Hacker News new | comments | show | ask | jobs | submit login
Jason Fried: Why I Run a Flat Company (inc.com)
435 points by duck 2476 days ago | hide | past | web | favorite | 179 comments

Fog Creek and 37Signals are probably more alike than either of us would care to admit (ha!) and I could see Joel writing a very similar article a few years ago when we had 20some people at the company. But what works there, or at FC, is not a one size fits all answer.

Having managers when there are 10 people are at your company makes no sense. The hierarchy starts out flat, you add a few more people and you're at 20, and the idea of making someone a manager seems like a waste and something a 'BigCo' would do. "We need people who get stuff done, not people who sit around doing nothing but managing", you think. Then you get to 50 people and everything breaks down. You wonder why people are frustrated they can't get things done, while other people are doing things that embarrass your company or compete with other things you are doing. And you realize your company isn't a special little gem that is wholly unlike every other company in existence. You need management.

Just make sure you give the devs a professional ladder and compensation structure that doesn't involve moving to management, because managing isn't something everyone is good at or even wants to do. And make sure that management knows their job is a support role to the people at the company who are making things happen, not the other way around.

"But what works there, or at FC, is not a one size fits all answer."

Absolutely. I'm simply sharing our experiences with how we run things - that's what my Inc. articles are about. I wouldn't expect anyone else to do what we do because we do it. Find your own comfort zone. There are plenty of ways to do plenty of things.

I really like your approach, especially making everyone manager for a week. I think the only problem I see with it is that if people want to find another job, their 37signals title makes it seem that they essentially "went nowhere". So even if you are paying them a lot more then when they started and they are the best customer service person ever, their resume still says: "Jane Doe - Customer Service Rep 2005-2011" which looks unimpressive elsewhere.

We solved this in part by using "cocktail titles". So anyone who works here can call themselves senior. So we have "senior designer", "senior programmer". At least that has some ring of fancy.

Sounds like you are following Ricardo Semler philosophy(author of The Maverick). Are you? [http://mitworld.mit.edu/video/308]

If you liked Maverick, you should definitely check out Seven Day Weekend by Semler. I found it much more readable and relevant. It's crazy that it's out of print in the era of the Kindle


Yes, huge fan of Semler. Big inspiration.

My company handles this by giving you whatever title you want. When I started, my boss sent me an email: "Give me your full name and title, as you want it to appear on your business cards."

It looks unimpressive... if you don't know 37s, Honestly I can't imagine that even their CSRs have _that_ hard of a time finding more work.

If they worked in such a high profile company like 37signals and did good work, all they need is ask for a few references and they can easily land a good position (if not making full use of their experience to run their own company).

Although they will have had the experience of being a manager (albeit for 1/4 of the time), and having been in both the manager and employee role during the same time period, they probably end up learning more about being a good manager than most people who have the title "Manager". With 4-5 people, I think it makes a lot of sense.


I'd imagine that this isn't nearly as much of a problem for an extremely famous/well regarded company that's also very small.

I for one appreciate the insight into how you run things. I especially liked the point you made about encouraging mastery (horizontal) instead of managerial (vertical) promotion. So, thank you for sharing.

I also appreciate the perspective. Next time I'm asked if I've managed people I plan to say 'I manage projects, I collaborate with people.'

How did the employees feel about the various bits of organisational experimentation carried out? Did any balk at the constant change?

+1 "managing isn't something everyone is good at." How many bad developers get moved to management to keep them from causing trouble, and then cause even more?

Isn't that basically the Peter Principle?

The Peter Principle is less cynical, but probably more realistic. It's that you get promoted for doing a good job, but eventually you get promoted into a job you're not good at -- at which point you stop being promoted and you stay in a role that you're not suited for.

Note, there's nothing malicious that happens here. No one is trying to promote a bad developer. People are largely doing what they think makes sense -- promoting top performers.

The usual money quote for the Peter Principle is, "in a hierarchy, every employee tends to rise to his level of incompetence" -- but I like your summation so much better.

It is the Dilbert Principle http://en.wikipedia.org/wiki/The_Dilbert_principle

"[C]ompanies tend to systematically promote their least-competent employees to management [...] in order to limit the amount of damage they are capable of doing."

Ah, whoops - that's the one, you're right.

I like to call it the Alicia Principle http://thedailywtf.com/Articles/The-Speed-of-Code.aspx

This is really true, and something we are experiencing now in the company I am working for. Traditionally we've been a company with ~12 employees, small project teams of 1-4 people. Now the company has expanded to over 40 employees over two offices, with team sizes becoming larger than 10 people.

We have experienced the exact difficulties you have described, and have been slowly adjusting to create just enough structure to organize and motivate people. I would add that an additional area that becomes more important as a company grows is communication. Once the company gets past the size where the CEO/Director is getting around and talking with each employee on a weekly basis, and physical walls start to grow between teams, communication becomes much more difficult in both a vertical and horizontal sense (a side effect of added hierarchical structure i suppose?).

The article mentioned rotating managerial duties, and alternative methods of adding structure. Was your comment taking that into account?

Rotating roles is great for a small company when the amount of managerial activity is very small. But creative work and manager work don't mix well. Being a manager requires responding to things that are happening at that moment (interruptive), which essentially destroys 'the zone' required for getting things done as a creative (developer, designer, etc). The other problem is that most people aren't cut out to be good managers, and learning to be a good manager is a skill - the more you practice the better you get. You start to see patterns and figure out how to deal with them. But rotating roles could help when you are first starting out to figure out who has the manager chops, and who sees it as a burden. Ideally the less time your company spends managing, the more time you can spend making cool stuff, so if the workload is small - go for it.

"Even as we've grown, we've remained a lean organization. We do not have room for people who don't do the actual work."

That is a priceless comment. It exposes Jason's huge blind spot. Worse, it is on this undefended flank that great future pain may be inflicted. I look forward to the post-learning article.

The underlying premise/assumption is that a 'manager' is not only not doing 'the' work, but they aren't really doing 'any' work. Its a very common meme in engineers, "The company makes money on the code I write, it makes no money at all on this guy telling me what do, it just costs them money."

Let's reason about this using a fairly simple analogy. We will start by positing that we are all rats in a maze. Our maze is, unfortunately, filled with rat dung. We further stipulate that walking on dung would kill us so the only way we can move through the maze is by shoveling the dung in front of us, to the pile behind us and then moving into the space we opened up. All the shoveling burns up calories, if we don't eat we will eventually starve to death. Finally, we add that there is a cheese somewhere in the maze, and once any rat makes its way to the cheese, everyone gets to eat of the cheese. That resets the rat’s hunger level, after the the cheese is located the maze resets around all the rats and process begins again.

Now in our analogy our engineers are the rats. And writing code is shoveling rat dung. And the cheese is a monetizable opportunity. Eating the cheese is collecting money from the opportunity.

In a small company, having everyone shovel as fast as they can, is a great strategy for finding the cheese(s). Some mazes have more than one cheese in them, sadly some mazes have no cheese in them. A manager, whether its the founder/CEO, or someone in that role, is given the opportunity to stand above the maze and see if there is a cheese nearby or in the distance, by seeing both the maze and where the cheese is relative to where in the maze rats are, they can direct rats that have the best chance of getting to the cheese quickly in the direction they should turn, otherwise each rat would be following his/her internal idea of the best way to find a cheese in a maze like ‘always follow the left wall’ or ‘alternate left and right turns’ or ‘leave marks in the dung piles of parts of the maze you have already passed through so that you can pick new passages the next time.’

So the leadership role of management in any technology company, is measured by their ability to get teams to the cheese while shoveling the least amount of rat dung. Good leadership will understand that there are many cheeses (and flavors of cheese, some more nourishing than others) and be able to evaluate the choice of going further for a very nourishing cheese vs going out of the way to munch a nearby, but less satisfying, cheese.

So back to the comment tail … “who don't do the actual work.” briefly.

It is pretty easy for an engineer to recognize a problem in one of their colleagues, even though their colleague is ‘doing’ a “lot” of work, that work is inefficient and thus ‘poor’. Someone checking in version after version of a subroutine, trying to get it correct, when that subroutine is doing something that is provided by the underlying operating system. Lots of ‘work’, lots of ‘check ins’, but someone who had a bit more breadth might have done in a couple of hours what this loser is taking a week to do. As an engineer, one can easily appreciate that this person is taking up an employee spot that could be put to more efficient use by a better quality engineer.

