We Need to Talk About How We Make Decisions
This article previously appeared in Transport Times
Hands up who feels the way we make decisions in transport today is as good as it can get?
No one? Thought not.
Yet this is a huge part of our problem. We tell ourselves that the big challenges in our sector are political, financial or structural, but solving these requires us to make decisions. And what we're really grappling with is a crisis in decision-making.
But here's the good news: decision-making processes were made by humans and they can be remade by humans. They're not fixed laws of physics.
In my career, I've worked with amazing teams across both public and private sectors. Yet, somehow, put those great people inside our system, and we all seem to move too slowly. I'm not blaming anyone else. I've been a director of transport companies since I was 27: it's my fault as much as anyone's and more than most. But we now need to fix it.
The good (or bad?!) news is that this isn't just a transport problem. It's not even just a UK problem. It's a global phenomenon, described from different angles in books like The Geek Way, Failed State, How Big Things Get Done, Mission-Driven Bureaucrats, and The Friction Project. These books describe everything from the UK public sector to the US private sector. What links them all is the insight that large organisations are finding it increasingly hard to decide.
Now, let's be honest, it is somewhat harder for those of us in transport, living in the shadow of a long history. This year we're celebrating 200 years of the passenger railway. That means we've had two centuries of well-meaning, smart, diligent people adding governance, sign-offs and processes. It's a ratchet: easy to add, hard to remove. Each new layer solves a real problem. But collectively, they weigh us down.
Then, just to make things harder, we're drowning in data. In 2002 - the year I started my career - the world produced 0.0005 zettabytes of data. Last year, it was 147 zettabytes of data. That's a three million percent increase in data production in my working life. Data is no longer scarce. It's infinite. And that makes the act of deciding even harder. Because now, we're not paralysed by what we don't know. We're paralysed by everything we could know, if we just commissioned one more model, ran one more consultation, requested one more spreadsheet.
Meanwhile, technology companies are making decisions faster, bolder, and often more effectively. And they're not small: Amazon employs 1.5 million people and operates 100,000 vehicles. We often dismiss firms like Amazon as “just software” but it is, in every meaningful sense, the world's largest transport operation. But unlike many in our sector, they've taken decision-making seriously - as a core competency. They've built processes that support pace, agility, and responsibility. And we can borrow their tools.
Here are four tools (some, though not all, originating in Amazon) you can start using tomorrow:
1. One-way door vs two-way door
Amazon categorises decisions as either one-way doors (irreversible, high-stakes) or two-way doors (reversible, low-risk). One-way doors need process. Two-way doors? Just open them. If you get it wrong, walk back through. Most of our sector treats every decision like a one-way door - and that kills momentum. It's like treating getting a haircut with the same caution as getting a tattoo.
2. Minimum Viable Business Cases
If something might work, test it fast and cheap. Don't build a full business case before you know the basics. I once saw £13 million spent on a service that could have been disproved for £50k. We don't always need more modelling - we need more learning.
3. Break the rules (when you know why they exist)
If a rule is blocking a team from achieving a goal - and they understand the rule's intent and context - empower them to break it and tell you afterwards. We are a compliance-driven sector, so you won't create chaos by saying this. Most people will need persuasion to do it even once. But if you can get them to try, you'll unlock initiative.
4. Decide at 70%
Amazon encourages decisions to be made when 70% of the information you wish you had is available. The final 30% is usually diminishing returns. The cost of certainty is too high - and in a fast-moving world, delay is riskier than action.
This stuff isn't easy. But it works.
It's the reason I started Freewheeling and decided to focus myself on helping organisations change how they decide. I now run workshops, awaydays and strategy sessions with transport and mobility teams across the UK. We don't "train" in the traditional sense. Instead, we explore different ways of making decisions to find those that work in your context. The outcome, when we get it right, is faster decisions, more empowered teams and execs who finally get time back for the things only they can do.
Why now? Because this decade is our moment. Great British Railways is being created. Bus franchises are being rolled out. Combined Authorities are being empowered. New political leaders are taking the reins. Governance is shifting. In the next five years, more new transport organisations will probably be created than in the last twenty. But here's the rub: if those organisations inherit the old decision-making culture, nothing will change. We'll spend five years restructuring, not delivering.
To do that, we need more than new organisations. We need a new language of change; a shared vocabulary for agility. The world I'm trying to build through my work is that we all understand this shared language. So that if someone at a combined authority says to an operator, "Let's just do it — it's a two-way door", the operator will know what they mean. Or if someone in your team says to you, "This came out of our pre-mortem", you'll know what they did and why.
If you've got a team that you lead or an event coming up, I'd love to come and teach you this new language of change.
Because before we can transform transport, we need to transform how we decide.
If this post has resonated with you at all, just click here to put 10 mins in my diary for a chat about how I might be able to help
