How Many People Can Someone Lead?
patkua.com
patkua.com
She's a nationally board-certified kindergarten teacher in a public school.
I'm most familiar with China (I taught middle-school classrooms of 55-60 in Yunnan around 2010, and had some friends in Guangzhou teaching 75-80), but I assume there have been & are plenty of countries with similar situations.
Faculty at colleges get teaching assistants, but their job is hardly to help with conflicts or prevent violence, their job is to help with instruction, grading, writing homework assignments, etc. When I was a teaching assistant, I myself tended to have 10-20 students in my section, and the class as a whole had close to a hundred people.
It's crazy how underpaid teachers are.
The point was what makes it worthwhile isn't cash.
I think a lot of teachers do value the experience of teaching enough that they'll stick with it through thick and thin - but a lot of folks never even consider teaching due to just how poor the wages are.
Most people are not motivated by money, they reluctantly accept that they need to earn a living, but it is not a big driver per se in most people I know.
When I say paid, I obviously also mean in intangible ways, such as having enough time to go to the toilet between classes. Or being insulated from being yelled at by parents and meekly saying sorry for telling off their particularly concentration-destroying kid.
How is it you'd institute a merit system? By the volume-of-students-per-hour, or what, exactly? The least complaining parents? Test scores that depend on parental involvement to be any good? How are disinterested or absentee parents or those parents who have it in for teachers accounted for in such a system?
Like capital class could tell the difference
The education children get in public schools is abysmal.
Good teachers that actually educate are exceedingly rare.
The education requirements for teachers is a sick joke. If you get a master's degree in underwater basket weaving, have a tolerance for noise, you join a teacher's union, work 9 months out of the year, get weeks of holidays, get virtue by association, a retirement, and are basically employed for life. Unfireable.
The reason teachers aren't paid more is because what they do isn't worth paying them more.
Some of the smarter ones move to private schools, but more of the good ones burn out and either leave, or stop giving a shit. Formalized regurgitation and pavlovian schedule training prepares kids for 9 to 5 bullshit. Barely. The rest is hit and miss on a garbled curriculum of mostly irrelevant filler and fluff.
You can't continue inflicting the bullshit teachers on the good ones, or on the kids. Unions have to end for certain classes of employment, but especially teaching. A vast majority of teachers need to be canned and kicked to the curb.
Kids deserve better, and that starts with honestly addressing the problem in the classroom.
Let's just fire all the teachers and see if the next batch of applicants do any better.
I'm a software engineer and I'll happily admit that I'm grossly overpaid for the contribution to society I provide.
I'm curious what you think teachers should be paid for trying to educate our future.
Actual good teachers are rare.
Simply being able to fire the bad ones would improve things immeasurably, but for some reason people put all teachers on pedestals. It's grotesque. Give PTAs hire/ fire powers or better yet, give the public a vote, weighted by testing scores or some performance metrics. Union ride or die is a garbage way to do things.
Treating teachers like a special caste of unimpeachable holy people is nonsense.
They are already good at driving good teachers out of the profession without a formal direct personnel role, giving them more power would make things worse.
Everything going on in education over the past 20 years has been to add that overarching, cross school performance metric to measure teachers against each other. The problem is:
1. How do you measure performance? Tests. What will any educator or student tell you? Ends up test performance doesn’t really correlate much to the quality of the teacher. What it does correlate well to is income of the student’s parents. You teach in a school that serves low-income families? You’re fired. Also tests don’t factor that some students are special needs. You have a lot of special needs students in your class one year? Sorry. You’re fired.
2. No one could seem to find the money to incentivize good performance of teachers. Just fire poor performers based on dubious criteria. Not exactly motivating.
3. Outside of test scores, which are a bad representation of teacher quality and easily manipulated, there isn’t a driving force to fire teachers.
I don't understand why everyone believes teachers are so undervalued. On average, I don't think teachers provide that much value.
My pet peeve is teachers teaching children maths. How can you teach something that you don't understand?
Your argument is like hiring only minimum wage workers to build rockets, then complain that on average they don't provide that much value and can't solve integrals in their heads.
If we want better teachers, the solution is not surprising: pay more.
I don't say that teachers don't provide much value because they're useless, but because they are usually very generalized and most kids don't seem be to that interested in learning.
Even if you fix problem #1, you haven't fixed the arguably bigger problem. I also don't think that significantly increasing the quality of teachers would change much for most kids.
Even when the author describes dividing the responsibilities of "leadership" among multiple people, they describe dividing both leadership and management responsibilities between those people.
This is an authoritarian model that makes the local despot the bottleneck for the team: Decisions end up being routed through the "leader", individual growth and development and team value is constrained within the scope of the "leader".
These constrained teams have limited potential, and so, because of the leader's limited focus and the departure of team members who feel constrained, the organization develops cracks and holes in responsibility and ability to act on opportunities.
Managers who subscribe to this authoritarian model promote authoritarians, capable technical contributors who want more control over the part of the system they and their peers have been working in.