And yet it may be hard for that same engineer to understand that a manager is helping him, and his colleagues, be more efficient by working excellently on a component that will get them to a good cheese, versus working excellently on a component or a technology that does not proportionally have the business value they need to pay their own salaries.

A real world example was a shopping cart company that had, at one time, all of its engineers working on a universal language independent component for presenting product descriptions in over 100 languages and nobody on the team was working on making the shopping cart code play nice with various payment services. Which is the more nourishing cheese? English only and you accept any kind of payment, or any language but you have to have one type to payment card from one vendor ? The engineers were all writing excellent code, using all the latest best practices and the language support module they came up with was best in class, but product was a shopping cart and the “high order bit” for a shopping cart implementation is “can it take money from customers and put it in the bank?”

So when an engineer makes a comment like Jason’s about valuing ‘doing’ over ‘directing’, it can sound like the oarsmen in a galley complaining that he should be accorded higher status than the navigator since without him the boat wouldn’t go anywhere. But the reality is that without the navigator the boat wouldn’t arrive anywhere. Considered in the larger context, the navigator’s role is both more stressful and more important to the overall success of the trip than the oarsmen.

What Jason’s comment misses, and it sounds like a blind spot, is the understanding that you cannot successfully navigate and row at the same time.

Also, sometimes the cheese is Gouda. When the cheese is Gouda, it is the cheese of customer service, and the job of the person standing above the maze it to identify if there is Gouda and where it is. But sometimes the cheese is Emmenthaler, and when the cheese is Emmenthaler, then there is a rule† that the Gouda manager and the Emmenthaler manager must share the cheese or else by finding maze paths that don't connect directly to each other immediately but eventually do connect get more Gouda than Emmenthaler or vice versa, in which case they take all the cheese, and you can see now that this might possibly be the most least helpful analogy ever constructed to make a point on a message board ever.

I promise it is all love with me and this comment.

Without employing any additional analogies, I believe I can respond to your underlying point by suggesting that in a well-managed organization, one President or CEO can structure roles such that everyone is clear about the business value they provide and can manage their effort accordingly. This requires that you be careful about who you hire, what roles you create, and how you scale the company.

It does not however require that you not scale your company past 20 people.

That strategy, however, will not work in the (bad) kind of Silicon Valley startup that grows by buying Pasteurized Processed Product Managers. Your CS reps need to know what "service" means and how your specific customers perceive and value it. That can be hard to do. Thus, some companies probably do need management structures.

Though only in the US edition that Rio Grande publishes, not the original German game

Excellent. This is the key bit for me:

"...in a well-managed organization, one President or CEO can structure roles such that everyone is clear about the business value they provide and can manage their effort accordingly."

And once achieving this point, you neither change the people in the organization, nor do you change the direction of the overall effort. Life will be grand.

So 100% agreement on the goal of what every CEO/Founder is trying to achieve which is "Structuring the roles such that everyone is clear about the business value they provide and can manage their effort accordingly," Keeping the organization as close as possible to that goal in the presence of change, the limiting factor of using people that happen to be available in the job market at any given instant in time, and with the added constraint of operating within a viable business model. Well that separates the good ones from the bad ones does it not?

It has been my experience that people who are not able to evaluate the relative contributions of others who are in dissimilar roles make poor choices.

When I read a comment like 'we value doing over not doing' in the context of leadership, I wonder under what set of constraints is leading a group or team not a form of 'doing'? I can't come up with a case where someone who was, as part or all of their role, leading, felt that the 'leading' task wasn't a form of 'doing.' I have also not experienced anyone who, was in a leadership or management role felt that 'leading' (or in our case 'managing') they were doing was 'less hard' or 'less useful' than technical work (designing, coding, debugging, Etc.)

However, I've encountered people whose experience is dominated by technical work firmly believing that managing or leading was both easier and had less impact than the technical work they were doing. And I've known managers who have been managers for most (or all) of their career and felt that is was harder and more relevant than what the technical people did who worked for them.

So, to my way of thinking, if someone does value 'doing' over 'not doing' (which is good) and they understand the different values between what managers do and what engineers do, then the question of 'vertical movers' vs 'horizontal movers' is not a cultural thing its a 'what do we need right now for the business' kind of thing.

It's like reading an article that says "We had to fire Bob today because he came to me to say ask if he could spend more time on gcc, you see our organization is all linux kernel hackers and we like it that way." And yet that pre-disposition to kernel hacking completely misses the benefits that can be achieved if you your compilation environment improves. Anyway, enough analogies, I can tell they obscure the point too much.

He doesn't necessarily miss the value of managers, in fact he says that he has managers, just not one person who's sole role is a manager, they rotate every week.

When you mention technical workers being suspicious of managers and managers believing that there job is more difficult than technical work, you're talking about the exact situation that Jason mentions trying to avoid. One way of being less suspicious of the value of someone is to do the work that they do. That's what he's talking about.

"And once achieving this point, you neither change the people in the organization, nor do you change the direction of the overall effort."

Good luck with that one.

I think you bring up an excellent point, and I really like the cheese analogy. The only problem with this analogy is that it presumes that managers can effectively see exactly where the cheese is.

In your real world example, I would bet those engineers had some management team that decided localization into 100 languages was more important than interfacing with payment systems. That seems like a real world example of a management team that couldn't see the cheese.

Speaking from an engineer and team lead point of view, yes management is important, but not all management is created equally. For many large companies, the Dilbert principle is in full effect - non-technical people are promoted to management where they can supposedly do less damage. What does more damage to a company in the long term? One bad engineer checking in code that creates rework, or a manager that forces an entire team of engineers to head down the wrong path in the maze towards cheese that doesn't exist?

Also, I just wanted to mention that rowing doesn't mean you can't navigate as well. Bright engineers have a good picture of the industry they are in, and market direction. Some are far better at it than their managers. Just because you write code all day doesn't mean you don't know how to make a business decision, which is what I read when you say we can't row and navigate at the same time.

"The only problem with this analogy is that it presumes that managers can effectively see exactly where the cheese is."

Yes, its the difference between good managers and not good managers. And yes the downside of a bad manager is that a bunch of people have a poor experience rather and the rest of the team having to cover for a poor developer. So picking managers is hard, must be done carefully, and must be constantly evaluated.

"non-technical people are promoted to management where they can supposedly do less damage."

See the comment about the role of the CEO (or more senior manager) if someone isn't the right person for any job in your company, then you have to let them go. If you can't do that you're not doing the management thing correctly.

What? I'm lost.

"What Jason’s comment misses, and it sounds like a blind spot, is the understanding that you cannot successfully navigate and row at the same time."

I strongly disagree, and I think the metaphor leads one astray.

In a startup there is not a navigator and then an army of oarsmen. Instead there's a team with diverse skills attempting to understand and solve a complex problem. A manager can only direct rats to the cheese if they themself fully understand mazes, cheese and shoveling shit. But if each rat has diverse and individual ways to contribute to getting closer to the cheese, the job of a single manager providing vision and directing becomes geometrically harder as each new skill/role is added. Imagine each rat speaks one of a babel of different languages, and we start seeing how hard the job is for our imaginary leader.

It's incorrect to assume that the rats shoveling the shit don't have their own vantage on the cheese, or that the oarsmen can't see the stars and where the ship is headed on their own.

This is the difference between an industrial structure and a a research structure.

Top down command structures are extremely effective when everything is already understood. It's what you want when you're solving a known problem with known tools, like building commodity products on a production line.

But top down networks face enormous challenges when trying to achieve only partially understood goals within complex changing environments. Integrating feedback as it travels back up the tree is very difficult to do effectively. Top down command networks are at their peak efficiency when feedback doesn't need to go up very high. Otherwise, the top of the structure becomes overwhelmed.

This divide is almost entirely why large technology organizations end up innovating primary by acquisition rather than internal discovery (example: MS). They have to buy both solutions to new market opportunities as well as the people who understand and built those solutions, because their internal structure is very poor at learning and adaptation.

Collaborative problem solving is different. It relies on frequent and open ended horizontal communication. It often appears directionless. But what it facilitates is overlapping shared understandings far more detailed than could ever be centralized in one person. It allows points of command to emerge when necessary, and disappear when not. The ease with which a small team does this organic communication and discovery is one of the huge advantages of small companies. It's not one you should shed lightly.

The only way to do this in large companies is to maintain a very horizontal structure where individual business units have a great deal of autonomy in accomplishing their goals, and where employees can fluidly move between teams to spread knowledge.

There's another option for the big company: create an internal market structure. But AFAIK no one has a repeatable template for quite how to do that (and apparently Google's attempt at it has somewhat failed).

