Unlike Babies, the Best Managers Come with Instructions
torch.io
torch.io
1) It's manager-centric. Why would you give employees a guide to their manager when you could give them an "Employee README" that is focused on the new employee? When I join a company, I want to know what's expected of me so I can succeed. If the company delivers this in a form of a manager README then I'll still be missing the big picture goals of the company. Managers - Focus on your employees, not yourself.
2) Manager READMEs become an aspirational target. Describing yourself in text is challenging for the ego. I've seen too many manager READMEs that read like a picture-perfect version of what the manager aspires to be, rather than a description of how the person actually operates.
3) The self-promotion aspect. When managers post their README all over internal company chat, public Twitter, their LinkedIn profiles, Medium think pieces, and the company blog (like this article), the document becomes a brand-building exercise. This puts even more pressure on managers to only write down their most perfect, idealized vision for what they want to be. Or they spin their flaws as some sort of superpower with creative wording. I once read a public manager README that was nothing like that person's actual management style, but it read as if that person was a textbook-perfect manager.
My recommendation: Skip the manager README fad and focus on your employees. Write an "Employee README" onboarding guide for each of your employees, that you customize with individual expectations and information for each employee before they join the company. Keep it in a shared document that you can update over time to include an up-to-date job description, list of responsibilities, and key expectations for that employee. Don't try to make management all about the manager's personal brand.
Why not do both? "Hey folks, here's my README, I'd encourage you to write something similar and share it with everybody."
When it comes to team building, I think it's much more valuable to say "this is how we operate" instead of having everyone write down "this is how I operate". The former is a mutual guidebook for everyone to work together, while the latter feels like an edict that everyone else needs to work around you.
Check for any overtly abusive behavior that requires correction. Usually rare, but worth checking.
Provide basic coaching and suggestions for working together in a healthy manner.
And most importantly: Let them know that they're expected to figure out how to operate together on their own like professionals. Managers can't be mediators for every impedance mismatch between employees.
I'm certainly not writing anything about my personal life because that's personal and, quite frankly, none of your business. I have a tragic backstory that I don't want to share with coworkers. It's my job to overcome my past and be effective and productive at work, not everyone else's job to accommodate any unexpected or unprofessional behavior, etc.
There's a whole host of instruments that more mature companies use to uncover the different working styles and professional psychologies of employees. E.g. knowing that one colleague wants just the facts and isn't interested in the back-story whereas another wants lots of context can be very useful and ultimately help one become more effective professionally.
I'm thought experimenting on also having a README written by direct reports that the boss doesn't get to see (so people could be more honest). I'm guessing it would quickly leak to the boss anyway, or someone trying to score points with the boss would just whitewash it. Perhaps only the person who wrote something could edit it, so you get lots of perspectives?
Would be an interesting experiment tho.
tl;dr, Manager README's are more for the manager than the direct report. It's a great exercise in introspection if kept private and not used as something codified for employees to follow.
While it was unintentional in my original post, "report" seems to work better than "direct" here as IMO my statement holds for a manager's indirect reports as well and thus "report" covers both direct and indirect. But now I'm just amusing myself with word games :)
In companies where management has to explicitly direct the activity of their employees, something has gone seriously wrong. Either the company is not working on something useful to the market and so management is trying to motivate employees to do something pointless anyway; or the company's vision & strategy hasn't been successfully communicated by the CEO so that everybody knows where they fit in; or a manager on a power-trip was promoted into the position; or my values don't align with the company's and I'd be better off elsewhere. In any of those cases the answer is to leave (or if it's my company, fix the problems before hiring more people) and find a company that there's actually a reason to work for. You have one career and there are ~20 million businesses in America to work for (and you can always start your own if no suitable one exists); make it count.
When values are in alignment, the power imbalance shouldn't matter much. But when there are disagreements within the team, the manager is the one who defines the rules of conflict resolution, or in extreme cases decides the result.
When values are out of alignment vertically the report is left with 3 options: adapt, escalate further, or leave.
The person you escalate to has more power, otherwise it would be pointless to escalate to them. Ignoring that fact seems silly.
It's usually written as "direct report" to differentiate from your boss's boss, who you indirectly report to.
From the OED:
noun: report; plural noun: reports
4.an employee who is supervised by another employee.
I'm a manager, I report to someone (my manager, who also reports to someone) and likewise I've people who report to me (my reports).
It doesn't change that we're all people, with our own quirks, personalities and emotions.
