The Management Team - Guest Post From Joel Spolsky

Today's guest blogger needs no introduction. Joel Spolsky one of the best bloggers out there. He also runs one of our portfolio companies, Stack. And his approach to management is unorthodox at times but amazingly effective. I asked him to tell us a little about how he does it. I think you'll enjoy this post, it's great advice on many levels, and its is also full of chuckles. I told you he's a great blogger.


Very few company founders start out with management experience, so they tend to make it up as they go along. Sometimes they try to reinvent management from first principles. More often than not, they manage their startups the way that they’ve seen management work on TV and in movies. I’ll bet more entrepreneurs model their behavior on Captain Picard from Star Trek than any nonfiction human.

Most TV management is of the “command and control” variety. The CEO makes a decision, and tells his lieutenants. They convey this important decision to the teams, who execute on the CEO’s decision. It’s top-down management. All authority and power and decisions flow from the top. How could it work any other way?

This system probably works very well when you are trying to organize a team of manual laborers with interchangeable skills to sweep up the ticker tape in the street after the Giants parade BECAUSE THE GIANTS WON THE SUPER BOWL IF YOU DID NOT NOTICE.

Command and Control probably worked great in the toothpaste factory where Charlie Bucket’s father screwed the little caps on tubes.

This system is also pretty obvious, so it’s what 90% of startup founders try first.

Seductively, it even works OK for a three person company.

This is dangerous because you don’t notice that it’s not going to scale. And when the company grows from 3 to 30, top-down management doesn’t work, because it doesn’t take advantage of everyone’s brains in the organization.

Turns out, it’s positively de-motivating to work for a company where your job is just to shut up and take orders. In tech startup land, we all understand instinctively that we have to hire super smart people, but we forget that we then have to organize the workforce so that those people can use their brains 24/7.

Thus, the upside-down pyramid. Stop thinking of the management team at the top of the organization. Start thinking of the software developers, the designers, the product managers, and the front line sales people as the top of the organization.

Joel mgmt

The “management team” isn’t the “decision making” team. It’s a support function. You may want to call them administration instead of management, which will keep them from getting too big for their britches.

Administrators aren’t supposed to make the hard decisions. They don’t know enough. All those super genius computer scientists that you had to recruit from MIT at great expense are supposed to make the hard decisions. That’s why you’re paying them. Administrators exist to move the furniture around so that the people at the top of the tree can make the hard decisions.

When two engineers get into an argument about whether to use one big Flash SSD drive or several small SSD drives, do you really think the CEO is going to know better than the two line engineers, who have just spent three days arguing and researching and testing?

Think about how a university department organizes itself. There are professors at various ranks, who pretty much just do whatever the heck they want. Then there’s a department chairperson who, more often than not, got suckered into the role. The chairperson of the department might call meetings and adjudicate who teaches what class, but she certainly doesn’t tell the other professors what research to do, or when to hold office hours, or what to write or think.

That’s the way it has to work in a knowledge organization. You don’t build a startup with one big gigantic brain on the top, and a bunch of lesser brains obeying orders down below. You try to get everyone to have a gigantic brain in their area, and you provide a minimum amount of administrative support to keep them humming along.

This is my view of management as administration—as a service corps that helps the talented individuals that build and sell products do their jobs better. Attempting to see management as the ultimate decision makers demotivates the smart people in the organization who, without the authority to do what they know is right, will grow frustrated and leave. And if this happens, you won’t notice it, but you’ll be left with a bunch of yes-men, who don’t particularly care (or know) how things should work, and the company will only have one brain – the CEO’s. See what I mean about “it doesn’t scale?”

And yes, you’re right, Steve Jobs didn’t manage this way. He was a dictatorial, autocratic asshole who ruled by fiat and fear. Maybe he made great products this way. But you? You are not Steve Jobs. You are not better at design than everyone in your company. You are not better at programming than every engineer in your company. You are not better at sales than every salesperson in the company.

It is not, as it turns out, necessary to be a micromanaging psychopath with narcissistic personality disorder (or even to pretend to be one) if you just hire smart people and give them real authority. The saddest thing about the Steve Jobs hagiography is all the young “incubator twerps” strutting around Mountain View deliberately cultivating their worst personality traits because they imagine that’s what made Steve Jobs a design genius. Cum hoc ergo propter hoc, young twerp. Maybe try wearing a black turtleneck too.

For every Steve Jobs, there are a thousand leaders who learned to hire smart people and let them build great things in a nurturing environment of empowerment and it was AWESOME. That doesn’t mean lowering your standards. It doesn’t mean letting people do bad work. It means hiring smart people who get things done—and then getting the hell out of the way.

#MBA Mondays

Comments (Archived):

  1. laurie kalmanson

    the upside down pyramid is really rightside up: awesome.also: reward good behavior, do not tolerate bad behavior.

    1. Dave Pinsen

      I worked at a bank years ago that used the upside down pyramid imagery too, with its customer service reps (which it called “bankers”) at the top of the inverted pyramid. Joel’s use of it here is more appropriate.

      1. laurie kalmanson

        interesting; were they really empowered or did it just look that way on the chart?

        1. Dave Pinsen

          They were only empowered for the most minor issues. The chart was basically blowing smoke.

          1. laurie kalmanson

            it makes somebody feel better about something when things like that happen, but why? oy.

          2. Roland Hopkins

            I’ve run a successful company for 50 years. I never tell people what to do. I ask them nicely – say please and thank you.  I have NO rules – just strong suggestions.. Each employee knows their job and is their own boss (or think they are. Every six months I ask them all to submit up to 10 new ideas about improving the  company. Do I get great new ideas? Try it with your company and find out.

          3. laurie kalmanson

            sounds like you are hiring great people and letting them do great things

          4. JLM

            Banking is the most over controlled and regulated industry in the US both internally and externally.If you turned banks loose to make real loans, take real risks on business loans, job creation would explode.If you can absolutely prove you do not need a loan, then they will make you one.

        2. Emily Merkle

          years ago I had the best experience with  my CS rep – “campaign manager” … 1) personality synch. 2) respect 3) clear constant communication 4) appreciation & guidance to learn beyond JD scope 5) some room to freestyle….awesomeType your comment here.

          1. laurie kalmanson

            that’s awesome; it should happen more often

          2. laurie kalmanson

            awesome when that happens

  2. jason wright

    In the brain economy (rather than brawn) management is by consent, and ego monsters need not apply.Some of the most rampant egos I’ve come across have been in academia, not manageable by themselves or others.

    1. Emily Merkle

      hmm ego is everywhereI think that company CEO can make a huge impact by – setting tone for the culture by the way he relates to everyone he comes in contact with.simply respect.

      1. jason wright

        Ego ‘maker’ (producer, achiever, et.c.) I go for.Unfortunately there is that other type of ego, the ‘monster’, the one that belongs in the pram with with the toys.  

        1. Emily Merkle

          No doubt. Ego that impedes forward progress or taints culture is not ok.All relative. If ego is petulance and contribution is there – Keep ego away from all humans and limit public representation if necessary :)Calculate ROI.

        2. Emily Merkle

          If you got it and you earned it – work it. That is just akin to a little end zone Jig. It is okay to celebrate a win.Confidence is amazing …

  3. Greg G

    Fred, these guest posts – along with yours – are consistently super quality. This paragraph is gold:This is my view of management as administration—as a service corps that helps the talented individuals that build and sell products do their jobs better. Attempting to see management as the ultimate decision makers demotivates the smart people in the organization who, without the authority to do what they know is right, will grow frustrated and leave. And if this happens, you won’t notice it, but you’ll be left with a bunch of yes-men, who don’t particularly care (or know) how things should work, and the company will only have one brain – the CEO’s. See what I mean about “it doesn’t scale?”Thanks much.

    1. fredwilson


  4. Sara Chipps

    This is a great post. I learned this the hard way via Girl Develop It. As we have grown to over a dozen organizers and teachers, I started noticing that every time I told people what to do and then waited for it to be completed I’d get something substandard to my requirements or nothing at all in return. When I did more enabling, things like detailed emails about existing systems, putting people in touch with someone that could help them, and making documentation about how tings should be done; things that I considered below the role of the “fierce and powerful leader,” things get done quicker, better, and more creatively.A good leader of developers does these things as well, makes sure the build software is up to date, ensures that developers have a good environment to work in, and does all the menial tasks that keep a good team humming. It’s not as fun as being in the trenches, but it helps your developers to grow and get things done faster. One thing I’ve been wrestling with the past few months is: by all accounts Steve Jobs was a d**k to the people that worked for him. I want to think that being a d**k can’t result in good products and making the world a better place. The legacy he leaves behind forces me to challenge that belief. I’d like to hear what people think about this (without turning this into a SJ debate). 

    1. fredwilson

      i think Joel’s point on Steve Jobs is there is always an exception that proves the rule

    2. PhilipSugar

       I think its a cause and effect thing and Joel laid it out well, saying you aren’t Jobs, Ellison, Gates or Bezos.So I think its an effect not a cause.  This happens in many instances.  I give Fred great credit for not becoming an asshole.Its really easy to do.  Everybody tells you how great you are, everybody wants to work for you, everybody wants your time, your money etc.  Its easy to turn into an asshole.  There is no governor on the system.  You can be an asshole and things go on their merry way.

      1. fredwilson

        i give credit to the gotham gal.

      2. Emily Merkle

        i try to mentor not manage.sometimes in early lean days you do not have funding for A or even C hires.working with what you can afford is toughsometimes you have to babysit intolerance of BS tends to be perceived as jerky :)Jobs expected as much if not more from himself as anyone elseif you know ability is there it can be frustrating. rising tide lifts all boatsfierce passion and clear expectations – no matter how un PC can work I suppose…nothing to fear if you are rocking 🙂

    3. ShanaC

      I guess the question is what is actually the point of being nice. I would say there is more evidence that being nice in terms of reinforcement is helpful. Lots of people aren’t nearly as well motivated by fear as they are by love

      1. JLM

        People are motivated by competence, authenticity, integrity, certainty (payroll checks clearing) and hard work.People are inspired by vision, energy, communication and personal example.

        1. JamesHRH

          @fredwilson:disqus  seriously, how good is this?JLM – tweet these!!

        2. Robert Thuston

          Drucker:  People are motivated by “responsibility”

      2. Andrew Hoydich

        My opinion in terms of love and fear, is that it’s not an either or debate. People are most inspired when there is an equal presence of both in the person they look to for guidance. Too much love and people’s egos become inflated and they start wandering off in their own direction, too much fear and you create a hostile environment that discourages creativity and kills morale. Finding a balance yields a leader who’s mere presence brings out the best in his/her followers. 

        1. Emily Merkle

          care enough to be a jerk / care enough to do cartwheelspassion 



  5. William Mougayar

    Very pragmatic advice across the board. Re: the 2 engineers investigating SSD drives, I would put them in a room and have them run through the pro/cons of each approach with me (as the CEO or VP, Eng) and I would make that decision for them. But I will let them understand my decision-making process so they can learn it for the next decision.  

    1. Stephen Starkey

      What would it be like to put them in a room and facilitate a conversation where they come to the correct solution themselves?  Then they can learn how to make good decisions together and you don’t have to be involved any more.The world would be a very interesting place indeed if managers of all stripes learned how to teach and facilitate instead of decide.  Then they could focus on collecting, holding and disseminating the vision and leave the nitty-gritty details to the folks whose jobs are suited for it.In Deming’s view, the manager should:”5. Be a coach and counsel, not a judge.”…I agree 100%!  

      1. William Mougayar

        Sure. Lead them to the best decision. Often, they will not see the business/financial implications which are almost always a factor in every technical decision.

        1. JamesHRH

          quite often problem solvers struggle with playing a single role in an executive team.the right answer, to them, is the broadest answer. They can struggle to provide a narrow answer that allows the CEO to make the broader decision.As the Rock says, ‘ know your role’.