In any case, returning to flogging your metaphor:

The smart navigator will listen to the oarsman: perhaps they've noticed something I haven't? And the very smart navigator will teach all who are interested about his skills, so that the whole boat becomes more valuable than it's competitors.

And notice that no where in here have I mentioned any terminology about prestige, respect, or "actual work". This stuff is cancerous, and whenever you see it, get rid of it immediately.

Your comment presupposes that there aren't any proven models of self-management.

This is the most ridiculous post I've ever read on Hacker News.

I've started companies that grew from 1 to 50 people. I find that business advice from 37signals is often quite naïve and I hope people treat it as a data point, not as a set of guidelines.

Managing a company below 25 people is relatively easy. You can still talk to most people every day, you can gather them all in one room, information flow is unrestricted. But staying under 25 people means for most companies that you are stifling growth.

Once you get above the magical 25 people threshold, you'll find that it is simply impossible to manage the company effectively using a completely flat structure. Also, you'll discover a lot of new problems you never suspected existed before: you'll need internal PR, for instance, as people in one part of the company won't know what people in other parts are doing. There will be myths to dispel, growing animosities, lack of direction. And there is simply no way you can keep your pretty flat structure with 45 people.

I know that 37signals' advice resonates with people. They are the cool kids. But keep in mind they are an exceptional company, in every sense of the word.

It's distressing how often, especially among technical people, the equality

    flat == good
is taken as a given. Hackers have a healthy skepticism for arbitrary authority, but we often erroneously extend that attitude to hierarchical structures in general. In the limit of large numbers of people, hierarchies are the only structures that scale; moreover, they can do so over many orders of magnitude, from small businesses with 10^2 people up to huge organizations with 10^4, 10^5, or even 10^6 people (such as GE, IBM, Walmart, or the U.S. military).

That's why small organizations are preferable to most hackers. I don't see how that changes flat == good. GE, IBM, Walmart, or the U.S. military would be places I'd never work, given any choice. Scaling is not necessarily good, it usually means dumbing down to the average.

Fair enough. But in that case most hackers should also accept that they won't make a boatload of money. If you refuse to grow your organization, don't expect to be a millionaire, especially if you follow the philosophy of running a cash-positive business in the long term, without selling it.

You can't have your cake and eat it, too.

This really resonated for me because we have almost exactly the same company dynamics at play (we're of roughly the same size).

I'm not sure we have similar answers, but one response to the problem of not being able to afford people who don't do real work is to make sure everyone is doing real work. We're primarily a services firm, and everybody in that org, including Dave, our President, is billable. It's something I tell people in interviews, and that I'm sort of proud to be able to say; everyone's grounded in the actual work that our actual people are actually doing for actual clients.

In that spirit, one way to address this problem might be to have team leads instead of managers.

I feel like Joel wrote about this a few years ago too, and while I'm probably wrong about this, off the top of my head it feels like their answer to this is that when they have too many senior people, they think about new products. Isn't the highest level on their comp ladder (not a fan of that thing) reserved for people who can run products?

Would love to hear more about what people in the 20-40 employee bracket are doing here.

20-40 employee bracket here! All of our employees are billable (sort of). It's hard to explain our structure but needless to say our CEO/COO/Operations Director/Consulting Director/CFO all work as a team and are never really limited to their "designated" positions. Our Operations Director has been billing on a project for 4 years.

Our company has an interesting dynamic. I would say everyone has the ability to "speak up" but we have designated management roles. I speculate there are some reasons for this:

1. Leadership - "junior" people need a certain level of mentorship. Many of them don't understand process and need managers to guide them on those processes. Hiring the quality of people (even at "the bottom") that your firms may attract is VERY difficult for the other 99% of companies like us out there. We simply don't have the whole techie seen watching our firm day to day and thus don't attract them to our company. Hiring SUCKS. Is that our issue? Sure. But this utopian business world isn't always possible -- even though we want it to be.

2. External client view - Our company works almost exclusively with blue chip companies. They expect to see CV's with years of experience and fancy titles like "Principal Consultant", "Management Consultant", etc. etc. With these titles comes the responsibilities of managing people. When I used to work for the largest biz software vendor in the world I remember there was a title "Engagement Architect". I was basically told that this position only existed because the client contact (mainly the CIO) needed a contact at our company that was on his same "level". Also, some people flat out do not have soft skills.

I think the overall point is this... this is not a lesson in how to run a better business. This is a lesson in how Jason is happiest in running his small business and wants to keep it a small business. Many of us love the agility of being in a small company and fear if we add more numbers we will lose that agility. (and that's probably true) 37signals has only increased it's employee base to 26 over 11 years. With the amount of press they have drawn there is no reason why this number shouldn't be 260. For the record, I'm not saying that it should be 260, I'm saying that they have probably favored higher profit margins on their existing employee base rather than to scale the business out vertically.

Fred Wilson once wrote "Marketing is for companies who have sucky products." In order to do this with 26 people, you need to find 26 very good people. Do you know extremely difficult that is to do? It's even difficult with just 2! Face it, most people SUCK at business. I applaud Jason (and others) who have had the fortune and ability to do such a thing.

I know this isn't quite what either of you have said, but someone needs to point this out

"real work" != "billable work"

"Everyone does billable work" seems great from the $ POV. That doesn't mean that people who are never billable aren't critical too.

Yep, Joel wrote about this here: http://www.inc.com/magazine/20080901/how-hard-could-it-be-ho...

...and was inspired by this Fast Company Article about some GE division that had no middle management... http://www.fastcompany.com/magazine/28/ge.html?page=0%2C0

I really appreciate the idea of having a career path that doesn't involve moving "up" to management. I'm a developer because I love development, and my own personal hell is managing a team and never getting to code.

Imagine how the engineers at Google think now that they're "back in charge" (/sarcasm) I'm sorry, I know it's fun to bash on managers but they exist for a purpose and asking engineers to be in charge is a burden I don't think many want.

> I'm sorry, I know it's fun to bash on managers but they exist for a purpose

The point is that it's possible to run a company where managers don't have a purpose and aren't necessary.

At Google's scale? Successfully? I find that rather unlikely. If you have any examples of this, please do share. Otherwise, all I can imagine is reducing the depth of the tree, and in some cases not to make it a strict tree, but a somewhat interconnected graph. (that said, throwing people into distinct "manager" and "engineer" bins is disingenuous anyway, a manager managing engineers had better have some kind of engineering understanding)

The Mondragon worker cooperative federation has 85,000 employees. If you look at how it's structured, it's not hard to imagine how Google could operate similarly.


Semco SA, the Brazilian conglomerate, would probably be a more relevant example of corporate re-engineering.

I mention Semco elsewhere in the thread, but Mondragon is 30x bigger than Semco, and yet still operates effectively as a horizontal organization.

I fail to see how that will help keep a driven, ambitious person happy as a career frontline customer support staffer.

Mondragon has annual revenues of €174/employee. I have trouble viewing that as a proven model.

I'm pretty sure the revenues on Wikipedia are off, and I can't find what reference they're using to get that number.

This study from around 2002/2003 puts their revenues at $8 billion annually with 60,000 employees. That puts their per-employee revenues at around $133k.


This puts their revenues at $24.2 billion in 2008, which is almost $300k per employee, if you use the ~85,000 employees figure.


It looks like I got confused because whoever edited the wikipedia article came from a country that uses '.' as a thousands separator.

Looking more closely, they're a strange sort of conglomerate. At a glance, I see two banks, two insurance companies, roughly two hundred industrial companies, a dozen retail shops, a dozen food-related businesses, a dozen research firms, and ten education companies.

In a way, it reminds me of Idealab, in that each new company is a new pool of equity (with the associated incentives that result from ownership), and that failures are not punished, but rather result in retraining or placement in a different firm. (for full co-op members).

It's also interesting to see that the founder was a priest who started off with a co-operative polytechnic school and bootstrapped everything from that base.

174/employee was calculated by assuming that the period in the number of employees is a comma while the period in revenue is a period (e.g. 17M/85K employees instead of 17B/85K employees).

So those figures would indicate 174,000 Euros per employee, which makes a lot more sense.

>asking engineers to be in charge is a burden I don't think many want.

if your environment has a significant amount of elements that engineers (or generally - employees) wouldn't want to be in charge for, then you are already deep into the problem and the shortest solution here is to assign managers to be in charge of those things.

