The Wrong Trousers

“If you’re going through hell, keep going.”

Winston Churchill said that. Except, apparently, he didn’t. The line has been attributed to him for years, but there’s no verified source for it among his writings or speeches. It’s a good example of what British writer Nigel Rees called Churchillian Drift, the tendency for a quotation of uncertain or obscure origin to wander off and eventually attach itself to somebody more famous. Churchill’s particularly susceptible to it, although Mark Twain, Oscar Wilde and Albert Einstein have acquired their fair share of verbal hitchhikers as well.

I came across the expression recently while watching The Diplomat on Netflix, which, as a completely irrelevant aside, I’d highly recommend. Once you know the expression, you start noticing it everywhere. Somebody comes up with a clever line, its origins become a little hazy, and eventually it acquires an author who seems like the sort of person who ought to have said it. Give it another few years on social media and there’ll probably be a black and white photograph of Churchill underneath it.

Something similar happens with proverbs, except that instead of migrating towards famous individuals they migrate towards entire countries. If nobody knows who first said something vaguely philosophical, sooner or later it becomes an ancient Chinese proverb or a Japanese proverb. I can’t find a recognised name specifically for that variation, so I’m going to call it Proverbial Drift, which has the considerable advantage of sounding as though somebody much cleverer than me has already researched it.

If nobody knows who first said something vaguely philosophical, sooner or later it becomes an ancient Chinese proverb or a Japanese proverb

Which brings me, by a slightly circuitous route, to something usually described as a Japanese proverb:

“If you get on the wrong train, be sure to get off at the nearest station. The longer you stay on, the more expensive it will be.”

James Caan CBE posted this recently on LinkedIn. I’d heard it before, but this time it made me stop and think. I’ve no idea whether anybody in Japan originally said it, so it might be an excellent example of Proverbial Drift. But I like it regardless, because it describes something most of us have done. We discover we’ve made the wrong decision and then carry on because stopping would mean admitting that we got it wrong in the first place.

Jobs are an obvious example. People can spend years in positions they stopped enjoying long ago, sometimes in organisations where the culture has become thoroughly toxic. These cultures rarely arrive fully formed with a helpful announcement from HR. They develop gradually, until you learn who can be challenged, which subjects are best avoided in meetings and what happens to somebody who says something everyone knows but nobody is supposed to mention.

You stop challenging decisions you think are wrong because you know where the conversation leads. You watch what happens to somebody else who speaks up and quietly file the information away for future reference. Eventually, you might realise you’re behaving quite differently from the person who first joined the organisation, while Sunday afternoon has somehow developed an atmosphere of its own because Monday morning is getting closer.

From outside, it can look easy. Leave. But real life has an annoying habit of complicating simple solutions. There’s the salary, pension, mortgage and family. There are friendships and the security of knowing how the organisation works. There’s also the possibility that the next job might be just as awful but with a worse commute and nowhere decent to get lunch. Starting again can be frightening, especially when you’ve spent years building a career somewhere.

Starting again can be frightening, especially when you’ve spent years building a career somewhere.

So you give it another six months, which quietly becomes another year. Eventually, the years you’ve already spent there become one of the reasons for staying. Leaving after twenty years can feel as though you’re throwing those twenty years away.

That’s where we run into something considerably less exotic than a Japanese proverb, but probably more useful for understanding what’s happening: the sunk cost fallacy.

A sunk cost is something you’ve already spent and can’t recover. Usually we talk about money, but exactly the same thing can happen with time and effort. The fallacy occurs when what we’ve already invested starts influencing what we do next, even though we can’t change what’s happened.

Suppose you pay £50 to see a show. Half an hour in, you realise it’s dreadful and you’d genuinely rather be at home watching an old episode of Columbo. The £50 has gone whether you leave or stay. Sitting there for another two hours won’t recover it. You can go home having lost £50, or remain in your seat having lost £50 and another two hours of your life.

But leaving feels wasteful because you’ve paid for the ticket. The question is what you want to do with the time you have left, not how much you paid for the evening.