    2. JLM

      Bingo!  Well played on an intergalactic level!A wise CEO would use this as an opportunity to allow both engineers to learn how to “negotiate” between themselves and to see how a decision is framed, informed, evaluated and decided.Layout the question — OK. what are trying to decide?Identify the competing alternatives.  Make sure there are no others.Discuss the merits and shortcomings of each alternative.  Whiteboard.Give each engineer an opportunity to respond to the other.Summarize the pros and cons of each competing alternative.Ask for a recommendation — you will surprise yourself here because they may have changed their minds during the discussion.Outline your initial decision.Make a final decision.Make damn sure that everyone knows what that decision is and that they have bought into the final decision.Make them “brief back” the decision.Ask them if they got a “fair” hearing.  Thank everybody.Executive make decisions even when they don’t know anything about it.  Clever executives allow themselves to become educated but they still decide stuff. 

      1. JamesHRH

        Totally agree – execs are not quite as useless as Joel paints them. It is a bit of the SW Eng high horse.Some of the themes are off kilter, but the content on creating a ‘lead via support’ culture & structure is bang on.A great article – circa 1998 I think, in, say Fortune – compared Jack Welch to a generic high tech CEO and came to the conclusion that Welch could run MS or Oracle or Fb in 2009:- make decision with insufficient market data (all the time)- make technical decisions (GE does build airplane components)- manage people issues in competitive markets (al the time)

        1. Robert C. Barth

          Let’s not forget Welch was a chemical engineer by both education and trade.

      2. William Mougayar

        I rest my case.

      3. Robert C. Barth

        The two engineers have much more knowledge on the subject in this scenario than a third-party “decision maker.” This is a tactical decision that should be left to the tacticians. If I were the CEO and they came to my office with this, I’d kick them out and tell them to figure it out, it’s what I hired them to do. Assuming I hired correctly, they would figure it out since they’re smart people who know how to make compromises without having to do dog & pony shows to the “boss.”

        1. William Mougayar

          Agreed, but the way this episode was described was as if the 2 engineers were going to dick it out instead of logically arrive at the right decision.

  6. JimHirshfield

    “You don’t build a startup with one big gigantic brain on the top, and a bunch of lesser brains obeying orders down below. You try to get everyone to have a gigantic brain in their area, and you provide a minimum amount of administrative support to keep them humming along.”Gigantic brains in the right places, umm hmm.

  7. Carl Rahn Griffith

    Wow.Well said.

  8. Jon Lim

    I think this is spot on.It can become quite frustrating to be a part of something where you want to put in your own voice, your own opinion, your own mark, only to be quashed by “the founder” who decides that they’re going to control anything and everything that comes out of your mouth.Let the front lines people do what they need to do, don’t pretend to know more than they would.

    1. Emily Merkle

      I would hope culture that would be perceived during interview process – candidate vetting co.

  9. Pascal-Emmanuel Gobry

    I’ve read everything on Joel’s blog and so I’m familiar with his theories on management and broadly agree with them. I hope he’ll read the comments and respond.But it leaves little room for the visionary-CEO role. Joel addresses this and says in essence “You’re not Steve Jobs.” He’s right!But here’s a trickier question: Joel’s beliefs come from his time at Microsoft in the heyday, when managers did view themselves as having a job “moving furniture out of the way”–here’s my question, though: was that how Bill Gates viewed his job? When he wrote memos like the (in)famous Internet Memo, that didn’t seem to be the way he viewed his job. More broadly, you don’t have to be a narcissistic egomaniac to be a CEO with a strong product vision. I’ve met Dennis Crowley a couple times and he seemed like a lovely guy, who also has very strong product vision. And conversely, as Steve Jobs himself has said, Apple couldn’t have survived if it hadn’t attracted great people, and that meant giving them tons of autonomy. So I guess my question is (and I suspect there’s no real answer): how do you combine a CEO with a strong product vision with the necessary autonomy that outstanding employees need? (Oh and btw: I try (and fail) to model every aspect of my personality on Jean-Luc Picard. 😉 )

    1. fredwilson

      founders with a strong product vision are not always barking orders around dennis doesn’t do that

      1. Pascal-Emmanuel Gobry

        I know, that’s my point.

    2. steveasfdfsdfsd

      These general advice posts can’t be directed at the Steve Jobs’ or even the Dennis Crowleys of the world. That’s a constant quibble with any startup advice post: “What about Steve Jobs? He didn’t do it that way.”Those people are edge cases, and Steve Jobs is the most extreme of edge cases. He was a textbook bad manager who succeeded because his vision happened to be what customers wanted. If it hadn’t, the company would have burned to the ground under his totalitarian leadership.If you’re an edge case and you’re destined to be hyper-successful, you’ll probably know it and be able to sift through this advice for what applies to you and what doesn’t.

      1. Pascal-Emmanuel Gobry

        I think Steve Jobs was actually an excellent manager–witness the dozens of amazing people who jumped off cliffs for him. It was a management style that even if you could imitate it, you shouldn’t, because it was based on cultism, but you can’t deny that it was effective.Also, my point is precisely to avoid the “What about Steve Jobs?” question by pointing out that there are dozens and dozens of founders/CEOs who have very strong product vision. Presumably this post is also adressed to them.

        1. steveasfdfsdfsd

          Getting people to break their backs for you is not always the mark of a great manager. Fear and desperation can motivate people to do that, and the accounts I’ve read about Jobs indicate that he used quite a bit of intimidation as a motivator.There are many, many studies that show that having employees that are fearful of you or afraid to make mistakes impacts an organization negatively, as does micromanagement. Jobs was guilty of all of the above.As Fred implies below, a great manager can make people happy to work toward her vision without making them afraid of the repercussions of failure.I’m sure there are many reasons for this, but I remember seeing a study about a year ago that showed the way employees were defecting in the software giants. Facebook was the only company gaining more employees than it was losing to defection, and you have to wonder why an Apple employee would go to Facebook. I’d imagine much of it would be the culture.

          1. William Mougayar

            Having employees clearly understand the ceo’s product vision is key in letting them loose to help realizing it. 

        2. JLM

          Well played!To ignore that fact that Apple is a cult and that Jobs was a cult leader is to betray a real lack of knowledge as to what happened and is happening there.Steve Jobs invented the most iconic physical movement in the history of technology — the finger swipe.Second only in the history of mankind to the extension of one’s middle digit as an expression of vexation.Jobs was a cult leader and Apple is a cult.

          1. Pascal-Emmanuel Gobry

            …I agree with that!

    3. laurie kalmanson

      i think it dependssometimes a founder stays in the center of the org as visionary, sometimes as leader, sometimes as ceothe genius of the post is that it scales, from the founders who can fit around a table to 25 people to 100 to 200, etc. it’s more about the “what happens when we need an org chart” phase.the founder isn’t necessarily the ceo at 200, but the vision has to be stronger than ever.

    4. Dave Pinsen

       Better him than Q.

      1. Pascal-Emmanuel Gobry

        Ha! Yes, hopefully!

        1. FAKE GRIMLOCK


          1. fredwilson

            i don’t delete comments, but i might have to change my rule of they show up here

    5. PhilipSugar

      I think there are two parts to vision.The first is seeing the whole picture like a chess-board and communicating it.  Its why you have to get out as a CEO and broadly understand other viewpoints.  If you are heads down blasting through a problem you have no vision other than tunnel vision.  Understand that you damn well better have tunnel vision to get a job done.  Nose to the grindstone, but somebody better have their head up seeing what’s around.The second part is having the vision to actually implement.  This is the vision on keeping everybody on point.  Not letting people go meandering in the field.  This is like when you tell your little leaguers to stop picking daisies and look at the ball.The vision I think everybody thinks a CEO has which I know I don’t is the “go up on the mount up high and come down with the commandments” that I think only the one above really has.  Anybody else that thinks they have it is drawing lines from points.

      1. Emily Merkle

        ideally the whole crew has some understanding of the vision 

    6. Emily Merkle

      i suspect a strong product visionary wants strong product peeps – the mission is top priority you can lead without stifling othersa good leader is not inflexible but rather is still motivated and curious to learn from those around him/herso i hope

  10. Dave Pinsen

    Props to TNG and the Giants in the same post. Nice. Someone actually wrote a book about management lessons from Captain Picard (see image of it below).

    1. fredwilson

      of course you found that book dave. nice!

    2. Donna Brewington White

      Who needs Google when we’ve got you!

  11. Kasi Viswanathan Agilandam

    I really liked the way where you have put the CTO … considering all other CXO’s come under the administrative roles. Nailed it.  Everyone in the Chief category can  act like facilitator/administrator but the CTO has to be authoritative and hands ON…

    1. William Mougayar

      But they must be leaders too. The post over-emphasized the support role of management. 

      1. K_Berger

        Agreed.  That’s why the inverted org chart doesn’t work for me.  It doesn’t appreciate the value of leadership.It is important to emphasize that a building without a strong foundation collapses, but turning it upside down isn’t it.

        1. William Mougayar

          Reality is that Management is 3 things: Leadership, Management & Support.

  12. steveasfdfsdfsd

    I agree with just about everything in this post, but I take issue with the example of a university department.Universities are some of the worst-managed organizations in the US. Ask a professor. They constantly struggle against the politics to get anything done. No advice post should ever mention universities as a good example.In fact, universities show you what happens when you put giant brains in charge of managing themselves. They go haywire and lack focus. That happens to be the point of a university in most situations (to let the researchers scour ever corner of their field), but it’s wrong for a corporation.Look at Google: they use the style you’re advocating and take it too far, and they’re one of the worst-organized, least-focused companies. They have a tremendous potential, yet all of their most promising products are languishing (e.g. Google Voice).Giant brains are good, but you have to know that person well enough to know to what degree they can manage themselves. Management has to be tailored to the team you’ve built.

    1. fredwilson

      yeah, i wasn’t moved by that example either

    2. JamesHRH

      Just referenced Dealers of Iightning below – the takeaway there was to manage ‘giant brain’ innovators through peer pressure. It clearly worked.Another overlooked work on innovation management is… .It agree with Dealers – a deadline that ends in a presentation to people that the Giant Brain fears or admires is the correct culture for breakthroughs.I disagree with Joel and agree with JLM – you need to grow problem solving engineer types in to people who see how good decisions are made (and feel the impact and accountability of making decisions on behalf of a group).

    3. ShanaC

      Maybe that’s why the humanities have stooped being a refuge.

  13. Michael Elling

    It’s about employee empowerment (and teamwork).  Not sure if pyramid (right side up or up side down is the right construct).  A matrix might be better as each employee then has a 3D view of their decisions and the impact on the rest of the organization (resource constraints/availability).  The matrix itself is defined by technology/product (supply) and customer/market (demand) parameters.And I’ll throw in another plug after a GREAT win Saturday morning; google Sir Alex Ferguson’s (SAF) track record and management approach that focuses on teamplay.  Offense becomes defense, defense becomes offense.  The same holds for good companies and employee teamplay.

  14. Kasi Viswanathan Agilandam

    I think people have forgotten about “Neutron Jack” considering the extreme behavior of Steve Jobs….15-years back Jack Welch was considered the ‘ruthless, notorious, arrogant’ CEO of all time….

  15. Markjaffe

    Well said!  Assuming you’ve hired smart motivated and experience self starters, I believe that empowering startup teams to excel requires that you give them rope and hope they don’t hang themselves, support and challenge their decisions, and make sure they know that it’s ok to make small mistakes, as a means of learning, as long as they never make the same mistake twice.

  16. awaldstein

    The inverted org is a keeper.Don’t be a jerk is a good theme that runs through all of these posts. But you don’t need to be an ass to lead nor a jerk to inspire. Nor a tyrant to make the final call. Consensus is not necessarily the same as a group decision.

    1. JamesHRH

      If you look at the inverted org, the obvious question to ask is:What role has the most pressure on it?;-)#whygreatCEOsdeservethebigbucks

      1. fredwilson

        yup. i watch that up front every day.

      2. Emily Merkle

        the perception should be “all”each role should be felt to be that crucial – that impactful

        1. JamesHRH

          That’s just incorrect.Accountability for the culture rests on very few shoulders. While heroic efforts can make modest impact in a poor environment, the fact is that there is only one member of the organization (without equivocation or dispute) that can (always and under every circumstance) be held accountable for the organization’s performance – the CEO.Very few people desire that level of pressure and most people perform poorly in that situation.If you are not the CEO (yet), the perception should be that the senior team has positioned you for success and you just need to do what you are good at doing – and we will all be successful.