The key is to not allow the situation to degrade into creating "demand" for managers. For example, once you create one manager for a group, s/he would want to talk to somebody, preferably managers, from other groups. That burden on the other groups would naturally trigger response - creating managers in the other groups. That starts to structure the information and decision flow the way that rank-and-file employees lose much of the visibility into the product and business and thus lose the ability (and motivation) to take even small decisions, thus requiring even more management ...

"The fairest rules are those to which everyone would agree if they did not know how much power they would have."

This quote really, deeply bothered me. One of the major differences between alpha males and the rest of the population is that they will always assume they'll end up in the top spot. Your prototypical alpha male won't even consider the odds of being on the bottom in that lottery.

To push it further, those same aggressive types will have the passion and voice to draw support for their views, no matter the substance. A "flat" structure overvalues the opinions of the loud and aggressive, with little room for more pensive contributors, especially women.

In other words, if you leave the authorship of the social contract to the loudest people, you may end up with a rather oppressive outcome. This is a universal rule, often overlooked by the alpha males who spend their time talking to Inc. Magazine.

You aren't making a very clear point. Are you saying that : "Alpha males will always assume they'll end up in the top spot, so when they choose rules for a situation regarding two competing agents, their choices will favor the agent in with the most authority"?

If so, then you are assuming a context in which the agents of the situation can change the roles in which they are assigned permanently. That is not the case in this context, because the agents (Jason Fried's support team) are rotated between management and worker, and no one is getting promoted vertically.

I thought hapless' point was very clear:

"A "flat" structure overvalues the opinions of the loud and aggressive, with little room for more pensive contributors, especially women."

Coincidentally enough, Jason leads off the article with an anecdote about a talented woman leaving the company because of their flat structure.

He was referencing a quote, not the structure of 37 Signals.

//are rotated between management and worker, and no one is getting promoted vertically.

While that's true, i think his original point was that the alpha males will choose decisions that favour themselves....Introspective motivational awareness is hardly a profitable trait in our current society* (*-whatever i have seen is constrained to indian IT)

The thing I've always liked about Jason's writing and approach to business is that he isn't afraid to say that they might not have the exact right answer. At the same time, they refuse to accept the "conventional wisdom" as being the correct answer; I think too often we believe that because most companies do things a certain way, that all companies should be run that way.

The takeaway for me is that you should be constantly questioning whether there is a different way to run things that enhances the performance of your organization as a whole. I have personally been privy to how the people with the most impressive titles frequently have the least connection to what's going on in the business. Some of the methods taken at 37Signals seem to be aimed at fixing this problem, which I think is commendable.

At the same time, it seems a shame to have to let go of a good employee because they want to take on more responsibility. If their view on more responsibility is simply a bigger title, then perhaps they weren't the right fit for 37Signals. However, in my opinion, ambition and competence should be rewarded, so it seems like there may have been a better way to handle the situation than choosing between staying in the same role and leaving the organization.

Thanks for this comment.

You're right - we don't have the exact right answer. We're always trying to find the right answer for the right time. We may have a different opinion in 3 years or if we have 35 people. Who knows, but we're open to figuring it out once we get there.

In the end we want to build a company that allows people to do the best work of their lives.

Love the idea of rotating "managerial" or "lead" person in a small group.

I'm on a small team within a larger organization that we support (in dev and tool usage). A challenge for us is that people in the larger org are used to having a manager to route their requests to.

I may give a rotational approach a whirl. But right now, one of my primary roles is as s&!t umbrella and I don't want to overly burden my real producers.

So what he is essentially saying is, they're an anarcho-syndicalist commune, taking it in turns to act as a sort of executive officer for the week, and all the decision of that officer must be approved at a bi-weekly meeting by a two-thirds majority.

There is some lovely muck over here.

Is this getting downvoted because people didn't catch the Monty Python joke? Or because they didn't like this specific reference (but accept the parent's reference)?

It might be because a movie reference that's relevant to the conversation at hand can be humorous, but a movie reference in response to another movie reference is just regurgitating a script. The child comment doesn't really add anything to the conversation other than "I saw that popular movie as well."

Maybe because it's a misquote. It's "lovely filth."

(Also, the committee only has to ratify the officer's decision with a simple majority in the case of purely internal affairs.)


This fits nicely with the study which shows that organizations are more efficient if people are promoted at random ;-)


I would like to commend Jason Fried on establishing a source of free and recurring advertising in a massively distributed publication that has had his target audience cornered for years. Not only is this free advertising, but it is the highest quality of advertising congratulating small and medium size business on being flat (which they usually are because in a small army, even the Generals are on the front lines) and reassuring them about the benefits of being so (imagine you’re a 6-person company where everyone does everything and you have just read this article: now decide between buying MS Project and 37 Signals’ Basecamp for your PM needs - cloud over your head says "Jason gets me, man!"). Jason, if you’re reading this I know you’re smiling - you have my vote for strategy!

There is a PR lesson in this for all of us!

I'm the first one to be skeptical about the universal feasibility of 37signals bootstrapping principles (that's an easy philosophy to espouse when your product was bootstrapped by leveraging early success in client work and community mindshare in what was a nascent market at the time), but your dripping cynicism is quite offputting. The fact that Jason's ideals serve his interests is not necessarily an indication of of any machiavellian scheme. In fact if you think about it, this is a necessary byproduct of the marketing of ideas by anyone with integrity.

Since when is benefitting oneself by adding value to something a bad thing or a "Machiavellian scheme"? This is what you need to do all day: figure out how to establish yourself inside your customer’s head, no matter where he or she is - get the singer they listen to, get in the publication they read, get on the radio show they listen to etc. As a CEO you should chart out your customer(s) and go after them - establishing yourself as the best choice (every demographic has a different definition of "best" so be careful here) through all and any mediums.

Also, think about it like this if you feel "evil" doing this, as a CEO you are the mind of the organization, you must think for the organization because like a baby it cannot fend for itself. I am commending Jason for being creative - most companies throw away money year after on ads in magazines like Inc that barely get glanced at - Jason, on behalf of his organization, has sold you his product - correct me if he hasn’t.

37signals can do this because of their hiring practices. They need to be extremely picky with the type of employee they hire. Their hires need to be able to function within their unconventional structure.

This type of information doesn't translate well to most other companies. I hope 37signals' audience gets that. For software startups, many of their ideas are exceptional and it's fantastic to see real-world examples. For already-established companies and companies that can't be as picky as 37s, testing this type of structure seems unnecessarily risky. I believe 37s has addressed this in the past, and Jason has in his Inc. writings. I just hope people are paying attention.

What Jason Fried is expressing is something I've been pondering for a while. And I think we'll (hopefully) see more of it (sorry, manager-types).

In my experience, managers in most departments have essentially taken the role of sheep-herders. So, I started to ask myself: why do I need a manager when I work well on my own, making smart, educated decisions that are based upon the ideals and successes of other smart, educated and passionate people?

After all, I'm being hired for my prowess, no? If I am, do I need a manager? And shouldn't you always hire people who have these sensibilities?

I think the message I find within this rubble of contemporary and progressive ideologies is: Hire smart. If you have a good team who understand their roles and how it pertains to the goals of the company, you don't need managers - not for a small or mid-sized company, anyway. Basecamp has been the best PM I've ever worked with - alive or binary. Software has already begun to facilitate the role perfectly for me.

Though I am a dreaded project manager, I agree mostly to what you are saying. But in the field I work in, advertising, I do think that its necessary and that my value is additive.

My responsibility is mainly to be a shit umbrella, so my creative and developers can do work, especially when we're all working on multiple clients. I'm not doing my job when I have to have them in meetings, in fact we don't meet save for face-to-face discussions.

That and of course dealing with the rowdiness of clients, hammering out scope, keeping things to scope, and being the bad guy when necessary with the client. (This is actually harder then people realize, our default human reaction is to help people, and it requires taking abuse in the extremes that is simply part of your job)

This article sums up my views nicely: http://weblog.muledesign.com/2010/10/project_managers_not_ca...

I like the shit umbrella metaphor. I have recently taken up some PM-type work, even though I am technically an engineer, and a junior one at that. Now it is mostly cleaning up some stuff and keeping other engineers out of the wind. Things like making overviews of calls in a certain time-period, and making sure contracts are set-up. Stuff that most developers could not be bothered with, but I enjoy it.

So for me doing PM work is actually more of a horizontal promotion, doing some different work because I like it, while I am definitely not "in charge" of any people now.

I guess my point is that even picking up PM work can be a horizontal promotion.

I would say in high-output environments that a PM is a good thing, if the PM in question is good at being a PM. I've met a few, but many more terrible PMs unfortunately.

Oftentimes PMs (or better, their organization culture + workflow) have thrown a wrench or 2 into production and caused cascading failures because they forgot a detail, or couldn't admit to fault. The culture is surely to blame amidst most of these catastrophes.

The fact that you can admit that your PM skills need work already makes you a frontrunner for being a much better PM than most.

The central problem I see is most PMs don't want to be PMs, and are tricked into being hired as account people. I like what I do, and my team knows I have their back, we do quite well together.

At my previous work, the head of R&D was an awful manager that couldn't light a fire under foot-dragging devs or keep things from going off-track (and software was routinely released in beta years late). But he had an amazing talent to shield his devs from the godawful CEO and other pointless distractions. This is a much unseen function of managers - shielding you from paperwork and interruptions that really aren't directly related to your job.

Managers can be useful to say "no" to business. Otherwise you have business asking for a new feature every other day. Even if the engineer says no, it's still an unnecessary distraction and time killer to listen to everyone's proposals.

This is a good point to bring up. I agree in theory that managers are useful in gate-keeping roles. The trouble is that very few managers actually possess the power to be gate-keepers. Investors, clients threatening to leave and other fast-paced blame-culture mentality all but strips away their usefulness. In my experience, this at times makes managers feel worthless, expire, or worse - erratic.

My take is, manager positions are ripe for buck passing (to and from), collateral pressure for performance and abuse. Maybe some people can deal with that type of nonsense but the number of those who can't certainly outweigh them.

I can see how a manager might feel by this message. My thoughts are: If you're providing a great service to your underlings and your clients than you're in no position of regret. Contrarily, if you feel sick to your stomach when you go to work everyday and you're proud of that to a point of defense or fear, maybe it's time to reevaluate.

How much time can you spare learning exactly what each of those other people are doing and how it might affect your work? For that matter, how much time can those other people spare to explain yet again exactly what they're doing, in enough detail for you to realize how it affects your work? As I see it, much of management's job is to protect me from investing most of my time in conversations which didn't need to happen because they won't pay off.

While I totally buy what this espouses, I think it is probably incredibly hard to scale. They are doing well if they keep things flat(ish) for 26 employees. I can't really see it working at all for > 50.

And while they may not have anyone with the job title CTO, I'd be very surprised if DHH was anything other than the de facto CTO.

Why does it have to scale? 37signals took 11 years to get to 26 people. It might take us another 11 to get to 50. We can optimize for that future when and if it arrives. But making your environment worse today because you worry about some possibly maybe future is the essence of premature optimization.

For 37 Signals it doesn't have to scale any time soon, but the grandfather was (tacitly) in reference to how many look to 37 Signals as exemplars of how to run small to medium tech businesses (why else write about it in Inc?)

I think if I already had a business of 10-15 people this would be a good model to attempt to follow. If I had a business of 40+, I would be far more wary; the question of scale is for those who may try to fit the model to their existing business.

Many companies need to scale up people as they grow because their ratio of human involvement to revenue produced isn't as high as 37signals'. Almost anything other than the specific style of product development that you guys do falls into this category.

The assumption when you publish an article is that you're doing it to provide examples to others, not just talk about how good you are. In that situation, I think it's fair to point out the limitations and areas where this strategy might not work.

I agree it doesn't have to scale, and you certainly shouldn't setup an organization around projected growth 2 years out. That may never come, and you also may not want it to come.

However, if you want your company to grow into a Groupon or Zynga, and you have the rapidly expanding demand to support that, then you need to scale. I don't think Groupon could run on 26 people, so they needed to scale to get to such massive revenue figures.

I specifically didn't use Facebook as an example because I know your feelings about them. I figured Groupon at least has the revenue to support their valuation :)

You have it backwards. You don't design your company structure in the hopes or anticipation of becoming Groupon or Zynga. You deal with that growth when it happens. You optimize for that scenario when it occurs.

You can't "want" your company into becoming a Groupon or a Zynga by design a certain organizational structure. Startups try that all the time by preemptively hiring tons of people to sit around and wait for the hockey stick growth that never comes.

I know, I'm not saying you should design your organization or hire around the idea that you'll be at Groupon's level in a year. That would be a waste.

Well let me ask you this. When is it the right time to reorganize and evolve from the flat structure because you are growing like gang-busters and can't keep up with the pressure on all areas of the company (support, development, sales, etc.)

I'll tell you when we get there, if ever. Although, as mentioned elsewhere in the thread, there are some good examples of enormous companies running counter-intuative-for-thier-size organizational models and doing great.

Its a different industry, but take a look at Semco SA:


Their industrial democracy approach has scaled quite well, with 3000 employees.

Maverick: The Success Story Behind the World's Most Unusual Workplace, by Ricardo Semler. Costs $10 at Amazon. Worth it.

His book "The 7-day Weekend" is good too.

Definitely a great example... but I find it disappointing that it is pretty much the ONLY example of this kind of company at that kind of scale that is brought up when topics like this are discussed. If it worked for Semler, why hasn't anyone repeated it?

Well, there's also worker-run democratic cooperative corporations like Mondragon:


I think the biggest reason most companies don't recreate Semco's example is that nobody wants to think of themselves as expendable, especially people who can decide whether they're expendable or not, as the top management positions of a company are in a position to do so.

By saying "yes, our employees can make these decisions without us, we just have to give them the structure and support to do so", people at the top are admitting that they're not as important as our society tells them they are.

In Semco's case, Ricardo Semler was happy to admit that, because he felt the stress of running the business was killing him, and he welcomed the idea. A lot of people in his position would take anti-anxiety medication and something for acid reflux, and keep going, convinced that their position holding the reins is indispensable.

It's unrealistic to not promote people. If you run a "flat" organization, you're telling your employees "I don't care that your resume indicates no progression." That's a real career limiter, and it can be perceived as an underhanded way to retain talent.

Also, more pragmatically, how realistic is it to have 30 direct reports?

Is "unrealistic" a version of "that wouldn't work in the real world"? Of course it's realistic. Jason is describing how things actually work at 37signals, not how he'd wish them to be. So we have an entire staff of programmers, many who has been with us for years and years, who started out as "senior programmer" and has stayed with that title. The progression has been measured in salary.

Whoa... cultural whiplash. I get that there might be places where this kind of thing is a concern. Really, I do. But I can't imagine applying to work at a company where they'd look at the line on my resume that says "3 years at 37signals" and worry that I hadn't been promoted in all that time.

I would think working for 37 signals alone would be a big enough evidence of how awesome you are. Also the kind of company that is looking at your "progression" probably isn't the place to go.

Maybe for a designer. What about the accounting clerks, administrative assistants, CSRs, etc. that any company of size needs?

It's "unrealistic" because there are social realities that exist outside of the valley and tech blogs. It's true that it's a good company, and it's true that it's a darling company here, but there will be sectors that haven't heard of these kinds of places.

And, again, once you hit something of size, some delineation is a necessity. You don't have to model the Marine Corps, but when you have 30 people arguing over something, the buck has to stop somewhere.

Unrealistic implies that it's not going to actually take place. It is taking place. We've been running like this for years. We've had very little turn-over and remain good friends with people who left. So it seems to be working out excellently for people presently and formerly employed by 37signals.

The buck always stops somewhere. Whether you're two people or three people. At 37signals, the buck stops with Jason and me. Now it happens to be that when you employ smart craftsmen and women who care about what they do, there's very few loose bucks to stop.

Doesn't "My salary kept getting raised because I kicked butt" also mean something on your résumé?

You don't typically put a salary on a CV.

I would probably have no problem putting "My salary kept getting raised because I kicked butt" on my resume. If they can get past the fact that I worked at a porn company, a little grandstanding probably won't hurt.

> If they can get past the fact that I worked at a porn company,

Employers that worry about this are rather silly. I'd hire a candidate even if he was starring in said porn productions.

I'm kind of surprised at how most of my former co-workers will not put the common name of the company (which is not offensive or profane) on their resumes and put the legal name instead (the legal name is pretty generic-sounding). Some have even requested that I change the name on my LinkedIn profile or mark it as "I worked with so-and-so at different companies" so they won't be directly associated with it.

Maybe it's just a scarcity mentality thing; they're worried about turning off potential employers, but when you're constantly getting contacted by recruiters I guess it's hard to feel like it's a disadvantage.

I disagree that this would appear a "career limiter" to smart people. Most people in tech work for the same company for only a few years, so it's challenging to make any significant career advancement within an organization.

But give them an opportunity to serve as team lead on a successful project or several, and you've got plenty of resume mojo to spread around to everyone.

