Anatomy of a nudge – Applying behavioural insights

[Originally published on Australian Government DesignGov under a Creative Commons 3.0 BY AU licence]

The work of the public sector revolves heavily around behaviour – whether it be about ensuring compliance with laws or encouraging healthy living or safer habits. How can the public sector use insights from behavioural economics (and other fields) to help ‘nudge’ people to behave in ways for the common good?

On Thursday 2 May, Dr David Halpern and Dr Rory Gallagher of the UK Government’s Behavioural Insights Team (known as the Nudge Unit) spoke at some events in Canberra about their work and what they have learnt about applying behavioural insights. The following is my summary of a discussion at a cross-agency breakfast roundtable organised by DesignGov.

Dr Halpern spoke of the opportunity provided to them when the Nudge unit began – a UK public sector faced with financial cuts and a political focus on not regulating. How then to change behaviour? By applying behavioural insights.

Dr Halpern outlined the EAST framework – that if you want someone to do something, make it Easy, Attractive, Social and Timely.


We systematically underestimate the impact of multiple small frictions in shaping behaviour. A focus on making something easy might be about changing the default settings (e.g. having people in the UK have to opt-out of a pension scheme, rather than having to opt-in). It might be about removing frictions (e.g. providing someone in a letter with the link to a specific form rather than the top webpage). Or it might be about simplification (so giving consumers access to the data that utility companies hold about them and proving a simple means of choosing plans that suit their actual usage).


If you make things attractive to people, they are more likely to act. Elements of attractiveness include making it salient and personalised (so handwriting on an envelope that the recipient needs to open a letter) and changing the messenger (e.g. someone with relevance and authority).


Link the behaviour to social expectations and norms, use networks and notions of reciprocity. Tell people about what other people do – “nine out of ten people pay their tax on time” increases the rate of other people paying their tax on time. A patient ‘hotel’ in Sweden includes a second bed in the patient room for a friend of family member – it lets people act in a social way and helps get better clinical outcomes and better patient satisfaction.


Make things timely and relevant at key moments/decision points. For instance asking people about leaving a legacy gift/donation when they are developing their will increases the number of people who will do so.

Dr Gallagher spoke about an example he had been involved in with employment services where they focused on three elements:

  1. Cut down process – instead of having the first meeting between a job seeker and the job centre adviser being about filling in forms, they ensured that it was about finding a job
  2. Commitments – this involved having the job seeker write down what they would do in the next week/fortnight rather than be about what they had done in the previous period
  3. Strengths identification – this involved working with the job seeker to identify their strengths and what they could offer, as opposed to what they could not.

Rory noted that the initial results have shown a 15-20% increase in people getting off unemployment benefits after a period of three months.

Dr Gallagher also spoke of some of the projects that he is working on while in Australia with the NSW Department of Premier and Cabinet.

There was some useful discussion after the presentation including:

  • The risks of desensitising or over-using these approaches – in most instances it is about helping people do something they actually wanted to do, and it is of most use at key one-off decision points. If it is used by every agency, all of the time, then yes it will be less effective, but we are a long way off from that now
  • How do you integrate this approach into agency operations? – getting results is the major means, and demonstrating them through the use of randomised control trials is particularly important
  • That the key for success in using these approaches is to focus on the citizen and helping them – if it is used as a means of just making things easier for the bureaucracy, then it will lose support and effectiveness (and lose its purpose)
  • Communications is a very powerful policy tool, but it is not currently placed as being such in the policy repertoire.

It was a great discussion, and the Dr.s also presented to a cross-agency forum organised with the Department of Finance and the Australian Taxation Office, as well as to some other groups.

We look forward to continuing to build connections between agencies that are interested and active in applying contemporary approaches to challenging public issues, and to contributing to the discussion about behavioural economics and how it fits with design-led innovation and other emerging strategies for achieving better public outcomes.