          1. Emily Merkle

            I am well aware that I have had a VERY niche “onboarding” into this amazingly flexible and dynamic world we swim in… but I just have a very difficult time swallowing that Fred and his henchmen would allow such an exaggerated depiction of – perhaps and I am finding my assumptions are dead ass wrong about suburbua – a real ii that is a trick question – not impressed too obvious.:)i do not think people get my humor and i have brought my family of … to anger so – how foes one convery JOKE in your world! and also this was not a joke the smiley wa t indicate not seriously harsh … need commicate 101 pls

          2. JamesHRH

            Emily – not a henchman. not whacking you. Stating an opinion that I think is accurate, based on a fair number of trips (in startup land at least) out to sea.And, stating it because I think it is dangerous for you to believe what you are saying (all humour based in honesty).It is not exaggerated, my opinion.There is an old saying “you can not survive two layers of incompetence”. What it means is you can work around a bad boss, but you can’t work around that person and their bad boss as well.Who is responsible for there being two layers of bad bosses? Not a front line engineer, that’s for sure. So, it is unethical to let a front line engineer hold the perception that they should care as much as the CEO – they cannot have the same impact.The front line Engineer needs to care enough to do the required job (hopefully they care enough to do it awesomely, but doing it well is 10 out of 10).The CEO needs to care more. For the CEO, doing the job is never good enough – getting the result is the only thing that is good enough.So, you will cause yourself an unfair amount of anguish if you take this perception with you into a company run by an unprepared CEO.Not an attack – a heads up.

          3. Emily Merkle

            I appreciate the fact that you feel as strongly as you do in looking out for others 🙂 nice I do not know yo guys and I really have had some very unfortunate experiences that my views band philosophy are shaped by. ! I sure hope I am the exception!

  17. Ronald

    file://localhost/Flexfocus/Site/Creationcourt.pngIs not only to flip the structure. Knowledge management and success is through connecting people. Across several Businesslines and without hierarchy. Its about making everybody “CEO”. This impact the culture and management style. I believe Google is a better example how to do be successful with employees who make the difference. 

  18. Kinshuk

    For a good model of a good knowledge management organization, try a “monastery” instead of an university. So you have a few old ancient people who exist mainly to advise a passing king or such riff-raff. But in the main, all monks are supplied with food at regular times, and most of the day they sit in some corner with their favorite IDE and tools to play with. There is reasonable silence and no one disturbs them much – certainly not to ask “hey, how are you getting along with Koan Xyz ?”. In the evening all monks irrespective of arthritis get together to play some basketball (or table tennis). Sometimes they do some serious archery and learn skillsets from each other. There are many many similarities. No king or merchant could ever run a manastery, only a monk could. There isn’t much to play politics about when you are already playing God.

    1. Dave Pinsen

      “People tell a lot of tall tales about the Shaolin Temple”.

    2. tomwest

      There’s a Babylon 5 episode where they use monks to solve a particular computing problem. It works for precsiely the reason you identify – because they were a group of people used to solving mental problems after being given the goal from a ‘senior’ person 🙂

  19. csertoglu

    Great post. Thanks @fredwilson:twitter , @spolsky:twitter . I have always seen the CEO to have three functions: vision, recruiting, financial health. The post essentially deals with the last two.  The first is a tricky one; it involves being an evangelist and a face to the outside.  I do believe that the CEO should be the best person in the company in that role.

    1. fredwilson

      I agree with you on those three. I wrote a post about that. Vision is saying where we want to go, not where to turn at every intersection

      1. John Revay

        Read/listed to that post  – it went something like .. CEO needs to do three things well; Vision, Hire and Retain the right people, Making sure there is Cash in the bank.I think this post points Hiring and Retaining the right people.   If you spend the time to hire the right people up front – that will make sure you have the right team in place  to manage the company using the upside down pyramid.

        1. Anne Libby

          And while “boxes” and “org charts” raise up feelings of resistance in some people, whether the pyramid is inverted or not, part of the leader/manager’s mandate is to make sure that people understand the scope/boundaries of their roles.   (Good fences make good neighbors.) 

          1. JamesHRH

            Absolutely agree Anne.I worked with a group who felt that openness was paramount.They had a receptionist – never got why, but there you go – and I asked “when Senior Exec from MS comes next week, is she sitting in on the meeting?”That helped clear up the whole fence issue.

      2. PhilipSugar

        Yes.  And its also saying we need to get this stuff done that nobody likes working on because its important to the customer.

      3. Reddy_s

        yes Fred , I remember that post  because  It is  All  time TOP 10 list from Fred in my book.So much wisdom packed in that post , I posted it as brand new comment .In my view this CRISP thinking/expression  is what differentiates  you from all other  VCs and Bloggers.

    2. awaldstein

      Istanbul–great city, Lucky you.

    3. Emily Merkle

      should – but may not be best face universally to every audience.

  20. PrasannaKrishnamoorthy

    Typo in your intro “and *its is* also full of chuckles”Nice piece, misses perhaps that hiring needs to be strongly in tune with this, and not just get ‘smart people’. People who can work in this structure and deliver are not that common.

  21. K_Berger

    This is a good post (thanks Joel).  I would suggest heading over to his blog for anyone who has not already done that in the past.  Lots more great stuff over there.One thing missing, and which doesn’t quite work with the ‘upside-down’ org chart, is that management provides the vision and business direction that is often not understood at the ‘do work’ level.  So let the engineers fight out the details but when an impactful decision needs to be made, the person at the top should be involved, preferably offering insight and guiding everyone together to the best solution. The best leaders know the faults and strengths of the team and especially their own.  Then they make sure to draw upon the right one at the right time.  The image I have in mind is a conductor leading an orchestra.For the record, I don’t think Joel would disagree with this, just that it wasn’t covered here.

  22. David Semeria

    It’s true that Jobs (by his own admission) could be a real jerk, but he was also brilliant at stripping out unessential features and settings priorities.I agree it’s engineers who need to decide whether 5 small SSDs are better than one big one, but it’s up to management to decide whether the whole issue is a priority in the first place.Delegation implies accountability on both sides: in what is delegated, and in what is delivered.Jobs was also very good at this too, and it’s a real pity he was frequently such a dick – he really didn’t need to be.

    1. Aaron Klein

      Jobs would have listened to the arguments, translated the technology tradeoffs into consumer experience, and said “look, five SSDs is going to make the device a millimeter larger. So find a way to get better performance out of one SSD even if we have to buy Samsung to do it.”Not all companies get to be Apple just like not all CEOs can be Steve Jobs. 😉

      1. David Semeria

        No, Jobs would have asked the engineers why they were pissing about trying to shave a few microseconds off the server response time when the client-side experience was so “totally shit” no-one was going to use the product in the first place.

        1. Aaron Klein

          Ha! That is definitely what his jerk side would have said.The biography was incredibly revealing and more than a little sad.

          1. JamesHRH

            Suggested reading –… – if you are interested.Its not Jobsian, its a problem solving, intellectual personality trait / learning style / accepted culture.Literally, in the book, people are yelling at a presenter, telling the presenter that what they are doing will not work (and that they are crazy / stupid etc.) and then they go:”oh, I get it”.No apologies.

          2. Aaron Klein

            Looks awesome. Thanks!

          3. Emily Merkle

            been therethat is called Passionno time for politics respect can exist with debateif egos are that easily bruised we need to evaluate team

          4. PhilipSugar

             No his jerk side would say you two pieces of shit are fired on the spot along with your boss for having two pieced of shit working for her.

          5. Aaron Klein

            This is better than the biography…

        2. PhilipSugar

          That comment is right on.  That’s the vision part of being a leader.

          1. JamesHRH

            Vision = What Really Matters 101.

    2. Matt A. Myers

      Yeah — in one sense it likely took a toll on him for having to play that role and create animosity towards himself — or it just fed his ego, perhaps even created his ego and kept it going.The story of him being a Buddhist and going on his journey and coming back and ‘Apple’ being created was popular, though it’s possible this lead to a “spiritualized ego.” He certainly was brilliant though.If you’re a spiritual person and you also know business than you can identify people who’re running moreso off of a spiritual ego than anything else. All that really means is fear of survival is still driving them — they haven’t let that go yet and don’t trust they’ll be fine afterwards. A lot of people reach success from the energy anxiety ‘gives’ you, and then are afraid they won’t be able to continue their success without that driving force.Edit: Does anyone know if he practiced yoga regularly?

      1. JamesHRH

        Maselow’ s Hierarchy of Needs.

        1. Matt A. Myers

          100% right there James! 🙂

    3. Emily Merkle

      there is no everymandick is in the eye of the beholderI fired a client four times – am I a dick? 🙂

  23. andyswan

    Universities are a disgusting model to follow, unless your business model is funded by the federal government–without regard for performance.I don’t understand why everything always has to be flipped upside down.  What was good is now bad, weak is strong, sin is virtue…. and the underdog always wins…we get it.But reality is that leaders with vision are extremely important.   Leadership is extremely important.  And yes…so is hiring, trusting and empowering high-quality people that can “run with it” in a flat org…Steve Jobs wasn’t JUST a dictator.  He was a dictator that formed a flat organization of competing business units full of smart people that he very much trusted and admired.  He demanded the most out of them and they RESPECTED the hell out of him.Love stack, and I get the idea behind this (trust smart people)….just not a fan of diminishing the importance of leadership, vision and passion.Believe it or not…people still want to be inspired.

    1. markslater

      i agree – be a jobs basher all you like – i assume you have worked with him. I haven’t.i find that there is no substitute for straight up HARD WORK. issues solve themselves much more quickly when people are performing. 

    2. Aaron Klein

      You hit the nail on the head with what is wrong with this post. Joel is a smart guy and gets a lot of things right here. His overarching point is correct.But he clearly doesn’t manage Stack like a university. I speak from experience.In this university model he talks about, here’s how it really works. I AM NOT EXAGGERATING THIS AT ALL.You decide there is a problem that needs to be solved. You form a 27-member committee to solve the problem.Each interest group shows up at a meeting before the meeting to protest the level of representation they have on the 27-member committee. When you say “27 is enough” they write letters decrying the unfairness of only 27 people making a decision and declare the results invalid (in advance).The committee actually meets and the first order of business is to establish the rules of the road. All 27 people must agree for anything to go forward. You can vote “sideways” which means “I hate this, but to avoid being the outcast of the group, I will express my displeasure but allow deliberations to continue, for now.”Everyone brings their “interests” to the table and it is an unwritten rule that the greatest sin is not to meet every single interest at the table. The highest praise goes to the person who fashions the compromise and says “does this meet all of the interests?”The end decision is watered down to the lowest common denominator of mediocrity until all 27 people vote “yes” or “sideways.” Many test votes are taken until the pinnacle of mediocrity is finally reached.If the iPhone had been designed using this process, it would weigh seventeen pounds and the web browser would have been killed at the last second because “if we can’t have a web browser that views all sites equally, we just won’t have one at all.”

      1. andyswan

        Oh believe me, I know.  It’s awful.  I see this a lot in non-profits, especially the ineffective ones.  I believe people that sleep in the park use this strategy as well.And yes, I do agree that Joel is VERY smart and worth listening to.  I understand and agree with what I believe the intent of this post is:Don’t be a jerk, hire smart leaders and empower them to do great things.

        1. Aaron Klein

          And THAT is the essence of leadership, not administration.

          1. Donna Brewington White

            Does the term “the administration” ring a bell?  And we wonder why we have a leadership crisis in government.

          2. andyswan

            +1 donna, +1

        2. fredwilson

          non-profits and universities are certainly not the vanguard of management excellence!

          1. ShanaC

            Nor are hospitals. A pity, since all of these institutions could because force for major change. I wonder what happened.

          2. Ben Kamens

            This struck home, but challenge accepted. Currently spending every day trying to build a dev team in Joel’s upside-down pyramid image at a non-profit.

          3. thomasknoll

            Sounds like an opportunity…

          4. Donna Brewington White

            It is.

      2. jason wright

        The Security Council of the United Nations comes to mind, the grandest of committees where the irresistible force meets the immovable object…every single time.