The best managers don't have a hard time finding talent for open positions on their teams: They already have networks of former employees and peers that they can use. Indeed, the employees of the best managers recruit for their teams as soon as positions open up.
But virtually all of the job postings you'll see publicly (especially for replacement positions or incremental growth) are described solely in terms of the project and the expected technical skill requirements rather than or in addition to the attributes that make a team more than a collection of extra appendages for a "leader". E.g. team values and non-values, the existing roles and expertise of the other team members and how the open position will complement those, the team's norms around work-life-balance and communication, etc, etc.
I think that the average manager at a tech company is not a good manager, and that the larger the company the lower the average (smaller companies just fail with bad managers). But by far most open positions at any time are positions reporting to bad managers.
>The best managers don't have a hard time finding talent for open positions on their teams: They already have networks of former employees and peers that they can use. Indeed, the employees of the best managers recruit for their teams as soon as positions open up.
It's also fairly easy to tell from talking to a team if the manager is actually good or not. If many people don't accept offers to your team then you may want to do some soul searching.
Well, that's because there's a difference between autonomy (e.g. "I'm going to let you approach the task in your own way without micromanaging your efforts.") and responsibility (e.g. "Your job is to do X and Y and Z"). What most people want from their leaders is some degree of autonomy, not necessarily extra responsibilities (unless there's a promotion attached...)
It's easier to leave work at work that way.
1. Show up 2. Do work 3. Get paid 4. Leave
Your brain is freed to think of important things rather than work minutia. Of course, jobs like this often are mind-numbing if you're not the type to think with one half of the brain and do with the other and don't pay particularly great (in general).
The number of direct reports that function well as managed or led employees or any other organization hierarchy is about 5-7 but at most 10-12. This is a biological, neurological and psychological limit. It's also related to information theory and the UI-limits of the brain. It's related to chunking but also related to other limits of cognition and the finite number of hours in a day intersected with the minimal required cognitive focus/attention required to guide or manage an employee.
At the DevLead level one is a technician - diving into the complexities and trade offs and making a decision that will meet constraints that are the end point of a political process the devlead probably was not a part of
The politician you seem to be thinking of is someone (possibly the same someone) who is active in the political process that ends up deciding the constraints the technician works under - internal and external to the organisation
So this will be some of email threads or chats amoung some or all other devleads, it will be "being helpful" answering requests, assigning resources or time to help other politicians, or worse architecture diagrams, or it will be speaking at conferences etc etc etc. In the lingo it's becoming an "Authority". But one is only an "Authority" by being accepted by a "constituency" - maybe "all the devs in the company who think we ought to start using git" is your leadership position (ok maybe that's a decade old now but it's an easy example). Betray the constituency and lose the leadership position.
Neither are better than the other. Some political processes are better than others (in a democracy we perhaps appreciate this). Open decision making is likely to lead to long term better decisions - even if it is messy.
Linus Torvalds is an obvious leader example - using one political process (this is what I think - essentially using email as Thomas Paine used pamphlets)
But yeah - "managing" 5 people is reasonable - by the time you hit 12 it's a full time job listening to the moaning and they stop being a devlead. Only some can be a politican and a manager at same time.
This isn't an absolute conclusion, I don't think. I can speak for my own organization -
I lead 2 functionally-aligned teams, meaning they own internal product/systems for two different partner orgs. I also have a technical program management function which owns initiatives related to the above when they span more than a single team or are company-wide initiatives. I also have an engineering team with its own EM who roughly cover the scope of whatever my organization is doing - could be single-function, could be company-wide.
In my model, decisions from Principal/Director/Manager team members don't wait on me. They can if they want extra eyes or extra air cover, but otherwise, my job these days is less about setting technical direction and more about unblocking via executive alignment, budget, people, etc.
So far you might be thinking this is uninteresting and just means that my individual managers are the local despot, but I've done two things to (so far) eliminate that -
1. The TPM is responsible for getting us through the design stage, and they manage no one (or maybe other TPMs). This means they herd the cats and work with engineers and SMEs to get aligned on whatever design will make us functionally and technically successful.
2. This one is more on my qualities as a leader, but I also ask the managers on my team to err on the side of delegating more and letting their staff impress them, and being there as a resource when they need help. I think this is called servant leadership, but I don't read a lot of management stuff.
Team composition (experience and horsepower) plays a large role in whether this model can be successful and how hands on you'll have to be, but in trying to distill a decade+ of experience into a couple of paragraphs, this is generally how I like to run things. YMMV of course, I don't hold myself out as an authority, just offering a counterpoint.
> The best managers don't have a hard time finding talent for open positions on their teams: They already have networks of former employees and peers that they can use. Indeed, the employees of the best managers recruit for their teams as soon as positions open up.
If you adhere to the standard 1 year non-solicit, this isn't true right away. Hiring in my first year at a new company is one of the hardest parts of my job. But I'm probably a bit of a rule follower when it comes to the non-solicit - moreso than others, I've observed.