That sounds painfully obvious when it involves £50 and a bad night at the theatre. Add several hundred thousand pounds, a supplier, a project board, senior executives and a few professional reputations, and the calculation becomes considerably more difficult.

The same thinking can keep pointless work going for years. Most large organisations have reports that probably qualify for listed building status. Nobody’s entirely sure why they exist, but they’ve been around so long that removing them feels faintly disrespectful. Ask why one is produced and you’ll often get its biography instead of its purpose. A director requested it years ago. The board used to see it. Somebody might ask for it again. Eventually, you arrive at that wonderfully durable piece of organisational reasoning: “We’ve always done it that way.”

Most large organisations have reports that probably qualify for listed building status. Nobody’s entirely sure why they exist, but they’ve been around so long that removing them feels faintly disrespectful.

There’s nothing inherently stupid about continuing to do something we’ve always done. I’ve been making tea in roughly the same way for decades and haven’t yet seen a compelling case for a beverage transformation programme, although I should probably keep quiet or someone will establish a working group and appoint a Director of Hot Drinks.

But the fact that we’ve done something for years isn’t, by itself, a reason to continue. One useful question is whether anybody would suggest starting it today, knowing what we know now.

Reports aren’t the worst example. Producing something pointless wastes time, but producing something misleading can affect real decisions. An analyst might know that the figures don’t support the conclusion somebody wants to draw, or that a target encourages completely the wrong behaviour. It’s easy to imagine the analyst explaining the problem, everyone thanking them for their expertise and the request being reconsidered. I may have spent rather too long working in the real world to find that sequence of events entirely convincing.

I remember working with a particular government minister who regarded himself as something of a statistical expert. At the time, I was a national director of analysis, trained in mathematics, statistics and epidemiology. In one meeting, the minister repeatedly gave instructions about analysis that I considered misleading at best.

I wanted to challenge it. But the civil servants in the room made the position fairly clear. I could choose to challenge the minister or I could choose to keep my job.

It wasn’t dressed up as a fascinating discussion about statistical methodology, professional independence or speaking truth to power. I was left in very little doubt about the consequences of continuing the argument.

I chose to keep my job.

I’m not particularly proud of that, but I’m not going to rewrite history and turn myself into the fearless defender of statistical integrity I might prefer to have been. It’s easy years later to imagine making a principled stand. It’s considerably harder when you’re sitting in the room, you’ve got responsibilities outside work and the people around you have just explained what that stand is likely to cost.

It’s easy ... to imagine making a principled stand. It’s considerably harder when you’re sitting in the room, you’ve got responsibilities outside work and the people around you have just explained what that stand is likely to cost.

And I think there’s something worth admitting in that experience. Organisations spend enormous amounts of time talking about openness, constructive challenge and evidence based decisions, right up until the evidence becomes inconvenient. Staff notice what happens to people who speak up. If challenging decisions damages careers while keeping quiet doesn’t, an organisation shouldn’t be surprised when its employees gradually become remarkably agreeable.

Cowardice is an easy explanation when it’s somebody else’s mortgage. Sometimes people continue doing things they know are wrong because they understand exactly what getting off the train might cost.

I stayed on that particular train.

Some time later, I made a very different choice with a project. I was sponsoring the development of an IT system for which there wasn’t an obvious off-the-shelf solution. We needed something reasonably quickly, so we went through a tendering process to find a supplier who could build it.

The winning company appeared to understand what we needed and gave us confidence in its ability to deliver. That’s why it won the tender. We didn’t select a company we expected to fail, and there wasn’t a queue of colleagues outside the room warning us of impending disaster.

But once the work started, the reality wasn’t quite so impressive.

There wasn’t a spectacular collapse. In my experience, there rarely is. The problems emerged gradually. Things weren’t ready when expected, delivery didn’t match what we’d anticipated, and meetings produced explanations and reassurances. Eventually, the collection of individual problems started telling us something none of them quite said on its own.

