fascinating because of the rapidly evolving way in which software designers try to make complex tools intuitive to use. The results are patchy but, at their best, rather brilliant. And the way in which the best software is created is fascinating, too: it’s a potent blend of thoughtful design with constant experimentation. The design gurus brainstorm and create; the experimenters see what works. In fact, this process – supported by the relentless improvement of our silicon infrastructure – has become so successful that when somebody says “technology”, we immediately think of computers and phones, rather than aeroplanes, vaccines or nuclear reactors.

Perhaps it is no surprise that “behavioural design” is becoming a buzz-phrase. Nick Chater, professor of behavioural science at Warwick Business School, argues that the combination of basic scientific research with user-oriented design thinking is a powerful one. Behavioural scientists in the fields of psychology and economics are producing reams of research about human behaviour, and designers have the skill and experience to turn those insights into products and services that make our lives happier or safer.

One new home for such collaboration is the Behavioural Design Lab, a joint venture between Warwick Business School and the Design Council. Another is Ideas42, a Massachusetts-based “design lab” established by economists and behavioural scientists. (The “42” is a Douglas Adams reference: we need to improve the questions we’re asking before we can find useful answers.)

A new policy paper by Saugato Datta and Sendhil Mullainathan of Ideas42, published by the Center for Global Development, uses the example of fertiliser use to illustrate behavioural design. African farmers use much less fertiliser than one might expect and suffer low yields. The natural explanations for this situation are that the fertiliser is unaffordable (solution: subsidy) or that it’s less useful than agronomists might think (solution: agronomists should back off) or that the farmers don’t understand the benefits (solution: education).

But behavioural economics suggests another possibility, which is that the farmers want to use the fertiliser but, being human, are tempted to fritter away their cash and can’t afford the fertiliser when they need it. One possible design solution is to offer commitment savings accounts to allow farmers to lock their cash away. When this idea was tested in Malawi, the farmers signed up with alacrity and fertiliser use increased sharply. Subsidising free home delivery of fertiliser also works very well, much more than a price discount of equivalent value.

These are design options, making it easy to do what we hope is the right thing. Free home delivery of fertiliser is an opinionated subsidy, just as surely as Mailbox is an opinionated Gmail interface. And when we combine the slick influence of design-based thinking with a humble willingness to test, learn and adapt, we have a powerful alliance.

Tim Harford is the presenter of Radio 4’s ‘More or Less’

Copyright The Financial Times Limited 2017. All rights reserved. You may share using our article tools. Please don't cut articles from and redistribute by email or post to the web.