How do deliver change

Whether it is the climate crisis, economic crisis or the crisis within the health service, the need for change is obvious. In fact, the need for radical change is one of the few things people across the political spectrum seem to agree about. There is certainly no shortage of people promising to deliver change.

But how easy is it to really change things? And I mean, change them so that the new system is so much better than the old one, that it justifies the trouble of the whole process.

Spoiler: Very difficult.

My friend and former colleague Tony Duckenfield has done some interesting research on how to get more people walking and cycling.

He sets out a five step process for change in that area:

1. Understand the current situation - using data, observations, and research;

2. Design the intervention taking into account the wider context;

3. Test options where possible;

4. Implement, with clear objectives and targets, a project plan, sufficient resources, and a plan for engaging with local people & stakeholders;

5. Monitor & learn.

It is well worth reading the full note on his website.

What strikes me is how generally applicable this process is, well beyond walking and cycling.

In recent years I have done a lot of work in investigations. A typical chain of events is:

1. Something bad happens

2. Everyone agrees that something needs to be done

3. Management/government do what they wanted to do anyway, using the event as a justification (often irrespective of whether it relates to the event itself).

In short, the opposite of the sort of process that Tony suggests.

As Tony says, “the biggest single problem is that the solution tends to come first - someone wants to build (or fund) a cycle lane so looks to justify that irrespective of whether that's appropriate to the context. There's very much a more haste less speed challenge here!”.

Now in many areas there are well organised processes for investigating problems and evaluating solutions - from pedestrian crossings to medical interventions. But how much do we apply that kind of measure to management decisions? I have lived through enough "restructurings" to know that the answer is not very many. Certainly the re-organisations of the NHS that I have seen didn’t trouble themselves with an evidential basis!

Who has an incentive to discover what solutions are really effective? This is a particular issue where lead times are so long that no-one knows or cares whether the proposal is actually effective or not. Leadership elections will be long gone before the efficacy of proposed solutions becomes clear. If building a new tram or rail line is popular, who is interested in whether it is effective or not?

I'm not sure how you can break these misaligned incentives which lie behind policy often being ineffective. We can only start by looking at decisions that are made and asking what data lies behind them.

The prevalence of problems facing us at the moment demonstrates that if we don’t tackle problems in a systematic way, they just don't go away.

Time to tackle them for real.