In Silicon Valley, people managers can easily be a waste of time and money, and can cause unnecessary drama on your team. Think about the kind of so-called developer who would rather manage people than write code.
We still need founders, executives, directors, advisors, etc. don't get me wrong. But has anyone ever come across an engineering manager worth their salt? Why did we start deploying people managers en masse to engineering teams? I miss the holy trinity setup: Dev, designer, PM. It worked well IMO.
Our industry is shooting itself in the foot with Engineering Managers. The ones I've come across are self-obsessed, as evidenced in these cringeworthy READMEs and this article that was written.
What is the difference between a manager and an executive? A manager is an executive of a small part of a large org.
BTW no-one is deploying managers out of thin air. At least not in startups. What I can often see instead is that someone in the higher-ups needs OKRs, and he also wants to know your performance (and also others' on the team) so that he can decide who deserves a salary raise, and he would like to read a summary report about the highest priority project's progress during last week. Plus he would like someone to represent the team's technical standpoint next week when meeting a prospective client, plus he needs someone who goes around the team and asks what bothers them and what would the team like to see changed in their work environment, plus he also needs someone who hires new team members, plus he would also like to organize the mentoring of the junior members and would like to see someone be involved in the product roadmap discussions from the engineering side to make strategic decisions, plus he wants to build a new team and he wants to understand who would be the best candidate to participate in it, etc. The list goes on and on.
So who will do all these things? Most devs absolutely hate doing these things because it's not technical enough for their fastidious taste. Designers and PMs OTOH have no technical knowledge to make strategic engineering decisions on higher level than the individual level (and not even on the individual level many times).
When almost everyone bullshits, there is no danger for the bullshitter.
So quite a number of people have a big interest in keeping the talents out of the game, and to have as many layers of bureaucracy as possible as this makes it more difficult it is to point the finger at any one person.
The same happends in school classrooms. If you are quite a lot better than the other pupils in class, then you better keep your head down. Don't think you know better how to tackle the work ahead.
I think schools educate us into accepting only bad management.
- Whos going to push back and sales and marketing for harassing you directly or taking direct blame for failures?
- Who will represent the engineering team in leadship meetings to make sure policies/decisions are made with their needs in mind?
- Who's going to fight for your raises and promotions?
- Who's going to make sure that there is budget for what engineering needs to do and keeps a proper record to keep finance happy?
Please thing about it... any company that grows to 30-50 people (even startups) need to have management and processes that eventually lead to (even unintentional) politics. The other departments will have leaders/managers defending/supporting them... its foolish to not understand or respect competent leaders.
Managements role, technical leadership role and politician role are different roles. The required skill sets and experience are different. The outcome of grabbing a wrong role is usually poor. Not confusing politician, technical leader and manager roles is probably a good idea to improve outcomes. As checking for required experience, to take the role.
On "strategic long term decisions", I think this needs clarification. There are many areas in which such decisions can be made.
To give an example, think of a flying airplane. Would you give a power to make "strategic long term decisions" regarding the flight to someone who bothered to go to flight school, ground school and had been certified as a pilot? Or to someone who hadn't done it, but is an excellent politician, people's manager and could mediate conflict well?
To give another example, if a conflict between passengers and crew had happened onboard of a flying airplane. Would you prefer a pilot (say, with no political or mediation skills) to stop flying the airplane and engage into the conflict? Or would you prefer a professional people's manager and mediator to step in?
> To give an example, think of a flying airplane. Would you give a power to make "strategic long term decisions" regarding the flight to someone who bothered to go to flight school, ground school and had been certified as a pilot? Or to someone who hadn't done it, but is an excellent politician, people's manager and could mediate conflict well?
I don't think this analogy works. The manager decides which routes to fly and ultimately who flies which ones. Presumably some routes are preferable, and there are competing priorities between routes. So the manager picks the routes to fly, addressing external conflict, and then alloctes people to routes, handling any internal conflict.
The manager isn't on the plane. The rest of your comment doesn't make a whole lot of sense.
I think your analogy works. And it seems that we agree, it'd be stupid for a manager without pilot's training and expert knowledge to make or impact "strategic long term decisions" that impact the flight in a flying airplane. And we likely agree that interference of a manager (who is used to exercise power) will likely make negative impact on rational expert decisions of a pilot.
Regarding "strategic long term decisions" outside the airplane - we can just look at the recent example of Boeing debacle with 737 MAX. Again, a good question would be - would the situation be better, if the company board had people that had bothered to go to flight school, ground school and had been certified as pilots?