Having lots of peers is not enough anyway; first you need to have a budget to lure those peers.
Some people are really good at leading large teams, others are better suited to smaller teams. It really also depends on the makeup of the team and how much you can trust them to execute without you intervening.
I've lead teams of ~20 without issue as a first time lead. It went very well, both for me, and based on feedback, for the team as well.
However what worked well there was that I could trust my team to both execute their work and stay on top of things.
What I found helped was making sure that everyone on the team was made aware of everything going on (in summary of course) so that there was less need to act as the central knowledge store, which is quite common for a lot of leads to turn into.
This is key. Do you have people you have to take by the hand and guide through every step, or do you have competent, capable ones, able to think for themself (who will get annoyed, if treated like childs).
Reality is usually a mix and a valuable leadership skill is finding out early, who needs close supervision and who not. With very competent people, you can have a very flat hierachie.
But what I've repeatedly seen is that the moment you have conflict, if you don't have a manager per ~4-10 people, you're screwed unless there's a lot of organisational support pulverising the people management over extra people. Because suddenly a manager is spending half their time untangling some conflict and still need to have bandwidth to do all the other things they were doing.
As such I always look on teams with more direct reports than ~5-8 or so with deep suspicion. They may work well now, but they're often one crisis away from total meltdown. They're also often an indication of dysfunctional organisational leadership, who fails at promoting people to distribute responsibility.
You can lead far more people than you can manage.
I took on a lead role earlier this year and unless the 3 people on the team were all inexperienced or I had enough other things to work on, I don't think it would be enough. I do feel like my balance of attention is better with a couple more relatively autonomous people involved.
The Marines have this number set to 3.
Marines obviously have different scenarios they have to deal with than most organizations. But it's still an interesting idea.
https://www.inc.com/bill-murphy-jr/the-us-marine-corps-uses-...
For firefighters in Germany this is, more or less, defined to be 5.
A group (which is one of the pre-defined tactial units) consists of 9 people: The leader, three troops of two, one machinist, one "Melder" (kind of the guy/girl for "special tasks", helping with whatever is necessary).
The leader commands the troop leaders, machinist and "Melder".
This is expected and taught as the amount of people you can actively manage and take care of in stressful situations like deployments.
edit: for larger deployments you'll have platoon leaders which will command a number of group leaders. for even larger deployments there'll be another layer of command so each platoon leader will only have to lead a set number of group leaders, too.
I don't see how people are supposed to reasonably manage 10+ reports. At that point they are just a figurehead.
True! I'm leading 5 people at work and I find that enough most of the time, as I still have projects that I'm actively working on, too.
If there are more people to come, I'll ask to integrate some kind of "team lead" for some of them, before all the work I do is micromanage and respond to mails.
I think there are minimums too. It depends on whether your reports are experienced or not. If they're experienced, any number is fine. However, if they're inexperienced (i.e. interns, or fresh out of college), you really need to manage 2 or more. 1 is weirdly a bad number.
Managing a single inexperienced direct report is truly challenging and suboptimal, both for the manager and the report. Having no peers to calibrate themselves to, the report feels insecure and is either too eager to please or becomes overly dependent. It's hard for the manager to know how to manage too -- you want to avoid micromanaging but everything you do is going to carry undue emotional weight on the report because they have no other reference points because they have no peers. If I had to do it again, I'd rather have at least 2 direct reports or none at all. 1 is just not good.
https://randsinrepose.com/archives/seven-plus-or-minus-three...
I've been on a team where one manager was over 10+ people and I felt neglected and I felt like the people who didn't pull their weight got away with it and that work was pushed off on me.
All this is to say that I got turned down recently for a place I really want to work because I only manage 5 people. Hiring manager wasn't interested because of that single data point. The job was to take half a team from someone who was managing 12 people. Bullet dodged, maybe?
Sounds like they wanted to scale the team in the future and the current manager isn't able to do that. Rather than demoting them they're bringing on another manager to take half the team. Eventually the new manager would grow their team enough to be promoted and the current manager would report to them. Messy but may be the least messy approach that keeps the current manager from leaving.
> "Some management experts advocate strict limits to the number of people reporting to a common superior—generally five to seven. But if one has capable people who require but a few moments of his time during the day, there is no reason to set such arbitrary constraints. Some forty key people report frequently and directly to me. This enables me to keep up with what is going on and makes it possible for them to get fast action. The latter aspect is particularly important. Capable people will not work for long where they cannot get prompt decisions and actions from their superior."
Probably the author meant "supervise".
There is fundamental difference between leadership and management and it is pretty poor performance on author's part to write about leadership and management and not realize this.
It’s like the hacker = break into systems vs. tinkerer debate
It is fault of the author to give a misleading title, not mine as a reader to pick on it.
The average person? I'd say 0.
Can it be taught outside of a highly organized situation? I'd say no.
Can it be taught from a book? I'd say no.