Maybe your title won't reflect your progression, but if you list responsibilities and things learned on your resumé then that would be a clear indicator of progression.

Jason seems to imply that managers are useless when he says "We do not have room for people who don't do the actual work".

I think this is very naive of him, and a little selfish. He's enjoying the title of "President" which, to me, is a purely managerial position. I doubt if he considers himself useless, but he's happy to label other managers useless. I might be wrong, but it seems that he's either too selfish to see other people take away some of his control or he's afraid to tackle the problem of a growing company. Both of these will have negative consequences in the future.

Just for the record, over my 14+ years in a technical field, I have been a manager for 5+, and have given up that title twice before to focus on more technical work.

He's enjoying the title of "President" which, to me, is a purely managerial position.

I doubt that what he does in that position is purely managerial. I think you may have a hard time seeing why they value what they do if you impose your own viewpoints on them.

Perhaps you're right. With a smallish company, it's possible for everybody to jump in and do whatever needs to be done. As a company grows, though, this will become impossible. Many a great company in history have suffered spectacular deaths simply because they couldn't scale correctly with their success.

In my opinion, it will be impossible to have 200 employees with a flat structure.

I don't care about what 37signals might/could look like at 200 employees. We have 27 employees today and we grow slowly. We're optimized for today. If we have to change down the road we will, but there's no sense in spending even 5 minutes worrying about if what we do today will work when we're 10x as big.

> What we learned is that adding a dedicated manager and creating a hierarchy is not the only way to create structure.

I think the misconception with the article is that the title (and focus) is "Why I run a Flat Company" as if this should be the standard for modern business. This works for 37signals, because (1) you have actively chosen to grow very slowly when you could have otherwise grown much faster and (2) you've had the fortune and ability to attract some very good people. Many businesses don't desire to do number 1 and find it very hard to do number 2.

Jason is a designer by trade, and is routinely involved with the design of 37signals' products. In the same vein, DHH is a partner in the company, but still writes code (as I understand it).

To clarify... "President" isn't a title I use, it's a title often ascribed to me as the founder of the company. Whenever anyone asks for my title I tell them we don't have titles. If they require one for whatever they are writing, I say you can say president or founder if that works, but I'd prefer if they didn't say anything other than founder since founder isn't a title as much as it is a fact.

Re: design. Yes, I design. It's the primary thing I do. Design, sketch, work on product simplification, design/think about user experience, etc. DHH still codes as well (he's working on some nice enhancements to Highrise right now, in fact). Both David and I would prefer to design and code all day, but it's not always possible. Lately it's been more possible so we've bee doing it more often.

> To clarify... "President" isn't a title I use, it's a title often ascribed to me as the founder of the company.

I apologize if I got things wrong. I was just going with whatever is here: http://37signals.com/about

"Jason Fried is the co-founder and President of 37signals."


"Jason co-wrote all of 37signals books, and is invited to speak around the world on entrepreneurship, design, management, and software."

with that kind of workload, I assumed you wouldn't have enough time for anything else. I never said whatever you do is useless. Far from it. Non-technical work that is focused at growing a company is more important than any technical work. After all, what can you do with an insanely great product that you can't sell because you didn't prepare for it properly?

Good catch there. Gonna pull that president mention.

Re: speaking. Not speaking anymore. Really enjoying getting back to product design.

What about creating a career path that doesn't involve moving people into management? One that involves more prestigious titles (e.g.: sun had a 'sun fellow' title), and significant salary growth (20% here and there isn't significant, imho)

As the organization grows, i can't see a totally flat structure working - you're going to end up with people who have been there 5 years, wake up one day and realize they have the same role and similar salary to what they had where they started, realize they have no career path with their current employer, and will move on.

"you're going to end up with people who have been there 5 years, wake up one day and realize they have the same role and similar salary to what they had where they started"

1. The people who are worth the most to us are the people who do the best work. We pay accordingly. "Management-like" salaries go to the best workers.

2. You assume people want different roles. Some people do, but we work hard to hire people who love their craft and want to keep crafting. A programming being pushed away from programming, just because they've been at the company for five years, is not the desired path for most programmers who really love programming. Same goes for designing, etc.

Speaking of titles: We tend to avoid those too. We've had a good bit of debate inside the company about this, but we're sticking to staying as title-less as possible. I don't think they're worth it in the long term. They set up boundaries between people that I don't believe are necessary. Might that change if we had 100+ people? It could, but I'm concerned with the now instead of the imagined future.

Looks like Sarah has started her own thing: http://cosupport.us/

If you're interested in hearing more about it, she was on Dan Benjamin's Daily Edition in February:


I like that he mentioned in 2 places how he/they supported people in the best way for those people even when it was clear they weren't going to work for 37signals anymore. (In one case they helped someone find another job, in another they helped someone start her own thing.)

It's important to set other people up for success, whether it's success at your firm or at someone else's. They're not "human resources", they're people!

I believe some sociologists found that personal relationships begin to break down at about 150. Beyond that it is very difficult to maintain meaningful interaction in person. Online relationship numbers are much higher so maybe a distributed team like 37Signals can get away with this for a while longer.

You're referring to Dunbar's Number:


Thanks for the concrete link! I never knew the name.

You should never say never, but I don't think we'll ever be 150 people. I think there's a good chance we'll never even be 50 people.

Although, as many others have pointed out, there are some great examples of huge companies running similar strategies to good effect (Semco etc).

What happens if you're inspired to branch out in another direction? Would your plan be to form a new company that also stays small and nimble? What would you see as the benefits to separate small organizations v. a larger one with smaller divisions?

Skimming the comments, it seems a lot of people are missing a significant point - the employee is in customer service. Customer service is a dead-end job. It doesn't take long to get on top of your game, and there's really nowhere to go.

Managers and marketers get new products and changing business conditions to keep them interested, developers get new tech to explore and tech debt to resolve.

Customer service... is easy to master and once done there's no new fields to conquer. It's ultimately boring. Fine if you want a job to show up to and just do, but if you want to be interested in developing/advancing skills, it's not going to happen in customer service.

I actually find it a little insulting that the tone of the article is a little "well, the developers can handle 'not advancing', why can't the customer support person?".

The professional development tree for customer support looks like a stump.

I dunno, they could get more involved in the product so that they can diagnose more simple issues without having to pass it on to one of the devs. They could get more involved in design based on the feedback that gets passed through them, they would have a good sense of what is well received and doesn't work well.

For straight up customer service person you would be right, but I think there is a lot of room to move there in a flexible organisation. Depends on the person I guess, it doesn't surprise me really they would see more turnover here than they would of developers.

There is still only so far you can go as customer support. I wear that hat in my current company (along with a few others, including planning) and no, there really isn't much depth you can get into with devs without becoming a dev. Most devs (not all) largely disregard customer support and feedback anyway, since support doesn't 'see the big picture' (a less charitable person would say 'because support works with existing products, and we've already covered that now-boring ground' - after all, how many laments do we hear of devs that have to work on 'maintenance' rather than new prod development :).

The key point is that even if you're interacting directly with the devs, you're still not really being proactive or learning concrete skills. The job is still pretty much reactive. If you're driven or feel the need to develop, there really isn't much to strive for. You're not the one that gets to figure out code, or create marketing campaigns. There's not really any ongoing feeling of "I made this!". And if you move into design/planning, that's not really support anymore.

I don't mean to rag on support - they are the ditch-diggers of the IT world - but if you're driven, then it's got nothing to offer you after a fairly short while.

I was one of those experiments, as the Operations Manager, and I like to think it went pretty well. As many people here have rightly pointed out, though, being hands-on is the key. I continued to do a lot of the day-to-day technical work.

The main difference between what I did, and what the rest of my team did, was that I had the added responsibility for dealing with partners and vendors, negotiating contracts, scheduling hardware installations, and the like. The rest of the team was able to remain completely focused on the system administration issues that we cared about while I split my time.

For our team, it made good sense. For the other development/design teams, the way they're run makes sense. It all depends on context.

There's a saying in team sports along the lines of, "Players play the game; Coaches coach."

Employees play the game (ie do the work), but I think a good manager/leader can make a big difference.

Sure, Michael Jordan was a great basketball player who had good teammates, but Phil Jackson must be doing something right in order to extract the talent out of his players in just the right way to win year in and year out. And that is hard.

Perhaps there's just not very many good coaches or managers, which is why there's such distaste for "management."

But every good team, organization, or company has a great "manager" or "managers".

In the case of 37signals, it seems that person is Jason Fried (he is the CEO).