No, I asked for an example of a management role that was apolitical. There are certainly responsibilities that managers have that aren't political, but someone claimed that people, political, and technical management were different. So I'm intrigued by the idea of a people manager who doesn't handle any political problems. My point is that by virtue of being in charge of people, political arguments become unavoidable.
As for your Boeing example, I fail to see what this has to do with managers having to address political issues in the company. It seems like you're having a conversation about how managers should have been ICs at one point which while an interesting discussion, is completely unrelated to anything I've said.
Can you give an example of unavoidable political problem that people manager unavoidably has to handle?
- Two of your reports want to work on the same project, perhaps because it will look good. There's only enough work for one of them
- The team fails to meet an objective. This needs to be addressed, somehow
> Political problems in that case would be implicitly handled by company culture
Someone(s) still need to make the decisions that align with the culture. Those people are making political decisions, they just have values that help guide the decisions.
I agree, this is a political problem. But it doesn't seem to be unavoidable. Perhaps, increase rewards for long-term gains for the company and remove rewards for short-term personal gains? It just wouldn't happen then?
Team failing to meet the objective. I don't see how this is a political problem.
You've just described a way of dealing with the issue when it arises: paying the person working on the less sexy project more. This may sometimes be a tool the manager can use, but it isn't always an option. So presume you don't have this tool, or the value of the project is larger than the value of the compensation you can provide.
> It just wouldn't happen then?
You're presuming that the company can perfectly reward people for work, and that every employee believes that they'll be rewarded optimally for work in the company's best interest. I've yet to encounter a company that can accurately reward employees for their performance, and because of that, I've yet to find a company where the employees believe that their performance will be perfectly fairly rated. If you can fix this problem, I'll buy your product.
You look at my examples as negatives... its the wrong way to look at it... thoes other managers of other departments? Are the bad people? No... but they have a job and responsibility to represent their departments and do from their point of view whats best for their team and the company.
I feel like there is a segment of workers in engineering has this weird self damaging hatred for leaders/managers other departments do not have... then thoes same people get pissed when they dont have a leader and their department is glossed over because no one represents them when/where some of the most important decisions are made.
- Who has the unchecked authority to blame others for their mistakes?
- Who will take direct credit for the work of others?
- Who's going to write a bad performance review of someone they don't like that much?
It goes both ways!
No one, its done by committee and/or senior leadership, the team/department with no seat at the table is always at a disadvantage.
>Who will take direct credit for the work of others?
A managers job is to distribute credit and shield blame/failure. This may sound generic but its true... a managers true strength/power/greatness or what ever other BS comes from the strength of their reports. Its extremely short sited to not pass on credit and grow those under you.
>Who's going to write a bad performance review of someone they don't like that much?
A good manager tries to be very objective and doesn't bring personal opinion into reviewing performance. I have had people who I did not like personally who i brought though multiple jobs... good managers respect effective people not people they can get a drink with after work.
The best place to observe the need for management is in teams like, 20 relatively inexperienced engineers, all reporting to a single manager. I have never seen a team like this that is operating at 100% of the engineers’ capabilities. People end up wasting time on disagreements, building things that end up not needed, jumping from emergency to emergency, not using best practices, all sorts of things.
If you find a good manager and split those people into two teams of 10 engineers, that change will make that team much more productive.
The problem is they’ve turned into something that justifies bad behavior more often than not. So many of them have this undercurrent of “I’m an asshole - don’t take it personally!” which is just totally not OK. The solution isn’t to proudly broadcast and justify that you’re an asshole and expect others to adapt — the solution is to not be an asshole.
That being said, things like “I tend to work late so if I email you on the weekend, I don’t expect you to reply until Monday” are totally reasonable and I think, on the whole, a good thing.
I've used OKRs, Kanban/agile and many other tools for me at an individual level and loved them. I've had the exact same tools at companies become a waste of time, a point of contention, and only be a "value add" for management.
An Engineering Manger/Director/VP/CTO's main job is 2 things.
- To be the over arching intermediate between engineering and other business units (note: this does not mean walling off engineering, engineers SHOULD understand and work with business units when it makes sense).
- To support, mentor, shield, and basically do what ever is necessary for their team/department/org to be as successful as possible.
This sounds VERY self centered and I honestly believe product and engineering managers need to be to most selfless people possible to actually be effective. I would call it backwards even... a leader/manager needs to tailor himself and his approach to both his people and his cross-department equals in order to be the most successful.