        1. veti

          But the UNSC isn’t supposed to make decisions, that’s not its function.Its function is to give everyone who matters a forum to express their displeasure at what other countries are doing, without actually starting a war to make their point. And you’d have to admit, that’s worth doing in its own right.When the UNSC agrees that “yes, there should be a war”, that means the five permanent members *and* a majority of the whole have signed up to it. It’s a demanding level of consensus, by design.

          1. Aaron Klein

            The UN shouldn’t be declaring war anyway. That’s not its purpose.But it’s really hard to defend their mode of governance in any way, since countries like Cuba, China and Libya have been or are still voting members of the UN Human Rights Council.

          2. Rhazes

            And so is the United States, considering their human rights record in Vietnam, Iraq and Afghanistan for example.

          3. Aaron Klein


          4. jason wright

            Yes, that’s what a committee does, not make decisions. That’s why the idea of committee structures in business is a bad idea. Businesses need to make decisions. My reference to the UNSC was not intended as a precise fit here, just a topical reference to how a well known committee structure we’ve all heard of grinds it out.  

        2. JLM

          The UN is a fantastically expensive idea which has outlasted its usefulness and is completely irrelevant to the real world.Could be completely replaced with about $1MM in conferencing hardware.

          1. jason wright

            I agree.I think the UN would pass a resolution on you JLM if you tried it 🙂

          2. Aaron Klein

            I could get them a better deal on the conferencing hardware.

      3. mikenolan99

        Aaron,After 20 years as an Entrepreneur, I find myself working in a University trying to build partnerships, serve small businesses and foster an entrepreneurial spirit across campus.Your comments are right on point, and at the same time my spirit has not been broken.  I’ve seen our organization do some great things.  For the most part, people see a need to improve the system – it just takes some firm leadership to make it happen.

        1. Aaron Klein

          Best of luck, Mike! Tough mission.

      4. JamesHRH

        This happened – it was called the Newton.

        1. Aaron Klein

          LOL – true!

      5. ShanaC

        So why do you volunteer at a university again?

        1. Aaron Klein

          I was elected to the Board of Trustees. At some point, I may be released for good behavior.

          1. Donna Brewington White

            BTW, the college president search I mentioned a while back?  They hired someone with a strong business background!  And you’ll never believe it — one of his former roles was CIO.I believe that nonprofits — and especially academia — are much better off when successful business people are giving vital input, if not running the organizations.  I’m glad you are in the role that you are in.  But I do not envy you.

          2. Aaron Klein

            Wow, that is absolutely stunning if it was a public institution. Not quite so surprising if it was a private one. The private colleges seem to understand the value of diversity in thinking and experience.

          3. Donna Brewington White


      6. annmariastat

        I think universities are this way now and there are far more problems than I  could begin to state in a comment. However, I think the university model 30 years ago when I was a student was  very different, it did resemble much more what Joel was describing and much less the sad state of affairs as we see them now. (This is one reason I left academia years ago to found a company.)

        1. Aaron Klein

          I can totally believe that. Otherwise there’s no way all of their wonderful output could ever have happened.

      7. LE

        You have to watch the 60 Minutes segment on Duke that aired last night. The independent committee they put together to investigate a fraud said there was no fraud (but in the end there was).

        1. Aaron Klein

          Wow. Just wow.PS: I know I owe you an email back. Sorry for my delay.

          1. LE

            I figured it had to do with your laptop needing repair! I have many Macs. I travel with two (one spare) plus the one I make my wife take. If you weren’t way out there I would have lent you a cpu.

          2. Aaron Klein

            Smart. And yeah, I am running at reduced productivity right now, that’s for sure. Fortunately, most of my stuff is in the cloud. 🙂

          3. LE

            By the way in that Duke story lot’s to learn PR wise.a) Get in front of the story.b) Throw researcher under busc) Give bullshit about how you are doing a&b to prevent others from doing same instead of the real reason “d”d) Story goes cold because not much more to say about it. Nothing meaty left for other media outlets other than to talk about 60 minutes.Very unusual. No Pr person spoke that I remember. Just the guy who essentially made the mistake. No spin. “We fucked up” nothing more to see.

          4. JamesHRH

            LE, have you heard of the Cavello school of crisis messaging?

    3. Rohan

      Agree with you, I do..(Except for one small thing.. More probably feared Jobs than respected him.. just a guess)

      1. Emily Merkle

        they may not have respected SJ as a personality, but hard to imagine no props for results. It is not personal. 

    4. Chris

      Note leadership is not the same as management. Leaders will emerge from multiple areas of an organization. Of course your CEO needs to be a leader, but a system where the *only* leaders permitted to thrive are those prescribed by rigid hierarchy will be hard pressed to respond adequately for all the reasons Joel points out above.100% agree, we certainly need to be wary of blindly vacillating between dualist positions; however for this instance, archaic assembly line inspired “management” models do need to be reevaluated. 

    5. fredwilson

      i don’t think Joel said anything about not needing visionvision is telling the team where you want to gonot ordering them to turn right and left and right and left

      1. Luke Chamberlin

        “Administrators aren’t supposed to make the hard decisions. They don’t know enough.”Aren’t vision-setting decisions much harder to make than which SSD drive to use? I agree with your interpretation but I’m confused by some of the language Joel uses.

        1. JamesHRH

          This is the single weakest statement in the post.See JLM below.

      2. andyswan

        I agree with that. I agree with where I *think* Joel’s head was at when he wrote this…but disagreed with flipping everything upside-down and saying the CEO/COO are just there “to support” and should not be relied upon to “make decisions”….because they “don’t know enough”?Not sure if I read it too literally or if it was written that way. Maybe needs to draw a line between “big decisions” and “small decisions”…strategy and tactics…Good discussion in any event.

        1. Mark Essel

          The crucial elements of the CEO decisions are different than the specialized decision making Joe speaks to early in the post. I’m gonna dig into this one after hours, but enjoy the “trust your hires to make good decisions” vibe I’m getting.Bonus points for mentioning Picard, rewatching STNG now.

    6. Kasi Viswanathan Agilandam

      where did you get your internet and fiber optics from? business and innovation were always been in love… both of them were always been liars. Neither of them can live without each other … unfortunately they never accept.

      1. JamesHRH

        This is a deeply cynical and poisonous outlook.Great business product teams are aware that they are Yin/Yang in nature.Woz got it pretty early. Steve got it later.

        1. Kasi Viswanathan Agilandam

          Cynical yes i accept … Poisonous NO.”Ideas are worth dime a dozen … execution is everything” – MBAs”Marketing is required only for shit products” – Innovators.

          1. JamesHRH

            Kasi – count how many times the word vision or leadership is used in this post.Vision and leadership are about telling people where the future lies.That is marketing, at its highest level.It is not a mystery. It is a discipline.It requires an understanding of people, in a broad and complete way, not a narrow and compartmentalized way. Both your quotes are specious.

          2. Emily Merkle

            life is marketingembrace the horror

        2. Emily Merkle

          no duh 🙂 how do you think MIT Media lab keeps the lights on?

    7. JamieEi

      I’m curious why you are so “disgusted” by universities. I wonder where you think that the research behind the silicon, network infrastructure, software, etc. that allows us to have this conversation comes from.Also, you assert without proof that university research is inefficient. As a research Ph.D. my experience has been just the opposite. Federal grant money is awarded via an *extremely* competitive system that is much closer to a market than any system governing private research. If you believe that I am wrong I would challenge you to back up your claims with actual evidence.

        1. JamieEi

          Your data addresses the cost vs. value of an undergraduate education, whereas Joel’s comments are primarily about the independent research and publication activities of faculty. It seems like you are using very strong language about an issue that is important yet largely orthogonal the the current discussion.

          1. LE

            You said, “Joel’s comments are primarily about the independent research and publication activities of faculty. “What am I missing? What Joel said is below. Is there something else you are referring to?Think about how a university department organizes itself. There are professors at various ranks, who pretty much just do whatever the heck they want. Then there’s a department chairperson who, more often than not, got suckered into the role. The chairperson of the department might call meetings and adjudicate who teaches what class, but she certainly doesn’t tell the other professors what research to do, or when to hold office hours, or what to write or think.How does that make Joel’s comment “primarily about the independent research and publication activities of faculty”?

          2. JamieEi

            Hmm, I’m not sure what I’m missing either. All of the parts that he calls out as distinctive seem to be about research and publication:”just do whatever the heck they want.””she certainly doesn’t tell the other professors what research to do, or when to hold office hours, or what to write or think.”Even “when to hold office hours” is about the freedom to have office hours at a convenient time that does not interfere with the research and publication activities that bring home the bacon.What part of that do you take to be about the practice of undergraduate education?

        2. sigmaalgebra

          Andy, the NSF and NIH money to universities is supposed to be for research, research, and research.  It’s not for teaching or at all directly for education. Better research is supposed to lead to more important content in the teaching and a more valuable degree.NSF and NIH research grants are TOUGH to get, and on average it’s TOUGH to do better research than those grants fund.It’s true that much of even NSF/NIH funded research looks like generalized abstract nonsense, make-work, junk-think, busy-work, prof-scam nonsense. Still in nearly every field, if look back 20 years will find some really good results that look like they justify all the funding including for the nonsense. E.g., NSF funded a lot of work in genetics; likely so did NIH. Now we are starting to understand DNA. DNA drives some amazing chemistry. We like chemistry: That was the origin of the gasoline in the tanks of our cars and little parts of our economy such as energy. Well, there is a geneticist at Harvard who helped work out how to do some DNA engineering to get some cyanobacteria to pump out Diesel, ready for the engines. They claim that they can put up essentially solar cells with bugs, a few nutrients, some CO2 and some H2O and get off 25,000 gallons of Diesel per acre per year. So, about 3% of the US SW could pump out about 25 million barrels of Diesel a day for the US, into pipelines at less than $1 per gallon. Presto: Cut US oil prices by 2/3rds, achieve US energy independence, and heavily f’get about ‘oil wars’.For more we’re making progress on cancer.Results big enough for you?Maybe the Federal backed student loans are providing loose money for academic waste, but the research grant money, first cut, is spent relatively well.There is the issue of ‘overhead’; maybe 60% of a grant goes for ‘overhead’. One of my physics profs, with grant money from the USAF “to further the technology of the infrared”, money very well spent, went back to the university adfunks and got the 60% back. But usually the 60% helps fund the humanities departments, mow the grass, furnish the offices of the adfunks, etc. So, some of the 60% is wasted, if that’s your point about the waste.Another point about the grants was: At one point in my career in software, I was making 6 times what new Camaro cost. So, seening such things, the NSF set up a committee of economists to stop such a situation. So, they wrote into NSF grants that so many grad students had to be supported. The results were overwhelming …. So, yes, at times some of the research grant money gets used for things other than research.In a crucial sense, it’s even tough for the US to waste money on good research: It appears that take some good researchers, fund them, f’get about them, wait 20 years, and then pick up the gains, gains far beyond anything else in our economy. Examples: (1) For US aerospace, pour buckets of money into electronics companies south of San Francisco. Result: Silicon Valley. (2) Pour buckets of money into battlefield communications. Result: ARPAnet, NSFnet, and the Internet. (3) Pour buckets of money into basic research into biology. Results: Genetic engineering for energy and drugs. (4) Pour buckets of money into military navigation. Results: GPS. (5) Pour buckets of money into research into pure math. Results: A shelf of beautifully polished textbooks that are the foundation of a large fraction of our science and technology.

          1. andyswan

            I am talking about the gov-backed student loan and grant system

          2. Emily Merkle

            research IS teaching. as phd student – i learned more from the research dance than any class. research is fundamental to advancing science. research imprints budding scientists/teachers/practitioners with a respect for not just experimentation but the whole damn process. how to think. no one “tells” profs what to investigate, but in large part the work a dept. is pursuing is what attracts the right students you need who want to do the work you do too. and they do what they do as part of a life’s body of work. not a job. yup competitive – my frosh year was 8. research builds program cred. “Difficulty” of research is not necessarily correlated with the difficulty to get fed funding; there has to be a perceived need and a corresponding ability and opportunity to investigate. grants pay grad students, yeah. to teach – TA – and for research hours. not actual hours spent doing research, tho. averaged 10$-!5$ / appointment max “20hrs”/week. hour. then again … i was a soft humanities dud 🙂 

    8. JLM

      People want to be inspired.  Amen.

