A catalogue of things that are stopping change

James Reeve
Mar 4 · 6 min read

By James Reeve and Rose Mortada

This is part of a series of posts about how digital leadership is changing. Last time we talked about some trends we’ve observed. This post is a two-parter with this bit coming today and Rose’s bit coming shortly after.


GDS showed us the way. The low hanging fruit is now gone. We’re increasingly seeing the need to:

  • develop services and policy in tandem because services are an expression of policy
  • offer services without making them ‘mandatory’ — and make them so good people choose to use them
  • reflect on and change up our approach to leadership and governance

The problems

In this post we describe a set of difficult problems. These are cross-government issues that we think civil servants are longing for help with. But these aren’t just problems for the civil service. These are very human problems which will be present in any large organisation.

We see three themes in the problems:

  • bringing together policy and delivery (in this blog)
  • engagement with ministers and influencers (in Rose’s blog )
  • governance (in Rose’s blog)

Bringing together policy and delivery

If I could push a button and solve one problem in government, this would be it. It’s the root cause of many team dysfunctions and often delivery failures. It’s difficult because it means changing how thousands of civil servants see themselves and relate to others.

We’ve given some of the problems we see names so that we can talk about them more openly.

The Best Reasonable Choice

Sometimes there’s no time to do all the research we’d like to do before deciding on policy. Lots of people in the civil service have become excellent at making ‘Best Reasonable Choices’. These are well intentioned but ultimately limited choices based on what they know and their own life experience. It’s understandable: We expect it and reward reasonable choices. We recognise it as a valuable behaviour and even call it ‘making effective decisions’.

A extract from the civil service competency framework: Making Effective Decisions
A extract from the civil service competency framework: Making Effective Decisions
Making decisions with incomplete evidence is a useful — but risky — skill.

This means that often in policy we value people who are excellent at proposing new ideas to ministers or senior stakeholders — despite incomplete evidence. This risks people becoming more attached to their own idea than the problem they’re trying to solve. We seek rational arguments that support our intent over the process of testing our ideas in the real world and being open to scrapping our initial idea.

Then once a decision is taken, a strategy published or a target announced, the policy is often considered ‘done’: Done well before the outcome has been realised in the outside world. This means that attention naturally falls away before we can know whether the policy will live up to its intent.

Delivery arrogance

Unsurprisingly, people who do delivery very well form a high opinion of themselves. They sometimes think they should have a go at making some ‘Best Reasonable Choices’ themselves. This can end up looking like people with post-its and MacBooks telling established policy people how to do their job using ‘Agile’ and ‘canvases’. It’s not a good look.

If digital people choose to have a go at making policy or user-centred design labs, they should be careful not to fall into this trap. Tools and techniques that work well for service delivery don’t always naturally transfer to policymaking.


The civil service is set up around individual performance. Pay & reward, line management and performance structures incentivise individual performance. They disincentive good cross-team working towards shared goals:

  • pay and promotion is often assessed on the basis of individual performance, not contribution to a team.
  • accountability is linked to posts, not people: So with civil servants moving regularly, there’s little incentive to plan ahead for sustainability or delivery.
  • line management can be used as a source of power, (with objectives and responsibilities flowing downhill) rather than being used as a way to support good people.
  • paying the market rate for specialist skills is difficult and can lead to grade inflation, often confusing skill with leadership. This can set up unhelpful hierarchies or create bad line management.

So people aren’t incentivised to share in success or make it easy for others to learn from their mistakes. This undermines effective team working and sets up unhelpful competition between people for attention and responsibility.

Policy-delivery conflict

It’s common in policy circles to hear the phrase “Let’s not make ourselves hostages to fortune”. What this means is “If things change and we have to do things differently, let’s not make things too hard for ourselves”. Wise words. But in practice policy teams often have to launch national policies and can’t experiment in small increments. What they can do instead is avoid closing off options. When you combine this with the need to announce things, some policy can become an exercise in creating uncertainty.

When I was a policy person I was once told by a senior manager to ensure that when I spoke, I allowed for a ‘cone of uncertainty’ about what I said to project into the future. “Don’t try to create certainty” he said, “it’s not yours to create”.

So we put off making irreversible decisions for as long as possible — but don’t do enough in the meantime to get closer to that decision.

A cartoon showing a person speaking with a cone protruding into the future.
A cartoon showing a person speaking with a cone protruding into the future.
Me 5 years ago

On the other hand, good delivery is about removing uncertainty through experimentation. This means we should know what will happen before we set out to spend larger sums of taxpayers money: Agile is for people who don’t like surprises.

An antique photograph of Marie Curie
An antique photograph of Marie Curie
Marie Curie in her laboratory — putting experimentation first

For example, think about what we often ask policy people to do — take long and complex plans and generalise them into one compelling sentence for a minister to say in a speech. On the other hand, we ask delivery people to do the opposite: take this one sentence the minister actually said and then flesh it out to deliver a feasible, usable, accessible and valuable service. The status quo of the civil service literally asks delivery and policy people to do conflicting, often opposing tasks. No wonder things don’t always gel.


Given all of the above (including tribes of people behaving as individuals), there’s a friction that arises between tribes, partly from disparity in the perceived value of the policy and delivery professions. There’s glory attached to the policy profession and advising ministers. This makes delivery tribes feel resentful, like the poor relation. At the same time, delivery professionals are seen diminutively as the ‘builders’ or ‘technical’ (clutching an oily rag) and by using different language (jargon), we often don’t help themselves.

And what’s worse, sometimes people inside tribes don’t even talk to one another. If I had a penny for every time I’d heard someone in the civil service say ‘we just need to join up more’, I’d be a gardener by now. Joining up is a lot harder than it sounds.

Picture of an allotment garden which needs a lot of work.
Picture of an allotment garden which needs a lot of work.
It needs more work but I’ll remain a civil servant for now

Things that could make a difference

We believe:

  • evolution on this scale will take a long time to pay off and involve making some painful changes: This is first and foremost a leadership issue.
  • though difficult to implement, the solution is simple: shared goals, values and a process to achieve them are what’s needed.
  • we’d like to experiment with team-based recognition over individual recognition.
  • the policy should never be considered done until the outcome has been achieved.

This post is a two-parter, so find the second part covering ministers and influencers posted by Rose Mortada here.

We hope this doesn’t sound too negative: It shouldn’t — the civil service is already doing great things. The final post in this series will look at the new forms of leadership will tackle the problems in this catalogue head-on.

James Reeve

Written by

Husband, Dad, Chemist and Civil Servant @ DfE, all views my own.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade