Say your boss asks if you can handle a couple more members on your team. Or say that — for whatever reason, the manager of an adjacent team leaves and you’re expected to fill in her shoes temporarily. The other team is five people; you’re currently managing three. Should you say yes?
In management circles, it is common knowledge that the ideal number of direct subordinates a manager should have is 7±2 (say it with me now: “seven plus or minus two!”)
Some people prefer five direct subordinates, so that they can spend more time on individual contributor work. Others are incredibly organised superhumans, and are able to handle the maximum recommended number: nine. But having more subordinates than nine is probably a bad idea.
There are two super interesting reasons for this special range of numbers. The first is the reasoning that former Intel CEO Andy Grove gives in his book High Output Management: essentially, that managers are expected to spend about an hour per report per week, summed up over all their interactions and activities.
These nine hours — plus whatever incidental number of hours you get from coordination costs in a nine-person team, should take up a solid quarter of the manager’s entire work week.
(Assuming a 40 hour work week, 9 hours is nearly a full quarter of your time).
A full quarter is the maximum most managers can bear. This is especially bad if you have individual contributor work on top of your management work, or if you hold other responsibilities in your company.
The second reason has its roots in a surprising place: neuroscience. In 1956, cognitive psychologist George Miller demonstrated that the limits of human working memory is 7±2 items. This explains why most phone numbers are between seven to eight digits long, and why most meetings are productive up to a certain number of participants. (Hint: you can only productively hold a conversation with up to 7±2 people).
7±2 has also been used to explain why professional sports teams never have more than eight people per function (for instance, six forwards and six backs in rugby). While this result has never been conclusively proven, the general argument is that people’s ability to keep track of their peers is constrained by the size of their working memory.
Actionable Takeaways
So let's circle back to the original scenario at the top of this blog post. If asked to shoulder the burden of managing more direct reports, should you say yes or no to your boss?
The right answer here is, of course, that it depends — and that it depends on you. What’s your limit? What changes would you have to make to handle the increased communications overhead from a larger team? Do you still want to contribute as an IC in addition to your managerial responsibilities?
I tell my new managers to experiment within five and nine direct reports, with the goal of finding their personal limits. It is incredibly valuable to know where your limits are — and to establish this knowledge fairly early on in your managerial career.
Knowing one’s limits helps because it allows you to mould your role to fit your career goals. Want to program? Take on less subordinates! Want to increase your team’s impact? Take on more people — up to your limit, that is.
My limit is eight people maximum, without any IC duties — that is, no programming work for me — and five direct reports if I do want to program.
I’ve learnt this the hard way. Sometime in mid-2016, one of the heads of the two teams in my company decided to leave. I took on his team in addition to mine — for a total of 13 direct reports. I became so overwhelmed with work during those two months that I found myself at the office on nights and weekends. I swore I would never allow this to happen again.
If you ever find yourself in a similar situation, do as I did and figure out a way to turn one of your direct reports into a manager. In my case, I set aside a large block of time to find a senior engineer to replace the team lead who left, and I offloaded part of my responsibilities to the most senior engineer on my team.
In this manner, you get to maintain your direct report limit, while still managing the output of the broader organisation.
The upshot isn’t that you should do as I did and jump in the deep end. The upshot is that you should test yourself while you still can — and for most people, this means at the beginning of their careers, where the stakes are lower. As you add new hires to your team, observe your workload. Are you comfortable with it? Has it grown too unwieldy? Would you prefer less people?
Knowing your limits allows you the ability to form your own practices in a thoughtful, explicit way. It prevents you from having these choices made for you.