  24. testtest

    “I’ll bet more entrepreneurs model their behavior on Captain Picard from Star Trek than any nonfiction human.” hilarious!i’d be inclined to figure out all the different ways to run management and then combine the best parts to fit the situation. there tends to be a lot of prescription–about everything, really–yet clearly one can’t carbon copy structures and hey presto, it works.there’s also the factor of abstraction. not parsing all information, but a simplified model. with the SSD drive(s), are technicians taking into account cost-benefit, risk, future scale (based on strategic decisions)? Probably not. which isn’t to slight the technician. they’re paid, and trained, to see the world thorough the lens of tech.

  25. William Mougayar

    Management includes 3 things: Support, Management and Leadership. I think this post over-emphasized the Support role, but rightfully so. It’s often forgotten, but it also assumes that the employee will have the right initiative to ask for support.

  26. Richard

    Great Metrics…Great Management: What is measured is improved. (Peter Drucker)

    1. Emily Merkle

      numbers do not lie

  27. shished

    Thank you!

  28. Tom Labus

    And vigilance too.When you set the table for your team to sit down to you want to make sure everyone is one is on the same team and not pursuing some personal agenda. Tom Coughlin gets the Giants to buy into team with pretty good results.

  29. Brandon Burns

    i agree, partly. this is how you manage a specific type of person. there are other types.some people want more active management, to have a heavier hand telling them what they should be doing and how. this is not always bad; and when someone is junior, this is absolutely necessary. it’s called training, and it’s something that most businesses today don’t give a shit about anymore. i’ve seen myriads of talented people, with proof of their talent via a body of great work, go into a job and suck because they didn’t know how to get with the bandwagon and no one wanted to take the time to mold them into the butterfly that would succeed in that culture. and then they end up leaving out of frustration, or let go due to not being understood. people say we have a lack of good, smart talent. i think that’s horse shit. what we have is a lack of willingness/ability on the part of managers to spot a diamond in the rough, realize that it’s going to need polishing, and take the time to do the polishing. everyone wants mr. perfect to start being perfect right away. it doesn’t work like that, not even for mr. perfect.

    1. Emily Merkle

      easy to pass judgement based on preconceived notions/views from ivory towers…you never know the story unless you askpeople with integrity, passion, and heart fall into viper pitshappens hard to escapeso I hear

  30. redsquirrel

    Great post, Joel. It was a great, validating way to start my week. 🙂

  31. Jake McGraw

    I’ve been a software developer for 6 years, exclusively NYC based startups. I have directly experienced a couple of management failings:1. Way too engaged CEO, dictating software functionality as I’m building a product. The result is that nothing gets done in his absence and I feel like a tool. Since almost all decisions flow through him, I associate all of his personal and professional weaknesses with those of the company and probably leave way before giving the organization a chance.2. Disengaged CEO, puts people in charge who have no clue what they’re doing other than “don’t bother the CEO.” Watched an entire product development team leave because the CEO was phoning it in and the people he put in place to manage the thing sucked. I think this is worse than an overbearing CEO… I can understand the ego of a CEO wanting to be part of every little detail, but the idea that they’d take the word of some bizdev flukey over mine is insulting.3. The Jobsian CEO, designed the product in a vacuum, ignored design issues during construction and we launched a functional but poorly designed product. It was obvious that any recommendation I made was immediately canned, I should have been more insistent considering how much I had invested in the company.I’m certainly not blameless in all of these experiences, I could have given 110%, stuck it out and tried to make it work. It’s just, I literally have 15 job offers in my inbox at any one time, so why put up with any of this bullshit for longer than necessary to justify a 20% bump in salary at my next job. This is probably the core of what Spolsky is talking about:If you don’t lead well, inspire and get out the way, you lose the people that will genuinely make you and your company successful and you end up with a bunch of unmotivated schnooks.

    1. JamesHRH

      Now you know why VCs look really hard at the person who is going to be CEO (and why so many lame VCs default to ‘serial entrepreneur’ choices – its a great CYA move).And, you also now know why super competent BigCo CEOs don’t do startups: they are more aware than anyone just how rare they are in the market.Keep swinging and, before you take the next gig, get some advice on what to look for (telltale signs) of an immature CEO.

      1. LE

        “now know why super competent BigCo CEOs don’t do startups”I think there is also something dystonic to people in traditional white collar business to managing a group of people playing games and being goofy.  Prior to web 1.0 there weren’t many examples of success in business by people not fitting the mold. Ben & Jerry’s comes to mind and maybe there are a handful of others (not talking about the entertainment business).IBM’s dark suit and white starch shirt wasn’t relaxed until the 90’s.People running public companies who do videos of yoga poses and have pictures of cats on their head was, um, not something you normally would see in the business press or the real world.

        1. JamesHRH

          Good point.So much of long term high level success is understanding where you should be, environment-wise.

          1. Emily Merkle

            indeed.i break out in hives thinking about politics and office hoursi wonder if I should even be public facing sometimes :)fast moving lean and mean logic/mutual expectations/clear rules of engagement / no bs

    2. PhilipSugar

      Agree completely.  Its why you mature, and I think that is a great point of his post.When I first started out I thought, make people put their desks together, make them scrounge for equipment, who needs an onboarding process?At some point I figured out this was just plain stupid.  Make them as successful/comfortable as possible.  Let them wow you with their productivity.

  32. pitosalas

    This reminds me of a meeting years ago with a few AT&T people, some top executives and some lesser folks. As they went around the room introducing each other they would say things like “I am Joe, and I support Jane and Jack”, and then Jane would say, “I am Jane and I support Fred and Clara”. As they went around we tried to write down the org chart and figure out who was who and who was in charge.It was total confusion. Because AT&T had done some business social engineering and decided that it was politically correct for a boss to say that he supported his underlings – I guess saying that they “work for me” or “report to me”  was too icky.

    1. Emily Merkle

      icky? hah! the bs is bafflingi avoid titles

  33. Matt A. Myers

    Great overview. But where does the visionary fit in?Naturally I would think they should be at CEO level — overviewing everything, however you say administration isn’t decision-making. So this conflicts.Visionary at VP Product?Product needs to be apart of marketing and sales, and you need to have an understanding of how much work (engineering or otherwise; what resources are needed) to be able to prioritize everything.You say “Administrators aren’t supposed to make the hard decisions. They don’t know enough” and this is where I am getting caught up.Maybe I am confusing “hard decisions” with “important decisions”? All decisions are important of course, however anything that will dictate the direction of the product is most important to me.For the other positions I’ve realized it really comes down to finding people you trust who understand each important position’s responsibilities, at minimum, as much as you.I think there’s been talk here before, maybe just what I’ve read in commentary, regarding Visionary vs. CEO. A visionary doesn’t necessarily mean they will be a CEO, though they can learn.

    1. fredwilson

      visionary should be the CEO or maybe the VP Product or Founder/Chairman

      1. Matt A. Myers

        Guess it might require a bit of playing around to see what works based on where everyone else best fits in an organization

        1. Emily Merkle

          i agree and i am confused with the strong emphasis on titles in this chat .. 

  34. BillMcNeely

    This post describes exactly the reason why the Army, between 2004-2007, lost a lot of young talent during our country’s most critical period in Iraq.The senior leadership could not make the transition from the cumbersome procedual command and control method of fighting ( reliant on pages long orders and PowerPoints produced by staffs of over 30 yr old Majors and above  (think consultants) to the more shake and bake interactive, 5 minute  back of the envelope method (dependent on 21-28 yr old Sergeants,Lieutenants and Captains) you need to win a counterinsurgency fight.Although the business enviroment had changed ( think back to the switch from IBM Hardware to EDS software in the late 60’s early 70’s ) the tactic techniques and procedures and the metrics used to measure effectiveness of the aformentioned had not.So young talent in the thousands like me were frustrated.  My viewpoint was “these guys are frickin idiots and are going to get a lot of people killed” and left at the end of our first term instead of staying through the 20 year mark.A few leaders got it ( GEN Petraus, GEN McChrystal, GEN Odenerio ) most did not. GEN McChrystal had an awsome Ted Talk on the subject called Listen, Learn, then Lead  Here is the link

    1. fredwilson

      thanks for the link to that ted talk. i’ve added it to my boxee watch later queue

      1. BillMcNeely

        The Atlantic ran a piece by Tim Kane entitled “Why Our Best Officers Are Leaving”The lead in was this:”Why are so many of the most talented officers now abandoning military life for the private sector? An exclusive survey of West Point graduates shows that it’s not just money. Increasingly, the military is creating a command structure that rewards conformism and ignores merit. As a result, it’s losing its vaunted ability to cultivate entrepreneurs in uniform.”Here is the link to the rest of the story:

        1. JLM

          Same articles were written in the VN era and probably since GW was C in C.

          1. BillMcNeely

              I agree that more officers from all branches should attend Ranger and Airborne School. Those schools teach spur of the moment problem solving techniques that come in handy away from the flag pole. Chad Storlie at does a great job demonstrating how these techniques can be converted to business success. He has written two books Combat Leader to Corporate Leader… and Battlefield to Business Success…If I may, I would like to contact you privately to discuss career options. Should I contact you via LinkedIn? 

          2. JLM


    2. JLM

      I agree with you more than you agree with yourself.Generals have a tendency to apply the lessons of and to try to perfect the last war in which they fought — “their” war.It is worth noting that Petraeus, McChrystal or Odenerio were not marked by Viet Nam.  They were too young to have served.  They saw their first combat as very senior officers — P saw his first action as a Division Commander.The type of warfare of the future will be more along the lines of the special operations, actionable battlefield intelligence and counterinsurgency of which Petreaus wrote when he supervised the re-write of the Army’s counterinsurgency manual FM-34.2 which had been untouched for over 20 years.Rangers in particular represent that bridge between conventional warfare and special operations warfare.  The fact that their modus operandi is the patrol — raids, ambushes, recon — makes them the most adaptable and quick hitting such force available.In my day, every Regular officer of any combat branch had to go to Airborne and Ranger schools.  That was a damn good idea.BTW, I have been hiring vets who are trade school grads w/ MBAs.  A strong source of talent.

      1. JamesHRH

        It makes total sense that the future of the military is fewer people, more specialized.I like the idea that, in 2025, a potential tinpot despot thinks to himself:”maybe I don’t want to enslave my fellow countrymen, pillage their treasury and flaunt my emotional shortcomings on the international stage………that has not worked out very well for the last 6 or 7 guys. The Seals & Rangers make the outcome far too certain.”

        1. sigmaalgebra

          No, no, no you have to understand:  It’s not the Seals, etc.  No, no!  Certainly not! Instead, it’s these backward countries.  When they start messing with explosives, things can go “boom”.  Just suddenly “boom”.  No reason.  No cause. Just “boom”. Even just ordinary cars on the street can go “boom” for no reason. And sick, people can get sick. Suddenly. No reason. And not the Seals, certainly not! Did I mention, they are backward countries? That’s just the way it is in backward countries!Yes, there was a B-2 flying overhead at 60,000 feet near midnight, but that had nothing to do with it either. It’s “backward spontaneous boom” — that was the cause!

        2. Emily Merkle


    3. sigmaalgebra

      Any VP candidates here?

      1. BillMcNeely

        I am interested to learn about the opportunity. [email protected]

  35. PhilipSugar

    I really think its great if you view one of managements roles as figuring out how to remove all productivity obstacles for the producers.  Totally agree.Another role and I’m stealing this directly from Joel (its somewhere on his blog) is to make sure the tough nitty gritty stuff gets done.  He correctly predicted that Linux would never go consumer, because given the fact that everybody got to decide what they wanted to do, nobody would go and build the crappy things like printer drivers.That’s where you lose me on the University.  I’ve pulled stuff out of the universities and they define work on what is fun and don’t do the nitty gritty, like figuring out this bug in Disquis where if you use Firefox the comment bar doesn’t scroll and where returns don’t work on Explorer so I have to edit :-)Also the tenure model absolutely sucks.  As I’ve wrtiten before many people work to a level above the bottom and if you aren’t kicking some ass when its needed you will not have a top performing organization.I also agree about not being an asshole but I worry that Jobs, Gates, Ellison, and Bezos are all renouned assholes.  I think you wrote that well in saying that unless you are one of them you can’t be…so that settles my cause and effect.  Being an asshole is an effect of growing a huge company not the cause.

