Big Picture

Why is Management So Difficult?

Feature image for Why is Management So Difficult?

When I first became a manager, I was struck by how counter-intuitive it was to get better. It seemed a lot harder than levelling up as a writer, or as a programmer: skills that I considered myself reasonably good at.

Management is difficult, I think, for two main reasons: the feedback loops are long, and the management as a discipline deals with people: complicated, messy, political beings.

Long feedback loops

When you're an individual contributor in a knowledge work company, it's often easy to get feedback for your work. If you're a journalist, you get feedback shortly after you publish, or at least shortly after you submit to your editor. If you design websites, you get feedback from your next round of user tests, or after your next client meeting. And if you code, you often get feedback the next time you attempt to run your program.

But management is a whole different creature. When you make a mistake as a manager, there's sometimes a span of months before you get the feedback that tells you that you've screwed up. And unlike other forms of work you used to do before you became a manager, some of these mistakes can't be fixed once they've happened.

I'll give you an example of a mistake like this.

In my previous company, we had a levelling policy that required level 1s to mentor another junior employee (or an intern) as a prerequisite for promotion. An engineer of ours had met all the other requirements for promotion come promotion season, but our head office in Singapore had ordered a hiring freeze for a few months. We were now put into a quandary: with no other junior engineers or interns incoming, should we promote the engineer in question, overlooking this one final requirement? Or should we hold off on promotion, possibly alienating this highly deserving employee and increasing the risk that he would quit?

My managers leaned towards overlooking the final requirement. There was also some discussion about how such a requirement was highly problematic given how our company might do hiring freezes again in the future. Perhaps the time was ripe to do away with this requirement?

I disagreed. I argued that the question of doing away with this requirement was a separate issue. In my view, I was more worried about the specific situation of this engineer — he was a promising software engineer, and we'd already invested a year in his training.

I was also worried about the precedent we might set. If we waived the requirement for this engineer, there may come a time where another software engineer would ask for a promotion waiver for some other category, citing extenuating circumstance and pointing to this incident as proof that we were willing to bend the rules. In such a situation, a manager would not have any defence against this charge. An explanation of "but that situation was more extenuating than yours” would not fly, and would raise the accusation of unfairness.

Sure, a manager might be able to repeatedly say "this is the last time, it was a special occasion", but over time the entire basis of the promotion system would have been lost. And in that case, why bother with a common standard for promotion? Why not just revert to the previous system where better negotiators got rewarded with promotions, leaving the better software engineers behind?

We eventually solved this problem by transferring a junior engineer from a different team to be mentored by this employee. But this was a lucky situation for us. If it weren't possible to do this, I argued that it was better to lose this employee than to have the entire levelling process be questioned a couple of months down the road.

This is a perfect example where the easy solution would have led to a disastrous outcome for the company in the long term. And in a couple of months, when the fairness of the whole system was called into question, the managers who had to work in this system would be left at a loss at what to do to solve it. Worse, they might not stop to consider that the downstream problems were caused by one wrong decision made months ago. Better to prevent such problems from happening in the first place.

Dealing with people

People are messy. They're illogical, emotional, political beings who can feel hurt, jealous, or may (god forbid) hold grudges against you. Management can be highly problematic if you're not able to deal with or prevent these problems.

The base skill in management is the ability to motivate people. There are good ways and bad ways of doing this. The problem is that the good ways and the bad ways are highly dependent on context — meaning that it depends on who you are, what your personality is, the kind of organisation you are in, and the culture you and your subordinates come from. It is extremely difficult to learn these skills from books. Worse, it is impossible to apply management techniques you've learnt without first going through the filter of inter-personal skills: a filter that you have to supply, yourself.

The upshot of this is that people with bad interpersonal skills tend to make for bad managers. It's difficult to motivate people without a good mental model of their internal worlds. Similarly, it is impossible to be a good manager without also being a good communicator. Both demands you to be able to empathise with and win the trust of subordinates.

My problem, though, is that interpersonal skills are incredibly hard to develop. You need to learn a large set of disparate, hidden skills in order to get better. There are no quantitative methods for evaluating your skills — and in fact most people aren't equipped to evaluate your interpersonal skills beyond “oh, he's easy to work with!” or “ehh, he's prickly". In fact, the higher you rise in your career, the more unlikely you'll receive actionable feedback on your interpersonal skills. This is something you have to develop yourself — but, like most things in this blog, I believe can be accomplished given enough time and effort.

I like to use interpersonal skills as a rough metric when training an individual contributor in management skills: is the individual respected by most of his or her peers? Does he or she know how to gain such respect without the mantle of their position? Is he or she able to resolve disputes effectively? If so, it probably pays to train this person in such skills in addition to the traditional set of management skills like ‘evaluating leverage’ or 'delegating effectively’.


Management is difficult because a) the feedback loops are long, and b) people are complicated.

The first reason explains why it’s easy to get stuck in a rut in management. As an individual contributor, it was clear to me when my manager made mistakes. But it wasn’t clear to them, because it was often awhile before the consequences of their actions percolated back up to them. This distance between action and consequence guarantees that only the most introspective managers improve over time.

The second reason — that people are complicated — is why management training courses have wildly different outcomes depending on the individual. Good managers get better after reading management books or attending courses, while bad managers languish in lousiness. The difference is usually explained by the amount of social ability these people had to begin with.

The good news, though, is that these reasons inform our efforts to improve.

First, if management is difficult because it has long feedback loops, then it directly follows that we need to set aside time to be introspective, in order to find the root causes of our problems. (As an aside: ever wondered why SWOT analysis or ‘chain of whys’ are so loved by management consultants? My take is that they're loved because they're forced introspection).

Second, if you're a manager who happens to be bad with people, the right answer might not be to buy more management books, or attend more management courses. The right answer might be to take things a level down, to the level of inter-personal skills, and improve at that level. You're not going to get better at management until you gain these sub-skills, anyway, so save your time and energy learning management skills, since these would be ineffective without a proper foundation.

This entire site is about getting good as a manager. I'll be talking about how one can improve in a bit. Stay tuned.