I became increasingly convinced that the supplier wasn’t going to deliver the system we needed. To be fair to them, they admitted they’d underestimated the project's difficulty. But by then, we’d spent a significant amount of money and faced the temptation to carry on because stopping would mean accepting that money had been wasted. So we’d arrived at a fairly textbook example of the sunk cost problem, although I can assure you that these things feel considerably less textbook when you’re actually sitting inside them.

we’d arrived at a fairly textbook example of the sunk cost problem, although I can assure you that these things feel considerably less textbook when you’re actually sitting inside them.

Imagine a project that has spent £200,000 and requires another £500,000 to complete. The original money has gone whether you stop or continue. The question is whether spending the next £500,000 makes sense given what you know now. If it’s likely to produce something worth the additional expenditure, there may be a perfectly good reason to continue.

But if the justification for spending another £500,000 is that you’ve already spent £200,000, you’re not rescuing the original investment. You’re simply finding a remarkably effective way of turning a £200,000 loss into a £700,000 one.

That was the position we faced. We could have revised plans, moved deadlines and lowered expectations until we eventually produced something called a Minimum Viable Product.

I’ve got mixed feelings, like a lot of people, about that term. An MVP can be useful when it means an early, workable version of something that can be tested and developed. But on a project that’s already in trouble, it can become the name given to whatever remains deliverable after the original expectations have been sufficiently lowered.

So, yes, we’d eventually have received something. But it would have cost considerably more, arrived considerably later and wouldn’t really have been what we needed. At least we’d have had a system.

And that matters more in organisations than it probably should. You can demonstrate a system or put screenshots into a presentation. Somebody can write “delivered” in the project closure report. Whether anybody actually wants to use it can be left for somebody else to discover, preferably somebody who wasn’t involved in the original project.

The project owner and I decided we weren’t going to do that.

We stopped the project and cut our losses. We couldn’t recover the money already spent, but we could decide whether to spend more. On the evidence available, we couldn’t justify doing so. We’d got on the wrong train and realised it while there was still time to get off, so that's what we did.

We’d got on the wrong train and realised it while there was still time to get off, so that's what we did.

I expected questions, and questions would have been entirely reasonable. We’d spent a significant amount of money without getting the system we wanted. But what followed felt very different.

There was a substantial internal investigation, and those of us involved were treated as though we’d done something seriously wrong. At times, it felt almost as though we’d misappropriated the money or behaved dishonestly, rather than made a difficult decision to stop spending it on something we no longer believed would deliver.

We hadn’t taken the money or hidden what was happening. We’d stopped the project because we thought continuing would make the eventual loss larger.

The investigation itself consumed considerable staff time. Meetings and decisions were revisited, events reconstructed and questions answered. More resources disappeared into a project that had been stopped to prevent further resources being wasted.

And there was a reputational cost that lasted much longer.

You see, senior directors didn’t necessarily know the detail. They didn’t know everything that had happened with the supplier, what we’d originally been told, what became apparent as the work progressed or why we’d eventually concluded that stopping was the least bad option. Some knew something much simpler: a lot of money had been spent, there wasn’t a system, we’d stopped the project and we were being investigated.

That’s quite enough information to do significant damage. Once that version starts circulating, the details have a difficult journey trying to catch up.. Years later, somebody might remember your involvement in “that project that was investigated” without knowing why the investigation took place or what it established. And they may never have known in the first place.

I’ve wondered many times what would have happened if we’d stayed on the train.

Suppose we’d spent considerably more, missed the original timescale by a year or more and eventually delivered an inadequate MVP. Would we have been treated in the same way? Probably not. The NHS and Government have a long list of failed or partially delivered programmes.

The NHS and Government have a long list of failed or partially delivered programmes.

There would have been complaints about the overrun, difficult meetings and lessons learned, because lessons are always learned shortly before being carefully documented and put somewhere future projects can avoid finding them. But at least we’d have delivered something. From a career and reputation perspective, we might have been safer.

That’s a concern because organisations teach people how they expect them to behave through what happens to those who make difficult decisions. If stopping a failing project leads to an investigation while continuing it leads to another budget allocation, people notice. It doesn’t particularly matter what the project management handbook says about openness, challenge, governance or responsible use of resources. People watch what happens to the person who puts their hand up and says, “I don’t think this is going to work.” If the result is that they're treated as though they’ve done something improper, the message travels surprisingly quickly. And the next project sponsor might keep quiet for another six months.