    1. fredwilson

      yeah, the university part is getting savaged in this thread

      1. PhilipSugar

        Yes, but its a good discussion.  Universities really can be brutal.I have a brother that when he came up for tenure got fired for politics.  That was good because he got all his IP.They had to rehire and tenure him on the spot because he was the main feature on the Universities Web page.  He won the GM Robotic contest with inner city Hispanic kids, had his students build a baseball catching robot that was featured in the NYTimes, and was featured National Geographic, ABC Nightly News and was considered a favorite professor.  Fired for politics.

        1. LE

          Sounds like he should be an entrepreneur actually. That tenure is golden handcuffs.

        2. Emily Merkle

          Universities are Crucial.

    2. Emily Merkle

      respectfully disagreediff between not suffering fools and being as assholelots of superassholes out therei will not name them 🙂

  36. reece

    awesome post, Joelit is hard to execute as a founder, especially when the founder is the one who holds the product vision and must articulate that constantly, but if the communication is good, then the ‘workers’ are able to execute toward that vision effectively

    1. reece

      follow up thought: i do believe though, that the leaders of any organization should still be executing themselves. CTO’s should still write code. CEO’s should still do sales etc. it’s important for them to understand, as best as they possibly can, what the people on the ‘front lines’ are doing and continue to exercise theirs kills in those areas$.02 from someone who’s still learning to scale 😉

      1. Emily Merkle

        absolutely – CEO is Chief of Sales every day

  37. Justin

    I’ve been reading “Quiet: The Power of Introverts in a World That Can’t Stop Talking” and an interested thing discussed is the difference between Introverts and Extroverts as leaders/bosses. They see performance differences in teams based upon whether the leader is an introvert or extrovert and the motivation of the team members. They find that Extrovert leaders and their teams tend to be more productive (particularly with unmotivated team members) except against Introverted leaders with motivated team members.Extroverts tend to be the command and conquer war-time leaders while introverts tend to be listening, encouraging, and supporting peace-time leaders. When team members are motivated and care enough to improve the organization/team enough to suggest good ideas it requires a leader that will listen and incubate those great ideas. By encouraging, supporting, and placing the responsibility of implementing the idea to the person who suggested it you create ownership, pride, and additional motivation.The organizations today that look for and hire bright, highly motivated people and then have introverted leaders that nurture and support brilliant ideas have a competitive advantage. They will not just survive like many operating in the old “command and conquer” mode, they will thrive as they will be leaders in innovation and implementation; increasing their own knowledge and understanding of what works and doesn’t work faster and more efficiently.

    1. JLM

      Do a little Myers-Briggs Type Indicator study for some additional illumination.All of us have a “comfort zone” either introverted or extroverted — a preferred style — but we also possess the same attributes of the other style.  We may just prefer a particular style.As you age, you become more adept at changing styles based upon the situation.

      1. Donna Brewington White

        In an interview, after asking someone to describe their extroversion/introversion in percentages (e.g., 65% extrovert/35% introvert since no one is 100% one or the other), it is then interesting to ask what the percentages were at age 12.  Indicates what is learned behavior and what is natural.

      2. Emily Merkle

        or a DISC

  38. Luke Chamberlin

    Joel on Software is one of my favorite blogs of all time but I’m scratching my head at this post.I would be interested to know if the management philosophy of Fog Creek (software consultancy) works in the same way as Stack Exchange (web product), which I see as two different kinds of organizations.From my own experience, consulting firms or firms that work with clients on a project-by-project basis tend to work better with the bottom-up distributed decision making system described in this post.But companies that produce a product with their own brand attached, that is to say most startups, need that strong vision and a CEO who is making those decisions.This goes back to FAKEGRIMLOCK’s ‘MVP’ minimum viable personality post. A product’s personality is created by one person or a small team with the same vision.Personality is not created by committee with “admin support” from the leadership team, and personality is certainly not created by the VP of Engineering and a bunch of smart MIT grads.My old boss used to say that vision is a leaky bucket, and that her job as CEO was to keep that bucket full. This means you have to pour vision into the bucket every single day. Maybe we’re using different words but I would not call that admin support.PS Thank you for teaching me the word “hagiography”

    1. Reddy_s

      ” vision is a leaky bucket, and that her job as CEO was to keep that bucket full. This means you have to pour vision into the bucket every single day. “I like the analogy . I only add ”  ONE of the CEO main job is to keep Vision bucket full” , meaning CEO has to make many Critical decisions everyday  in addition to Vision.

  39. Kid Croesus

    Glad to see I am not the only one who disagrees, although I am a big fan of Joel’s writing.This might work for software development — if every role needs A+ people who are in high demand, and you keeping those people happy in their jobs is your number one requirement for success… but it certainly seems counterintuitive for a lot of other industries, ( not just street sweepers and toothpaste cappers.)If you were running, say, a services firm (law, accounting, architecture, construction) on a project basis —  profits are generally made by having highly skilled people build processes that break up tasks into manageable parts, and then moving the simpler  tasks down to cheaper folks or automating them. Ditto manufacturing.  This is, you may recall from high school, pretty much the narrative  of the industrial revolutionTurner and Ford who figured out that you didn’t need to have highly skilled engineers (like the ones at, say, Rolls Royce) to build a car. On an assembly line–you can get relatively low skilled workers to do a single task well.  Creating efficiencies through process — which creates the same or even greater margins as you increase capacity — this what they used to mean when they said “the business is scalable.”   Joel may simply be in the business of making Rolls Royce’s…but most of us are not.

    1. ShanaC

      All of those fields are changing. Doc review goes off to India or is done by computers. Toothpaste caps are screwed on by robots.Even the most bespoke of these businesses have become less bespoke and more about the one on one experience. Much harder to scale unless you build rolls Royce

  40. Jess Bachman

    I love the university analogy, so when do we start giving employees tenure!?

    1. Emily Merkle

      i really do not see tenure as a big phd program professors and researchers – they pretty much all do both in my experience – are motivated by big things. tenure is a recognition of a level of achievement, perseverance, diligence, service, etc. not big money to be had either.

  41. Alistair Mackenzie

    Best book on management which covers the topics here is ‘Up the Organisation’ by R Townsend, recently reprinted.He covers all this, basically your management style probably mirrors your beliefs about people.The other point is that even Steve Jobs changed, SJII on his return to Apple was a much better manager than in his first incarnation at Apple.  He wanted the best possible product and ultimately that’s what managers should want!

    1. Emily Merkle

      stay hungry stay foolish he walked the walk

  42. Nick Grossman

    This is known as “servant leadership”…”Servant leadership is a philosophy and practice of leadership, coined and defined by Robert K. Greenleaf (1904–1990) and supported by many other leadership and management writers. Servant-leaders achieve results for their organizations by giving priority attention to the needs of their colleagues and those they serve. Servant-leaders are often seen as humble stewards of their organization’s resources: human, financial and physical.”A quick read that gets at this (which is kind of a cheesy parable, but it stuck with me) is James C. Hunter’s The Servant: http://www.barnesandnoble.c…

  43. TWAndrews

    To what size company do you think the “management as facilitator” scales?  It seems to me that once you get to be a big enough organization, if everybody is empowered to do what they thing is right, you end up with a bunch of disconnected teams who may all be doing cool things, but who don’t have any idea what most of the other teams are doing, and maybe don’t really care.I think that one of the things that can excite people about being part of an organization is feeling like you’re part of a group of people pursuing a mission, and I’m not sure that loosely held-together chaos probably conveys that feeling, no matter how much freedom the best people have.

    1. Emily Merkle

      how about management as 1) trust but verify – procedural spot checker / 2) admin 3) efficiency chief and opportunity spotter – finding ways to work smarter/better like eliminate dups, identify & coordinate interdept. efforts 4) education coordinator

  44. MartinEdic

    I’m a bit sorry this turned into a Steve Jobs issue as he is an extremely unusual outlier. I agree to some extent with Joel except that those giant brain developers, etc. are typically really good in a very narrow range of business activities. The C team should provide support in the sense that they lend big picture perspective to the decisions they are asked to make by their team leaders. Lead the leaders by applying a broader, longer term perspective.As for Jobs, he surrounded himself with extremely talented and capable people who didn’t mind being held to insanely high standards. You don’t hear his immediate day to day team making these complaints of autocracy. No Ives or Cooks whining about mercurial Steve.

    1. Emily Merkle

      there will always be sour grapes..outliers should not be discarded for not being shiny happy people holding hands…stories…

  45. andyidsinga

    what an awesome treat for the avc community – thanks Fred ( and of course thanks Joel ).Lessons learned : Hagiography and twerp stuff 😉

  46. RacerRick

     “Management team as support”.  I like that.But, Captain Picard? Joel needs to update his references. How about William Adama.

    1. andyidsinga

      and he should have used The Doctor as an example of the “right” management style to adopt 🙂

  47. Tim Metzner

    Great post!Especially love the image and the “Mysterious Marketing Stuff” department he includes. 🙂

  48. matthughes

    RE: JobsI was with friends from Brazil recently in Las Vegas and they commented that America is such an amazing place – “just look at this city”.True, Las Vegas is spectacular and is in fact an American city.But it’s a wildly sensational oddity – there is no other city in America even remotely similar.That is Jobs. Apple’s unprecedented success notwithstanding, is a case study in of itself and not necessarily a model by which to judge or compare other companies.

    1. PhilipSugar

      I think the point always is that you can have cities like LasVegas and people like Steve Jobs in the U.S.Its unimaginable to the rest of the world.  That is what makes the U.S. great.  There are people that build great wealth doing things that nobody else even imagines.

      1. matthughes

        I could not agree more Philip.I probably didn’t articulate that point very well. 

    2. JamesHRH

      I disagree, Matt, in this sense.You can see how Steve fit his success, just like you can see how Vegas fits its existence.The lesson is one of alignment.Both Jobs & Vegas have made reasonably large mistakes (some details, do, in fact, not matter THAT much & hedonism is hedonism, don’t kid yourself) that were born out of a lack of perspective.Finding the perspective that fits you and your opportunity – and then navigating based upon it. I think you can take that from both these examples.I am preternaturally opposed to singularity as a rationale for success!!

      1. matthughes

        I agree on your point about perspective – well said.

      2. Emily Merkle

        Steve did not “fit’ his success whatever that means…his magnitude of impact is not given to rationalization or navigation or plotting. Right place  / right time / passion and just a drive that defies wordshigher calling?great motivator – manipulator – distortion field creator /  what have swan done good

        1. JamesHRH

          Emily – you’ve been busy!Here is what fitting your success means:1) many of the first movers in major tech trends share a similar personality. They are oriented to open systems, as they crave factual inputs. They are flat – relational hierarchy does not appeal to them. They can simplify masses of data (that would overwhelm most people) in to clear, high value observations (Gates – SW most valuable component of PC, Page – index most valuable component of internet, Zuckerberg – ‘friends’ most valuable component of online connections). They are gifted at understanding “what is”.2) last movers in a major consumer trend, while rarer in tech but more common in other industries, are top down visionaries who utilize personal experience to create higher value consumer experiences. They tend to be very structured and visual in nature, relying on iteration and insight over instinct and volume of data. They are masters at seeing “what should be”.First movers make out like bandits, seeing around the corner and cornering the market before others can appreciate the situation.Last movers drop out of the sky, with an experience that consumers did not even realize they wanted – until they experienced it.If you really don’t believe that 2005 Steve Jobs fit the opportunity, then why was 1985 Steve Jobs thrown out of Apple and why was 1995 Steve Jobs on the brink of bankruptcy?