I like the idea of keeping people closer to what they are good at. However I'm afraid this strategy will backfire on your employees if they are forced to leave to another company for whatever reason. In an environment where every company does not have this flat hierarchy it is strange to see a person who has spent 10 years in one place and has not advanced to some managerial position. New employer will think that he is unfit in some way (even though the opposite is the case) and will probably not even give the person a chance to explain themselves.

We're optimizing for having the best work environment today. Prioritizing what some employer somewhere else might think a decade from now over improving working conditions today seems like a bad trade.

I don't understand why this has to be a binary choice. The real issue at hand appears to be that people get promoted to positions for which they are not well-suited, and then stay there, because they do a lousy job (agree that this sometimes happens... although eventually they probably get fired).

What if "promotion" doesn't necessarily mean domain over other people, which seems to be the other issue. There can only be one SVP of Technology, for example, in most organizations.

What if, like in school, you get "degrees" for achieving things within the organization? If you launch a major new product successfully, you achieve a "master programmer" honor, and the associated pay benefits. This doesn't necessarily give you any right to control others in the organization, but it rewards employees for stellar performance. They can then put that on their resume if they go to another company.

I would argue it is a little short-sighted to say that the company can exist in a vacuum, when every other company in the world, for the most part, operates with a title. That doesn't mean you have to play that game within the company, but you should at least accommodate employees' concerns that they work for you for 10 years, and then can never get a comparable job because it looks like they never advanced. If they stay in your good graces, it probably isn't an issue, since I'm sure you will give them a good recommendation. But if not, they could be in big trouble when they go to find a new job.

If anyone ever faces this thought-experiment problem when they leave 37signals, I'll be happy to bestow them a non-managerial fancy title of their choice. If someone leaving 37signals thinks they need to have a "Master Programmer" title, they'll get that. Problem solved.

I realize that particular suggestion regarding degrees is a little Mickey Mouse. My point was simply that it is quite risky for them to have to rely on your good graces when they leave the organization. What if you're not around to bestow that promised title 10 years from now? What if after a string of successes, that employee makes a mistake and tanks a product, and gets fired on bad terms? This system works fine internally, especially for management. It adds a lot of risk for employees, and that risk escalates the longer they are with the company. I'm not telling you I have the right answer, but I think it's worth considering.

What's the risk? That I don't take a job sometime in the future from a short-sighted HR drone who can only judge me based on my resumé? If the lack of an archetypal career path on my resumé is a barrier, it's probably not a job worth having.

There are a whole host of intangibles that will probably break down by individual employee. An employee may have to move to a different location for personal reasons where the only option is a job that hires via HR drone.

A large part of management structure in the first place is to try to insulate the organization against changes at any level. Want to replace the entire C suite and the board? No problem... business goes on as usual. With this structure, the company becomes highly dependent on the people at the very top. Any changes there might have the effect of blowing up the whole organization, because there are only a few people who are unquestionably in charge. I think there are lots of benefits from trying to reshape the corporate org chart, as I mentioned in another post, but it is also dangerous to be overly idealistic in terms of how your employees regard this structure and the risk they place on working under it. If most of them feel its appropriate, then I suppose there is no problem. However, I wouldn't automatically assume that's the case, just because they have chosen to work without a title. Also, a large part of their continued work at the firm may have to do with their relationship with you personally, which may put the long term viability of the business at risk.

The hypothetical company you've described here sounds very irrational to me: discarding a candidate with 10 years of experience (at a very highly regarded company, no less) without ever even speaking to the person.

Why should anyone optimize for the irrational job-hirers out there? This does not sound like a path to happiness to me. Besides I find it unlikely that the kind of person who is happy at a small company like 37s for 10 years would want to work at this hypothetical irrational company.

I'm the type of person who would work at 37signals. If I were to leave 37signals, any job that I would be applying to would have heard of 37signals and presumably my work there would speak for itself. If I got nixed because of detailed line-wise consideration of my CV I would take that as a sign that I was better off far away from that company.

There seems to be a rift here between what people are good at/what they want to do now and what they want to do later. I consider myself a good programmer, but it's not where I want to be forever. I have always wanted to branch out and learn multiple facets (management, service, sales, etc.) and this seems like it would be limiting in that regard. I guess I just wouldn't be the target of 37signals?

Without wanting to sound snide, do you look for people who want to stay in the same role forever? It surprises me that people's ambitions to branch out and take on more responsibility haven't caused this to come up before. A salary bump, more benefits, and more vacation time wouldn't help me placate my desire to learn about other skill sets.

People branch out all the time. Part of being flat is that there's very little division. If you are a programmer who has a good idea for a new workflow process (i.e. management design), you propose it, and we usually try it. If you're a designer who want to learn more programming, we encourage it.

You can easily have a big impact on all the other domains of the business even if you're just a programmer.

Hell, we just had Craig Davey switch from being a programmer to being a designer recently!

I think it's a good thing that a potential employee of Jason's can read his blog and be warned or at least aware of what kind of company he's running and potential future careers there. Personally as a developer I'm looking forward to gaining more managerial experience as I can see how you can make bigger more exciting things as you have more people contributing. Incidentally I work for a company twice as big as 37 signals although growing quite fast in comparison and I can see that the people in the nontechnical roles quite enjoy the possibilities (and realities) of progressing to different and managerial roles. It makes me happy that they can stick around and don't have to leave when they might otherwise stagnate.

This is a lot of BS, spoken by a Founder with no understanding of employee needs. Employees do not have their eyes set on always staying where they are. They are not reaping the monetary rewards the way a Founder is. To an employee, EVERY job is a stepping stone to the next one, eventually. This Founder is completely self-absorbed.

Job titles are free and it helps the employee along on their career. No, they shouldn't be inflated, but they shouldn't be held back either. Eventually the employee needs to put that job on their resume and if it looks like they were an entry level person because the Founder was a jerk about titles, it's better if the employees leave now instead of later.

Appears there are two types of people,

The 37Signal employees: * Like their field and want to be 'hands-on' * Don't mind staying in the same company for 10X years.

The supposed norm: * Prefer to advance to other positions vertically * Like moving between companies (for challenge/change).

Makes me wonder what is the difference between the two types of personalities and how those affect the organization.

For example, is there more or less innovation in 37Signals ? Are people more ready to step up and fix/report problems outside their immediate responsibilities ?

I don't think there's really a difference in employees. Everyone reacts to incentives, and it sounds like 37signals has removed titles with pay. Which honestly, I'd much prefer. In many ways titles are a substitute for flat pay.

I'd rather have a flat org (and no titles) with wildly different pay (and extreme bonuses, profit sharing, equity grants) -- rather than relatively flat pay and lots of titles (which tends to be the norm in industry).

What they removed is not just meaningless titles but rather all management positions. I, for one, don't want to code my whole life. Something 37signals can't offer me.

But there are people who are content to code forever, and I wonder how they differ from people like me.

" I, for one, don't want to code my whole life. Something 37signals can't offer me."

I'd say it's just the opposite. The lack of organizational hierarchy and title means that people can work outside their normal expertise more easily. Nobody is stuck in a box.

Designers pick up programming skills as their interests allow. Programmers with UI ideas can try them out. Both contribute with writing, with workflow ideas, and on customer support. We've even had people completely shift roles from programming to design.

This "flat" system sounds very nice, and I really like this rotation of responsibilities and "managerial positions" to all members. I wonder though, if the same principle is applied to the CEO and Business Owners positions as well (both in respect of decisions and profits). If not it doesn't seem that flat to me.

So preaching "flat" while being a business owner sounds a little suspicious, because you are preaching everybody about flat while standing above them.

I hope 37signals is truly different (I cannot judge since I don't know crucial details), because if at some point after ten years the business is sold to a big corp, and everyone finds himself trapped below ten levels of management, without a career path, 4 day workweeks during the summers of youth will sound like a bad joke.

Why don't you just split into two little 37signalses when the time comes?

In all the proposals / solutions mentioned here for dealing with growth while maintaining a flat culture, this is one approach I haven't seen yet. It worked well for a Dutch consultancy firm called BSO, which reached over 6000 people in the 90's, all organised into near-independent little companies of 50-ish people each, all targeting a different market, but each with the same culture and values. The firm itself was a flat company of these little companies (called "cells"), so effectively there were just 2 to 3 layers of management.