And there’s the sunk cost fallacy again. Another six months doesn’t recover the previous two years. Another £500,000 doesn’t recover the previous £200,000. Another ten years in a job you hate won’t give you back the previous twenty. The past explains how we arrived where we are, but it shouldn’t automatically decide where we go next.

The past explains how we arrived where we are, but it shouldn’t automatically decide where we go next.

And that brings me back to Churchill.

“If you’re going through hell, keep going.”

The quotation he apparently didn’t say has survived because it sounds good and, under the right circumstances, contains perfectly good advice. If what you’re doing is worthwhile and you’re still heading somewhere worth reaching, giving up simply because things have become difficult may be the worst decision you can make.

But the proverb tells us almost the complete opposite:

“If you get on the wrong train, be sure to get off at the nearest station. The longer you stay on, the more expensive it will be.”

One says keep going. The other says get off. And the thing is that both can be good advice. The difficult part is knowing which one applies.

There’s a considerable difference between going through hell on the way to somewhere worth reaching and going through hell because you’re travelling in the wrong direction. A difficult project with a reasonable prospect of success might deserve perseverance. One where continuing will simply make failure more expensive probably doesn’t. A difficult period at work might be worth riding out; ten years in a culture that’s destroying your confidence is another matter.

That judgement is much harder than either quotation makes it sound.

“Keep going” is particularly seductive inside organisations because perseverance and commitment sound admirable. Nobody wants to be known as the person who gave up, and leadership literature doesn’t generally overflow with inspirational stories about somebody taking a long look at a project and deciding to pull the plug.

Nobody wants to be known as the person who gave up, and leadership literature doesn’t generally overflow with inspirational stories about somebody taking a long look at a project and deciding to pull the plug.

Getting off the train also creates a visible moment that continuing doesn’t. Somebody has to make the decision and put their name against it. Continuing can happen almost by itself, supported by another meeting, another revised plan and another promise that things are nearly back on track.

Nobody has to decide to continue forever. They only have to decide not to stop today.

Which finally brings me to the trousers.

I called this article The Wrong Trousers, rather than The Wrong Train, because Wallace puts on something that’s supposed to help him and gradually discovers that the trousers are deciding where he goes. And I think organisations accumulate their own versions. We create projects, processes, reports and systems to serve us, then somewhere along the way they acquire enough history and momentum that we start serving them.

We create projects, processes, reports and systems to serve us, then somewhere along the way they acquire enough history and momentum that we start serving them.

A project we chose to start begins deciding how much more we’ll spend because we’ve already spent so much. A report becomes something that must be produced because it’s always been produced. A process designed to solve a problem survives long after the problem has disappeared. A job somebody freely chose twenty years ago becomes a job they can’t imagine leaving because of the years they’ve already given it.

At some point, the thing we chose starts choosing for us.

Do I regret stopping the project? We’d spent a significant amount of money without getting what we’d paid for, and I believed continuing would make matters worse. What was harder to accept was being treated afterwards almost as though we’d done something improper with the money, when our decision had been to stop wasting more of it. No, knowing what I knew at the time, I’d make the same decision again.

Perhaps the quotation wrongly attributed to Churchill and the proverb possibly wrongly attributed to the Japanese should be merged together:

If you’re going through hell but you’re still on the right train, keep going. If you’re going through hell because you’ve boarded the wrong train, get off.

Don’t stay aboard simply because the ticket was expensive or because you’ve already travelled a long way. Getting further from where you wanted to be won’t recover either.

Don’t stay aboard simply because the ticket was expensive or because you’ve already travelled a long way. Getting further from where you wanted to be won’t recover either.

When we looked out of the window, we realised we were on the wrong train. So we got off. We were criticised, investigated and suffered reputational damage because of it. Staying aboard might even have been safer personally, but it still wouldn’t have made it the right train.

Of course, had we stayed on long enough, spent enough and finally arrived somewhere we’d never intended to go, at least we’d have had something to show for the journey.

A very expensive pair of the wrong trousers.