          1. Emily Merkle

            oh busy! shush this is my vocation or some amazing coalescence of randomness that is so fun. finally. worth it and could not have been any other way. opportunity in the sense of Jobs’ travails…that is not pure and simple opportunist fit. it is stubbornness and will and passions and curiously and finally knowing what you know -and knowing what you don’ as always – and just … I do not know what you mean in expressing temporal “movers” .p[ls. explain. I was stealing my dad’s pocket change to buy the sugary I craved at 6. I have

          2. Emily Merkle

            a very very active interal life and swimming with petty lazy jerka just farces

          3. Emily Merkle

            sorry type too fast. with no brand no tech no press no ID zip you really learnt how to demonstrate value in as undeniable a way as possible 🙂 and ever evolving game

          4. Emily Merkle

            but in my experience the first an last are not mutualy exclusive and i hardly see cornering market as admirale of rme unless alignment with value ..can amy one answer some questions I have had for eons re: GOOG?

          5. Emily Merkle

            but in my experience the first an last are not mutualy exclusive and i hardly see cornering market as admirale of rme unless alignment with value ..can amy one answer some questions I have had for eons re: GOOG?

    3. fredwilson

      vegas is awful. it’s disneyland for adults. it sickens me.

      1. matthughes

        Totally agree – it’s awful.The very poor analogy I used here only got worse when my comment was not articulated well.

        1. Patrick Foley

          I liked the analogy.

      2. Joe Timlin

        Maybe so. But at least it’s well differentiated! It certainly appeals to those it appeals to.  

  49. Pointsnfigures

    Totally agree. Bottom of the organization will be closer to the market than the top. Same goes for government which is becoming more and more centrally controlled.

    1. Office Ant

      For some domains (tech for instance), the bottom may be closer to the market (I assume you mean in terms of knowledge and feedback etc), but for many others, that is not the case at all, and you need leaders with vision and deep understanding of the domain. I mean, are you expecting that someone who, say, writes software all days has an in-depth understanding of how to sell products for, say, health care of education?

  50. Nico101rsa


  51. Bob Warfield

    Lots of good comments here.  I too had a fairly visceral reaction to Joel’s view of management.  There is a lot more art there than simply inverting the pyramid, and there are important times when the pyramid, in the form of process, is exactly the right tool for the job.More on my blog:http://smoothspan.wordpress…Cheers,BW

    1. stacy murray

      A nice way to sum it up Bob. And “art” is very personal too.You can’t reduce good art to a procedural diagram. However it might be instructional to show how one person does it. But then to generalize, as in this is the way to scale, is beyond naive.Management is the ART of getting things done through other people. And how do you do that? Anyway YOU can. We all bring different attributes to the organization that combine into a culture. Good managers can “learn” rapidly and instinctively how to get things done within that culture to reach its goals. It’s a very personal skill. It’s art. 

      1. Emily Merkle


  52. Carl Youngblood

    Great thoughts, but I think “hagiography” is a little harsh. The book was actually quite critical of Steve.

    1. Emily Merkle

      i cried my eyes out 🙂 what is wrong with criticism if it is honest and not mean-spirited? 

      1. Carl Youngblood

        I liked Joel’s post, and actually hagiography is a great word that I try to use whenever possible, but it’s simply not accurate in this case. The book was far from laudatory. It basically made Steve Jobs look like a complete jerk with an uncanny design sense. Which seems like a fairly accurate assessment to me.

  53. Kasi Viswanathan Agilandam

    University is where innovation happens …that is what i suppose he meant.I don’t think anyone want to dispute that innovation does happen in universities.Then …Science is struck.Engineering which followed the science is in shit.Finance which followed Engineering is in deep shit.People who followed are in ….

    1. PhilipSugar

       BS = BullShitMS = MoreShitPhD = PiledHigherandDeeper in Shit:-)

      1. Kasi Viswanathan Agilandam

        That is the coolest definitions i heard of degrees. Nice one. 

  54. Jayanth Kashyap

    Loved this post!!

  55. Diego Mariño

    130 comments before this, and no one hasn’t mentioned Henry Mintzberg (… and his studies about managing organizations :/I truly recommend reading “The structuring of organizations” for those interested in knowing more about this topic. 

  56. JamieEi

    If individual engineers and salespeople are in charge, how do you make decisions that affect the whole organization? For example, the Giants, WHO ARE THE LUCKIEST SUPERBOWL WINNER IN HISTORY, may have a bunch of smart offensive coaches, but they still have to pick a play to run. I’m concerned that a purely Egalitarian Approach Generally, Literally Equals Stagnation.

    1. Emily Merkle

      everything affects the whole org to varying degrees

  57. Lee Blaylock

    My experience is if you hire great people, treat them like volunteers (just don’t pay them as such!) b/c the best people can work anywhere they want.  Volunteers move mountains when they are passionate about something.Share your vision and timeline with them, get their input, thus buy in, and when possible, help them form the team so they’re working with people who they want to work with.  Now make sure you remove all barriers that come their way – then give them ALL the credit when you are successful and take ALL the blame if it screws up.That’s the way I’ve found the best leaders to instill fire into the troops and do things that folks earlier thought impossible, plus you will engender a loyalty to these high performing people to take the next hill together.

    1. Emily Merkle

      I tend to take a different approach. no shame in things not working out – certainly no “blame”. i work with clients/partners/etc. sometimes – many times – things break/etc and you just take one for the team. no biggie. if your team is not “punished” for failing in good faith – then they are empowered to keep taking those risks that = great things at times. fall down check for blood get up learn keep going. I want a team that is truly in it together. and realistic.

  58. JLM

    Great read, great rant, lots of passion.  Thanks.There is a tendency for folks today to look for an easy answer to difficult challenges and to fail to recognize that sometimes the real answer is just hardwork.  Getting outside your comfort zone rather than abdicating your responsibility.There is no revealed truth in the notion that an organization looks to its leadership and management to provide it with the means to accomplish its mission — the upside down pyramid.Even the most autocratic organizations imaginable — the Army, as an example — operate on the principle that the CEO (green, shavetail 2nd Lt Platoon Leader) must ensure there are enough bullets, beans and bandages to get the job done.It is a bit of a cop out to suggest that C level management should not be involved in solving technical questions.  It’s hard and you will have to sweat the details but you still have to make decisions.  You have to evaluate and educate.  It is not easy but it must be done.  Executives have to make decisions.Knowing how to frame and make a decision is the critical skill of management.  That does not mean you have to know how to do something — you just have to pick through the alternatives and evaluate their relative merits.How these decisions are made is where experience provides guidance.Look to a guy like Eisenhower to see how he was able to forge coalitions amongst incredibly powerful egos whether allocating air power between day time and night time long range bombing approaches, rebuilding Columbia University, forging the first NATO alliance or being President and getting us out of Korea (warfighting being his core competency, he knew that was bad terrain for a modern war).And when it came down to making the final D Day decision as to whether to go or not wherein the inputs relative to weather were either nonexistent or inconclusive, the guy just made the decision.

  59. Matthew Moore

    Ug, one of my least favorite blog posts.  I get his main point, that new startups shouldn’t have a CEO who is dictatorial for the sake of being dictatorial.  However, most startups have the initial problem of choosing, building, and then marketing the right product.  Poor (or great) management doesn’t even matter until you nail that down.  Once you do nail that down correctly, you have to organize the management structure around the abilities of each individual of your team.  

  60. JLM

    A CEO’s or a company’s vision must be painted in color for all to see because they cannot see it in black and white.  Sometimes they cannot even see it.  Your job is to be their eyes.I hate the way this sounds — but in the tiny space in your head that is really you, most people want to be lead.  There are only about 0.000001% of folks who really want to be an entrepreneur.  Sorry.  Don’t feel bad personally because a whole lot of the folks who can be entrepreneurs are on this blog.Most people want to be able to follow a vision that is good, one they believe in and one they can participate in and one which adds to their own self perceived self image.This is why the Marines never have any trouble recruiting — their basic pitch is >>> you’re not good enough to be a Marine.  To which the recruit replies — “I’ll show you.”Don’t recruit someone by saying “this is a great company” — say, “you can do great work here, we’ve been looking for someone like you”.  Feed their ego.The choices in leadership/management are not being an incompetent soft pushover or being a prick.Be genuine and authentic — “hmmm, I don’t know, help me with that” but be prepared to make decisions and make them stick.Be a guileful, thoughtful leader who gets the best out of their subordinates — and yes, they are subordinates cause you are the boss — by coaxing it out thoughtfully and with just the slyest application of the smallest challenge.”Hmmm, do you really think you can do that?”Life is grand, don’t be afraid to take a few risks.  There are no “one size fits all” formulas.

    1. JamesHRH

      Said this above JLM and totally agree – most people want to be part of something great.Very, very few want to do what it takes to create or lead something great.It is better to know who you are and find the right part for you in something great, than to do things that go against your nature.Very well played – a hard truth but a truth none the less.

      1. jason marshall

        I’m reminded of a now famous line from Jamie Zawinski’s resignation letter from Netscape: And there’s another factor involved, which is that you can divide our industry into two kinds of people: those who want to go work for a company to make it successful, and those who want to go work for a successful company. Netscape’s early success and rapid growth caused us to stop getting the former and start getting the latter.

    2. William Mougayar

      Words of wisdom: “Don’t recruit someone by saying “this is a great company” — say, “you can do great work here, we’ve been looking for someone like you”. I’m a never happy manager…I always think that my employees can do more and will step up to more. And when they do, I quickly step-in and recognize them. I want my employees to compete with themselves, to keep pushing themselves to higher achievements, and my job is to give them the opportunity to do so, to be all they can be. 

      1. JLM

        The curse of the true entrepreneur is not absence of happiness or joy — it is contentment.  One is never, ever really content w/ any performance.There is always a better performance we all want.

        1. William Mougayar

          Absolutely…The pursuit of better performance is an obsessive thing. 

        2. LE

          “it is contentment.”From the perspective of managing employees have you ever (personally) found it a good idea to move around someone who gets to settled in their position (even if they like it and are doing a good job). Not a promotion but a lateral move?

          1. JLM

            Never had anyone who got that “settled” because if they did, I would go light a fire under their ass.  Literally.One of the most courageous actions I ever undertook personally was to completely restrain myself when I got a call that one of my office buildings was on fire.I was watching the news on TV and got a call from the portfolio manager who told me the building was on fire.  I had been watching the fire on the TV.I asked him how his job description began — “I am responsible for everything that happens or fails to happen…”I said to him:  “Good luck.”  And then I hung up.You cannot just preach empowerment and leadership and responsibility and hiring good people — you have to live it.  You have to force it to live.And when you live it, sometimes it will drive you absolutely fucking crazy because believe me I wanted to hop in my car and go down there and grab a firehose out of somebody’s hands.My guy had it completely under control the next day — building dried out, damage crew hard at work literally 24/7, not a dollar of lost rent and a great story to tell.The story became a legend in the local real estate industry.  You called up the CEO to tell him a building was on fire and he told you “good luck” because you are in charge.  He did not hold your hand and he expected you to perform.In my company, it reinforced everything that I wanted folks to know — you have to take charge and make stuff happen.It was real world not theory.The building was a midrise office building that was also our corporate HQ.That guy became the COO of my company years later.

          2. JamesHRH

            love it.

          3. Alex

            Literally? You’d literally light a fire under their ass. Somehow I doubt it.

          4. Casey Schorr

            Great story!

          5. Sandy O'Gorman

            This can be challenging, however, often the biggest learning opportunity as they will find they can do more than they thought and it is your support, inspiration, doggedness that makes them move to try it out…..move them early and often if they are smart, good and share your cultural values.

          6. Emily Merkle

            contentment is not negative – can be optimized to foster optimal safety / performance /etc

        3. Emily Merkle

          and the ability to reframe contentment to one as comfortable and the seeker as – boring. if you love what / why / how it is beyond passion, transcends wealth/can give birth to bonds that are stronger than any other and foster salves for seemingly fatal wounds that conquer all strive and are just – good stuff.

    3. Donna Brewington White

      I often ask clients embarking on a search, “What will be different about your company as a result of hiring the right person?”  I don’t always expect them to be able to give a detailed answer, but I like to challenge them to think about the search in this way.  Any response I receive is helpful in framing the search and to use in conversations with prospective candidates.  A great selling tool.I am generally recruiting at the higher levels of the company.  The question can be rephrased to encompass various levels of the organization.  The bottom line is that most people want to make a difference…and this is always true of people attracted to entrepreneurial environments.

    4. Tim Washington

      Couldn’t agree more with Spolsky’s management approach.