Then you're incredibly fortunate. I've had to deal with management issues at multiple companies in a shorter span of time.
Not to mention people are just horrifying bad at seeing themselves as others see them. I think of the scene in A Christmas Story when the kid expects to get a glowing review of his paper but instead got a D grade. It actually reminded me of a family member of mine who got called into his boss's office. He was bragging that he was going to get a bonus or promotion and instead he got a demotion due to poor performance and bad work ethic.
I've had five software jobs and five unskilled jobs.
The reason I thought it would be helpful is that I am not a high feedback person. I seldom praise (need to work on that), I am not expressive, I've been told I 'play my cards close to the vest.' by someone who directly reported to me. That really got me thinking that maybe the circumstances where I am not getting what I want is because my staff really cannot figure out what I'm looking for in general. So I wrote up a 'read me', before I knew there was this trend.
I included self-deprecating statements about things like assigning a task to multiple people because I forgot it had already been assigned 'Don't take it personally, just let me know and I'll fix it.' And, if they need me to attend a meeting, make sure it's on my calendar, or, no chance I'll remember.
In the end, it's only going to be as helpful as it is 'correct' and clearly described. That depends on each manager's level of self-awareness, combined with their sense of self confidence. You'd hope managers would have both qualities, but if not, this can really be a red herring, or even be used as a dark pattern.
The reason I would be afraid to standardize manager READMEs is specifically that I do not trust the average manager to publicly acknowledge and privately work on flaws like that. For managers who do that, I’m not that concerned about the exercise because it’s more likely to produce something useful for the employees. For the other type of manager, I’d be afraid of accidentally creating the expectation that the employee needs to adopt the manager’s style, which I would privately recommend to any employee, but which I would vehemently oppose as anything resembling an official expectation.
unfortunately, this is very unlikely to be true. and increasingly less so as the team size increases.
one confounding factor here is the heavy use of religion to guide technical decisions. the decisions we make now about the direction of the product and the attributes of the codebase are largely subjective. there are arguments on all sides of a position.
a good team, properly enculturated, understands the process. when the issue is important enough to justify digging in and when its just a personal bias. a good team helps all of its members stay on the path and not end up in giant ratholes. a good team uses its experience to inform future decisions rather than create artificial strictures. a good team has enough time and maturity to interact constructively with the rest of the business and the customers.
I shouldn't say 'good' team, I should say 'perfect'
good management greases all of these points of potential friction. but I think at the end of the day management is there to address the last point. its natural and expected for engineers to keep a narrow focus on the technical aspects of the project itself. someone's got to make sure its evolving in a direction that's meaningful in an external context.
The short version is that it worked amazingly well for a little while, with a small team of people that worked well together. However as the team scaled certain personality traits emerged in everybody. What basically happened is certain people became "the manager" in practice because everyone just deferred to this person for decisions and such. Various reasons for that, good and bad:
1. This person was technically very capable.
2. This person had a strong personality that was disagreeable, and people didn't want to cross him.
3. If you were wrong in any way, this person would publicly broadcast your failure, which made people afraid to ever challenge him.
4. Many people aren't really comfortable not having a boss. They will naturally look to someone to lead things, and usually the person they select is not the best one for the job (or even close), it's just the person with the strongest personality. I don't mean to get political here, but I can't help but notice how similar this is to President Trump.
So basically my conclusion is: we can't have nice things because human nature. More specifically, manager-less teams would great for small groups that work well together. It doesn't scale, and doesn't work well for everybody. Some people need management.
Generally, when someone suffixes their name in a non-scholarly context with "PhD" that's usually the first clue that it's going to be pretentious and/or self absorbed.
I also cannot understand the obsession with 30 minutes of 1:1 every week.
I consider that an extreme version of micromanagement. Give us developers the big picture so we know what we need to work on and can see what's important in which way, and we will find out how it's done.
If you don't trust developers to be able to communicate and plan complex jobs without 30 minutes of 1:1, how can you write "excellent communication skills" in the job posting?
Why is a meeting with a single developer so much better than a meeting with the full group? In group meetings you get more view points and more easily find out if somebody are talking past each other.
What is it that needs 30 minutes of personal time every week and that cannot be communicated at the meeting with the other developers in the group?
Unless your surname is .exe I don't think you need a readme file. You could do the human thing though and discuss expectations with your team though...