(http://www.extent.nl/articles/entry/origins-original/ if you care about the details)

In many companies the managerial titles are invented incentives, not necessarily they "manage" people. They exist as part of the incentive package, and certain companies attract certain personalities who would be happy with titles. Management position gives one probably a different satisfaction, "doing better than the other guy", and assumed better pay above the managed is all that is needed. It's a kind of a distraction. It is not bad unless it kills nurturing leadership environment. The best case is that laders become chosen managers by their peers. The worst case is that those who can not manage become assigned leaders (managers) by "the management".

"Besides being small, 37signals has always been a flat organization."


"We've experimented with promoting a few people to manager-level roles."

So they are flat, with no chief anything, but they have 'manager-level roles'? Am I missing something or is there a contradiction in his description?

Edit: Got it, 'experimented' meaning that they tried, didn't work, and they went back to flat. Thanks for the responses.

Besides that I find that even with no job titles or formal roles, people within a company tend to self-organize and take on de facto roles. The only difference is that it isn't formalized, and people who end up managing aren't being paid manager salaries or getting manager options.

We've experimented. That's how we've become more confident in our decision to be as flat as possible. We believe it continues to be the right direction for us.

Re: manager's options... We don't offer options so that's not an issue.

Key word: Experimented?

Past tense, too.

My cocktail title at 37signals is "Product Manager," so whatever experiments are going on with management aren't completely past tense. I do UI design, work with the programmers, help design and dev work together and contribute strategy ideas to the partners.

Despite the cocktail title, there still isn't a clear "path of authority" or anything like that. The buck stops at the partners, and the rest of us function like a meritocracy. The ones with a track record or expertise on a given topic have a respected voice, and there is always opportunity for expertise to shift or widen over time. It's that fuzzy-edged quality to the role that makes it feel experimental for me.

The only way to solve management issues with a high degreee of confidence is to stay small enough to avoid management. The NASA analogies are problematic, however, because large organizations do indeed manage to manage themselves while completing critical projects. Which is more interesting? Large-scale management, or head-in-the-sand? I could personally never work in a large-scale organization, but how can we all avoid these issues and still create a highly functioning economy, which produces both critical and lifestyle goods and services?

"And because we don't have a marketing department, we don't have a chief marketing officer."

37 signals is amazing at marketing. Their blogs. Their books. Video Lectures. Mission statements. Guest posts.

To clarify, we think everything we do is marketing.

From how the error messages are written, to customer service, to copywriting in our emails, to the cancellation process, to clarity of message, to how every single thing in each product works, it's all marketing. Marketing is everyone's responsibility. It's the sum total of everything we do.

37Signals is ubiquitous in a certain way on the internet - we are smiling! We are about our products! No look over here we are about education! Come to our conference and learn how to run a small business! It is like Martin Fowler and Joel Spoelsky had a child and it was nurtured by Saatchi and Saatchi.

This sums it up in my mind - this is just smart business. Having such a core need of a business being wrapped up in a "Marketing Department" contributes to a lot of corporate BS in many corporations.

While I completely agree that flat vs a vertical hierarchy should be assessed from organization to organization, I tend to prefer flat structures mainly because they allow for the culture to mold perception of progression over an existing structure itself. What I mean by this is that if you have a structured way for internal progression (usually vertical) people mold their perceptions around that ladder no matter how much you try convincing them otherwise.

You can clearly see this being played out within the big 4 accounting firms (I recruited for them and from them). Within the firms its extremely vertical in that progression is dictated largely by how long you stayed until you hit partnership where its strictly vacancy. Thus, you're basically looking at steady yearly promotions until you reach being a senior manager after 6-8 years within the firm.

This is where I was able to take the most senior people usually in a seasonal manner pretty easily. This is because after being a senior manager you really have only two trajectories within the firm, associate partner or partner. The AP is basically a position they created to please senior managers who sounded too old and weren't good enough to be partners. So what you generally see is 3-4 hotshot senior managers all vying for the 1 partnership position that will be available that season/year. Inevitably I'll have 3/4 partner potential senior managers leaving because they know they're better off leaving the firm and going to an industry position or worse a competitive firm. They already know the stigma of being an AP.

Thus, you don't simply see attrition at the top level, but the attrition of the very best at the top level and the rest being APs. What's worse is those guys who are the best usually have a loyal following within the firm. Well guess who gives me a call after placing that senior manager as a hiring manager where they're building a team? Now you see an attrition of even better people who you were probably underpaying at junior positions leaving the firm for better pay and better hours. The only guiding light there is you're hoping that senior manager becomes their client in the future.

Thus, you see a system where the highly vertical nature of the structure led to a culture where attrition was the norm. While it might be naive to think so, I think being a flat structure might give a better chance for the culture to shape that perception of the promotion and have them "feel" it rather than perceive it.

Since you run a fairly flat organisation, do you ever promote the exceptional people that have worked with you for many years into the 'partnership' area of your company?

I read somewhere that David was originally a contractor and then became a partner, but has this occurred for any other significant contributors to the company. I'm thinking 10+ years, has a solid and respected reputation within the company, etc.

But what if a young startup company want to use this approach. consider a programming team which its developers are not in the same level of expertise and ability. Is it possible for this team that the manager rotate among team members? Doesn't it lower the performance of the members and the self-management of total team?

Sarah Hatter, who is the employee described in this story, left an insightful comment that unfortunately showed up as a child of aless insightful comment and is therefore buried. Check it out here: http://news.ycombinator.com/item?id=2417278

I find it outstanding that they have 5m users and only 5 support staff!. I would love to know how they manage that.

Simple products mean less support requirements. Although, we're going to be growing the support team so we can reduce the ticket load per person and create more time for other support-related writing/design projects.

I would presume they use the same technique that Brent Simmons suggests: treat customer support as an engineering constraint, and design solutions to minimize common support problems. The quick feedback loop of the web makes it much more possible than in other industries.

My view on managers: Turn your org chart 180°. All managers are supporting those "above" them, this is their primary goal. Usually companies have this the other way round. This also means there is no problem with "useless" middle management (Scrum calls this manager type ScrumMaster).

How does such a model work for tasks where you need to perform long term planning? Similar to national elections, such a model could result in policies being changed every time a new manager is in place.

That idea about rotating leadership within the customer service team is brilliant! I wonder if changing it every week is too frequent though - what about doing it biweekly?

Way to reward the one critical person who actually has to do the hard work of interacting with your customers AFTER they've been sold and paid you money.

That's the hard part because it's only dealing with problems and never "great to have to call you".

I guess customer service these days is disposable and easily replaceable.

I actually feel like I got quite a reward for my time at 37signals - Yes, that's me Jason's talking about :)

Since this topic is sort of centered around a small part of my experience, I'd love to shed some light on a couple points:

True! I started my own business! For about 7 hours of my life I didn't know what I was going to do if I couldn't bloom at 37signals, and then I told David and Jason I was starting my own company, CoSupport.us.

True: I left 37signals because I really wanted to do more support-wise, and the needs of the company didn't mesh with my ideas. They need people full-time on support, they don't need someone who's not there responding to tickets. After 4 years of answering 100+ emails a day, I wanted to do other things. Needing me on tickets interacting with customers is what Jason means when he says they can't have people working on a team not doing the work of the team.

True: David and Jason may sometimes come off a bit stiff, perhaps even cold, when it comes to how they manage their business, but that's the unfortunate aspect of print, edits, etc. In truth, working for them was the most challenging, educational, inspirational time of my career so far. They are the opposite of cold when it comes to their employee's ambition, ideas, talent and goals.

True: My experience leaving was equivalent to a professional football player saying to his coach, "Hey I think I want to try being a referee!" The player isn't on the team to be a referee. If he wants to be a referee, he needs to go to referee school and become a referee, not say, "I'm going to stay on the football team and be a referee, too." That's not how it works in football, or at 37signals.

False: I was not at all "rewarded" for my ambition by being shut down, put down, held down, or in any way shuttered by David or Jason. I was rewarded by them telling me I had too much ambition to be contained there, and by them supporting my new company as best they can.

False: I actually did my first job for 37signals in 2005, and started full-time in March 2007. So a little more than "just about 3 years."

The only bad thing I have to say about 37signals or their management or my leaving is that they stole my frog, which you can ask Jason Fried about.

As a woman who has also been a CSR and has almost exclusively worked around men in the computer tech industry I think you are making the best of the situation and I applaud your noble efforts to move upwards. I think there are many subtle details most others around here will never understand (or even try to).

But I still think you got a raw deal no matter how they spin it and I firmly believe in the back of their minds you were always flagged as "disposable" (replaceable) if necessary. I've seen it over and over again over the years.

Four day work week in the summer? Holy cow that sounds awesome.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact