Asking ‘Why’ as a Career

A confession

I have a confession to make. It’s been bothering me for some time, and now I’m going to get it off my chest. I’m an imposter.

I know everyone has imposter syndrome these days, but I really am. You see, people believe I’m a transport modeller, and I let them. Some people even think I’m a transport planner, which I rarely object to. But I don’t have formal training in either field. I’ve not even worked in transport all my career like most transport modellers and planners. I started off in a small research group in the Post Office, then moved on to modelling water treatment and leakage at Thames Water, before stumbling (luckily) into transport modelling via a stint in the rail industry.

But I’m not telling you this because I’m ashamed or feel diminished. I want to explain why I’ve realised it’s a good thing, and why more of us should be proud imposters.

Two types of expert

First of all, I want to repeat something I’ve often said about transport modelling, which is that we need two sorts of transport modellers, coming from two different directions. If you’re not a transport modeller (or not currently pretending to be one) stick with me, this probably applies to your profession too.

First, there are transport specialists, who we can call domain experts. These are people with expertise and interest first and foremost in transport, who do modelling as a tool to solve transport problems. They have vital understanding of the context for models, such as engineering and design, knowledge of processes (e.g. TAG and TUBA), client contacts, organisational awareness and just knowing ‘what works’ in transport.

Then there are the modelling specialists who, along with programmers and data scientists, we can call technical experts. These are the people whose careers focus on capturing the data, processing the numbers, and producing meaningful outputs to aid decision-making, short- or long-term.

I hope it’s obvious that we need both types of expert. The transport domain experts know the industry and the problems, and they know WHAT they want to achieve through modelling. ?The technical experts know HOW to build new models efficiently, and apply fresh techniques.

In each of the industries I’ve worked, I’ve seen the magic that happens when domain and non-domain experts come together. Creative and well-targeted solutions to real problems emerge, taking advantage of new technology and techniques, whilst learning lessons from experience. (Of course, it goes without saying that many talented people comfortably play both roles, though normally one side is dominant.)

When experts go bad

I’ve also seen things go very wrong if either side dominates too much. If domain experts have too much power, there is too much emphasis on existing processes (‘the way we do things’), leading to systems not being updated, a lack of fresh ideas and an inability to respond to new challenges.

If the technical experts have too much power, then modelling and analysis quickly becomes unhelpful; models don’t address the key questions, overlook real life constraints, or work focuses on what’s technically interesting over what helps.

And there are (many) other ways the relationship between domain and non-domain experts can fail. Sometimes the model is treated as a ‘black box’ by domain experts, so that they don’t engage with the fundamental truths the model is representing (or maybe failing to represent). Or the technical experts oversell new and innovative methods as a panacea for existing problems, overlooking simpler solutions.

Where am I in all this?

As I’ve moved between industries, I’ve often been unable to identify as a domain specialist, so have tended to act as a ‘technical expert’ side. And unsurprisingly I’ve suffered from the imposter syndrome that comes with being a newcomer in an industry: “I can’t possibly challenge decisions round here until I know EVERYTHING the domain experts do!”.

However, as I’ve grown to know more about transport and the context of modelling, I now feel I know less about the ‘technical’ side of modelling. I’m not able to operate the models, programme in ‘modern’ languages, or do whatever the data scientists do. And there’s really no going back, I’m a senior modeller, a project manager, aspiring perhaps to be a ‘thought leader’ and mentor to the next generation, but not a ‘doer’.

Back where I began

Well, maybe it comes to us all, maybe most modellers and analysts ultimately become ‘domain specialists’: you know the industry you work in well, you realise WHAT needs doing and can manage the process; you’re less able to bring the HOW, but then it doesn’t matter if you can manage the people who do.

However, I think there is a third way, which a role that all too often gets forgotten between the WHAT and the HOW. It’s something that takes me back to my original training, and also explains my habit of asking awkward questions throughout my career!

?You see, I trained in Operational Research (OR), sometimes known as Management Science. OR is a discipline which involves analysing real world problems, usually with numbers involved but not always. It has always been hard to define: is it a branch of computer modelling, a pre-cursor to data science, a set of consultancy tools, or a bunch of nice heuristic algorithms?

Possibly because it is hard to define, many OR people drift during their careers from identifying in OR to being either a domain expert or a technical/analytical expert. But some OR habits never leave you, and for me the critical thing is that OR people don’t just ask WHAT or HOW. They ask WHY: Why do you want to do that? Why is this the normal approach? Why is this important? Why is that not important?

They also ask WHAT IF and HOW ABOUT, think about what solutions we really need, and strive to understand the whole process to produce solutions. Of course, this requires some appreciation of the WHAT and the HOW, but crucially a little distance to stand back and ask sensible questions.

One good friend and distinguished transport domain expert often talks of the ‘Philosophy of Transport Modelling’, and another highlighted to me the increasing need for ‘diversity of thought’ in transport planning. I think these are both great ways of thinking, which fit nicely in this third area.

It’s increasingly recognised that transport modelling needs to think really deeply about the right approaches, avoid falling into the trap of narrow-thinking, but not be distracted by new techniques which promise much but may not deliver.

Be a thoughtful imposter

So I’ve finally learned to embrace both my OR background and my itinerant career. It’s a position where I can happily step back at any point and ask important (awkward) questions, which I love doing!

Now, I’m not saying that WHY is more important than WHAT or HOW (see Figure 1). And I’m certainly not saying transport experts and pure modellers are incapable of stepping back and asking WHY. We all contain a mixture of these different approaches, applied at different times and with different emphasis.

What I am saying is that imposter syndrome strikes us all at some point. It’s a sign of our own self-awareness rather than weakness. An expert in a specific domain knows the standard processes and what hasn’t worked before. A technical or analytical expert knows how to develop and use new tools. But provided you ask the right questions you can come from either side, or even from outside, and help be part of the solution.

Asking WHY is an important part of our toolkit, and that knowing less can be turned to an advantage. Long live the thoughtful imposter, and long live Operational Research.

Postscript: Tim Gent is happy to be identified as a Transport Modeller, a skill he has been learning since 2005.