Skip to main content

Mastery, the scientific method, and policy design

Posted by: , Posted on: - Categories: Case study, PublicPolicyDesign, Thought leadership

A photo of public sector workers participating in a local event.

On the BBC radio programme, ‘The Life Scientific’ in early October, Sir Patrick Vallance argued for the application of the scientific method to policymaking. This wasn’t just about employing more STEM grads. It was an argument for a broader shift in the way that governments solve problems. When developing policy, governments at every level - including my own - should rigorously test their assumptions and hypotheses using empirical observation and inductive reasoning wherever possible.

Listening to the life scientific made me wonder whether Patrick Vallance would recognise policy design as part of this transition. It also made me wonder how many policy design practitioners would recognise themselves as being part of a movement for greater scientific rigour in policymaking. Our experience in the London Borough of Barking & Dagenham suggests that this is a missed opportunity.

The scientific heritage of policy design

Whilst policy design as an umbrella term is comparatively new, the disciplines associated with it have long and proud intellectual traditions. Each of these disciplines has its roots in a different branch of science and – in its own way – represents an application of the scientific method to policymaking. To take a few examples:

  • Human Centred Design combines the anthropological traditions of ethnographic observation, with experimental processes – also known as prototyping - that are common to a range of sciences, including mechanical engineering, architecture, and software development.
  • Behavioural science draws from behavioural economics and psychology to understand individual choices and behaviours. Like Human Centred Design, it uses experimental processes to test assumptions and hypotheses, this time drawing on the practice of identifying control samples to test efficacy.
  • Data science uses techniques and methods drawn from maths, computer science, and information science to spot patterns and uncover actionable insight from within unstructured data sets. Where the ethnographic traditions of Human Centred Design engage with people’s live experience, Data science looks for truth in the numbers.
  • Systems thinking is itself a collection of disciplines seeking to understand and influence the dynamics within complex human systems. In policy making terms, systems practitioners seek to understand points of maximum leverage and rely upon consensus building, learning, and adaptation, drawing upon a range of different scientific traditions, including complexity science. Again, observation and experimentation are important tools.

Embracing these traditions would help the movement for policy design generate both traction and credibility. But it would also challenge the community to practice these disciplines with the same level of precision that you would expect from any scientist. Any policymaker can – and should – make use of personas, user journeys, and prototypes, but the best results are achieved when these tools are applied with deep knowledge, expertise, and experience.

Develop T-shaped mastery

Mastery matters in part because it enables genuine multi-disciplinarity. IDEO have long pioneered the idea of the T-shaped professional, with deep expertise in one area and the ability to collaborate through engaging seriously with the disciplines of others. In Barking & Dagenham, we are fortunate enough to employ a group of ‘policy design’ practitioners – both within the Council’s Insight Hub but also across other corporate services - with deep expertise in one or more of the disciplines already mentioned. Through working together on real world problems, this group are becoming increasingly T-shaped.

Use science to improve the lives of citizens

I have been part of workshops in which frameworks from behavioural science have been applied alongside insights gained from ethnographic observation, and tools drawn from service design, to come up with new ways of minimising waste and maximising rates of recycling. And I have watched the same group develop a series of prototypes – informed by the practice of identifying control samples - to test the ideas that emerged from these workshops.

I have seen our data scientists develop models to identify who is at greatest risk of missing a payment to the Council, drawing together data sets that include levels of debt, frequency of contact with social care, and prior experience of homelessness. And I have watched them work with our behavioural scientists to pilot assertive outreach models that make use of these insights to ensure residents on the edge get the support they need - before debt turns into homelessness or any other form of crisis.

Together they are transforming the efficiency of the Council’s debt collection efforts. We are now much better at distinguishing between those residents that can afford to pay – with limited impact on their life and circumstances - and those for whom a repayment would tip them over the edge. We generate more revenue whilst providing the right kind of help to those that need it. But don’t just take it from me…

Just had first appointment with client. I explained about the pilot project and asked if she was ok answering how she felt about how we contacted her. Client explained that this has lifted a weight off her shoulders. She said that the support she has had since being contacted has really helped as she was suffering with bad depression and was afraid to open any letters but has now opened them all and has been trying to deal with them.

The next frontier for multidisciplinary working

These results have come from allowing expert service designers, behavioural scientists, and data scientists to practice their craft, working alongside those with deep expertise in policy and/or service delivery. Our next frontier is to make the tools, frameworks, and mindsets that come from these disciplines available to the rest of the organisation – we have a way to go to develop a deeper culture of policy design. But we will be careful to do so in a way that continues to respect the scientific traditions of the disciplines themselves, and that manages expectations about what can be achieved without mastery.

Join our community

We use this blog to talk about the work of the multidisciplinary policy design community. We share stories about our work, the thinking behind it and what policymaking might look like in the future. If you would like to read more, then please sign-up for updates. Join the conversation by commenting below.

Sharing and comments

Share this page


  1. Comment by Elena posted on

    Great article. Very informative, Thanks for sharing.

  2. Comment by john mortimer posted on

    According to the evidence from analysing much of the change work currently going on. Connecting service designers and those involved in change disciplines, is a step that service design has to take. It needs to move away from simply Digital ...and co-design using systems thinking, behaviour science, and complexity.
    Great to see it happening here.

    • Replies to john mortimer>

      Comment by Jonathan Mallinson posted on

      Couldn't agree more John. The disciplines of policy design could and should be used to shape all areas of government policy, including - but definitely not limited to - digital.

  3. Comment by Eoin Quiery posted on

    Thanks for the article Jonathan a really useful summation of ideas and practice. Doing a lot of thinking here at LB Hackney about how we can use systems thinking and existing expertise to better target our behavioural and data science interventions where they will have the maximum transformational impact.


Leave a comment

We only ask for your email address so we know you're a real person

By submitting a comment you understand it may be published on this public website. Please read our privacy notice to see how the GOV.UK blogging platform handles your information.