    5. Robert Thuston

      Eye opening about the Marines.  It’s a good way to motivate people. “sorry kid, you’re just not good enough”

  61. BillMcNeely

     When an organization/country’s success is personality driven, things go really well (Dubai, if you can forget the debt issue) or really poorly (Arab Spring anyone?) 

    1. JLM

      What everybody is missing about the Arab Spring is that this is the opportunity to bust up OPEC and the unified hatred of Israel.The first to go will be those countries who do not have a lot of oil, then those who have gobs of oil but cannot control their economies/societies and then Iran.I get a kick out of Iran — talking about how they are going to punish Israel.  What do you think is #1 on the target list?  The means of retaliation, stupid.Once they take out rocket launching sites, then they go get the centrifuges.  Centrifuges require a lot of electricity, the power plants go also.

      1. LE

        Back in the day my father was an importer of giftware from Israel. Every time there was a conflict, business went up. It was a form of support. He no longer owns the business (other family members do). Much of the products now actually come from China and other low wage countries. Last year my cousin was telling me how the factories in China are increasing prices for all their merchandise (I know this is common knowledge I just found it interesting to hear it from someone who frequently travels there and is experiencing it.)

        1. JLM

          Business is likely to be booming soon!Unless you have had 6MM of your tribe killed for just being in the tribe, you cannot imagine how much the Israelis are not going to allow another mad man to have dominion over them.I predict that Mahmoud Ahmadinejad is assassinated by his own people.

          1. jason wright

            The Persians know a thing or two about history;…History repeating itself is the lesson of history. That’s my thesis.Israel has approximately 200 tactical nuclear warheads in its arsenal.You reap what you sow.

  62. Office Ant

    This model is only true for products where the domain (and sales) knowledge also lies with the ‘work layer’. Like Stack. Or like one of these other companies who create (great) developer oriented tools. For most many domains though, if you want to make it big, you’ll need to get real good in-depth knowledge of markets, sales processes etc… the kind of stuff the work layer rather isn’t bothered with (and as an engineer, I happily include myself in that).

  63. vruz

    Re:  the SSD engineers know better.I tend to think of the role of management as administration too, but there’s another aspect that Joel seems to overlook here.Take for example Joel’s SSD engineers, they will come up with the most efficient solution in many regards… speed, reliability, availability… but they may not be in the know of other aspects…. for example, how much does it cost us to house the new supercallifragilistic data centre if we have to rent or buy property downtown?The engineers will have to talk to the finance guys, who will have to talk to the acquisition guys.”Management” probably won’t be able to outsmart a joint decision taken between those two departments.So management doesn’t have to be *only* an administrator but also an arbitrator.An arbitrator is not so much an administrator shuffling the seats around in a theatre, but more of a chorus director, choosing the best voices and arranging them in such a way that they sing in beautiful harmony.Yes management does have to take decisions, and the most important decision is who to listen to.In part the range of voices you will listen will be defined by your recruiting decisions, but in the long run the chorus needs to be arranged too.

    1. darius

      Not simply “Arbitrator”, either. Management must frame the question… help the team ask the right questions and look at a problem from the various important perspectives. I like the “Chorus Director” analogy. 

      1. vruz

        You’re right.  Asking the right questions and providing a vision of what the company as a whole is set to accomplish.

  64. jason wright

    Pyramids,… triangles.I prefer circles.If top management is at the point of a triangle (of whatever orientation – upside-down, downside-up, sideways-right, sideways-left) it’s a long way from the significant number of people in the organization. A circle puts top management within equal and shorter reach of everyone in the organization, and visa versa. This structure kills off unhelpful terms like “top-down” and ‘hierarchy’. The center of the circle should be where we find the product and the user. Wrapping around it the people who produce the product and serve the user. Around that the thinnest possible management ring.

    1. panterosa,

      Again we agree on visuals. 

  65. jbcolme

    AWESOME post. Really. 

  66. Modify Watches

    I need to be better at being the administrator. Awesome perspective, that nugget alone made this gold. Thanks Joel!

  67. anonymity86

    A lot of truth here. But, what about providing direction? I know of companies that just move forward on the path they were set, the engineers engineer, the salesmen sell, etc. but they never build new produvts or make great change. A CEO has to set the direction of the company he DOES have to make decisions… and then step out of the way.

  68. annmariastat

    I am amused by the number of people convinced of the importance of “vision”. I spent months putting together the design we just submitted on a proposal. I’m the president and majority owner – and I agree completely with this post. For much of the design, I coded snippets just to test if this or that piece was possible. I’d say I’m the second best programmer in the company. BUT there are people much better than me at other aspects and I let them do what they do. When they talk, I listen to them and in their areas of expertise I would never consider contradicting them or going against their advice, just like I wouldn’t consider their views on Java versus Javascript.When I think about the bosses I have had in my career, the most effective ones were those wh o got me the best technology, paid we well, let me work the hours I wanted (as many hours as you want, as long as it doesn’t start before 10 a.m.) and got out of the way. I appreciated that and tried my damnedest to deliver amazing results. That way, both me and my manager were able to continue along being happy.One of the best people I ever worked for was not a great engineer (though he was good enough). He said, “My management philosophy is to hire the best possible people, keep them happy and stay out of their way.”He had a very successful career.

  69. Reddy_s

    Great Insights  “The Management Team” from Joela/ Command and Control probably worked great in the toothpaste factory where Charlie Bucket’s father screwed the little caps on tubes.b/ This system is also pretty obvious, so it’s what 90% of startup founders try first.c/  You don’t build a startup with one big gigantic brain on the top, and a bunch of lesser brains obeying orders down below. You try to get everyone to have a gigantic brain in their area, and you provide a minimum amount of administrative support to keep them humming along.d/ This is my view of management as administration

  70. Fareed Mosavat

    The worst part of the Steve worship is that he didn’t actually lead this way, or if he did, only on the things he felt extremely strongly about, in a space he felt like he actually understood. At Pixar, he was the ultimate hands-off manager – hired great people, and facilitated their genius. 

  71. jfccohen

    Another really interesting article along these lines came from HBR about a tomato manufacturing plant that has no managers.  They enable all of the employees to make decisions and their peers hold them accountable.  Contracts are negotiated by peers, purchases are approved by peers, raises are evaluated by peers…it’s a completely socialistic model that works (up to a certain size, I think).Check it out:…

    1. Cynthia Schames

      That actually sounds to me like a big gigantic waste of time. I believe strongly in empowerment. I do not believe in decisions-by-committee, primarily because companies are made up of human beings.  Even the best of us can be petty, ignorant, egotistical, and vindictive at times or in certain circumstances. 

      1. jfccohen

        Ah what a sad remission from the potential of humanity.  First off, did you read the article?  If not, I highly recommend it and it addresses many of the things you mentioned above.  Most notably, empowerment – the thesis is that if you empower bottom up, you’ll get better decision making and more productive work forces.Second, you are forgetting the core elements of humans – that they are social and gravitate towards being in and forming communities (Rousseau), know that by helping others they inevitably are creating a tacit contract that will help themselves more (Tocqueville) and their actions are mitigated by the threat of accountability (Foucault).  So you are right that in a place of low accountability, lack of social structure and broken contracts, people are generally assholes.  This happens with scale as the tightness of community is fragile with numbers and there is an inverse relationship between size of community and accountability (think of commonality and lack of prosecution of petty crimes in NYC vs Sioux Falls, ID).Third, there is an equilibrium where all of these cornerstones of humanity perfectly coincide to make a “perfect” society.  This is what made Marx think it was possible at scale – he was wrong.  Modern day versions are Kibbutzes, egalitarian communes (Twin Oaks) and businesses with shared values.It isn’t easy to strike this relationship where it is a bottom up governing structure but it is possible.  The article addresses how hard it is for them to hire the right person.  The petty, ignorant, egotistical, vindictive, and brutish people get block out and/or pushed out by the community-minded in an effort to protect the community at large.  Much like white blood cells attack foreign bacteria and viruses, strong communities expel those who don’t contribute to the overall growth of the mini-society in favor of their own personal gain.Read the article.  Seriously.

        1. Cynthia Schames

          I did read, I’m not forgetting, and I think “perfect society” is a pipe dream.  I simply disagree.Don’t get me wrong, there ARE amazing selfless, fair-minded and equitable groups and organizations/organisms/communities made up of humans. But I believe they’re organic and not created, and I also believe that where there are dollars at stake, a business environment based on some utopian societal ideal has little chance of long term success.I believe in empowerment.  Truly.  I just don’t believe everyone has to have a vote on whether bagels or doughnuts are a better choice of refreshment for the morning meeting.  Let the carbs expert choose, and let everyone else concentrate on what THEY are best at, and get back to work.

          1. jfccohen

            Fair.  Disagreement is what makes the world go round.There is an interesting tangent here of organic vs. created.  Can a seductive vision organically create a movement or does it take a leader to create it?  Are the two mutually exclusive?I would argue that literally no human groups or communities are organic.  It’s growth could be (people voluntarily join), but at some inevitable point, it must be organized, directed and lead – now we’re back to created.  The question is whether or not someone can create the parameters for organic growth.  I believe that can, and occasionally, is done quite well.

          2. Cynthia Schames

            It’s really fun to argue with you :)I have to give some more thought to the idea of creation of parameters for organic growth.  My knee-jerk is that by simply performing the act of defining parameters, you cross from truly organic into structured.  Which isn’t bad–and certainly not when you’re talking about a company vs. an online community, for instance.  It’s just different.

          3. Emily Merkle

            i think there are parameters that can be set, influencers of population you want to attract,  empowerment, understanding of human motivation and reward the participation & engagement that will help community thrive…certain point members see and realize personal benefit…yay!see also social media & cultivating audience & engagementfun stuff to blue sky about

          4. Emily Merkle


          5. Chris

            I suspect you do not mean to imply we should strive toward  simply adequate societies. Striving towards a shared vision of perfection or utopia is precisely how we have advanced as far as we have. There are of coures many paths and many visions. 

          6. Cynthia Schames

            You’re correct, I don’t believe in adequacy.  I believe in striving toward perfect products, perfect user experiences, offering perfect value propositions. Striving toward a perfect society is a bit navel-gazey to me.  I’m a much bigger proponent of GSD. 

          7. Emily Merkle

            align self interest performance rewards with group interest performance rewards

        2. jmartins

          IMO, it makes little sense to speak of community “at scale”.  Corporations/counties/cities/states/countries will never reliably function as close-knit cohesive communities. Rather, they’ll continue to function as fractured and reluctantly interdependent self-interested clans. Most people have difficulty with context and concerns beyond the end of the blocks on which they live or the groups in which they work.As an aside, the notion of finding the “right” people is contextual. If the community at large is filled with asses that community may prefer like-minded, petty, egotistical, ignorant, greedy, vindictive brutes. And these days they’d have absolutely no problem finding qualified candidates. We can imagine a handful of industries that fit the profile.

  72. jmartins

    I’ve enjoyed many of Joel’s posts over the years but this is not one of them.  The front line is important and it should make decisions within its scope, but that is true of each level within whatever pyramid you happen to have…ten levels or two, upside down, down side up or sideways. Just as the engineers might be best equipped to make technology decisions and recommendations, project managers, sales, marketing and upper level management are best equipped to make decisions that affect the scope, structure, speed and direction of the company and its many projects.The pyramid is fine just the way it has always been so long as the folks at each level trust in the expertise and decisions of their colleagues and coworkers above and below them. No need to reinvent the wheel Joel. Just smooth out the bumps.

    1. Emily Merkle

      sometimes it is good system check to dust off the process theories you hold and defend them / debate with new voices, new data, etcprovocation makes you think

      1. jmartins

        I enjoy challenging the status quo Emily. It’s not that Joel’s suggestion wouldn’t work for others. He certainly claims that it works quite well for his group, and from what we can see publicly the results appear to agree with his claims. And I believe in doing whatever works for one’s own situation.But beware of those who claim X doesn’t work/scale and Y does, when X may not have been the issue at all. Oftentimes different people, groups, cultures and companies flourish differently (e.g. working with developers in Moscow was dramatically different from my experience here in the States.)Based on what little I know about the company, Fog Creek appears to have always been organized and run as a large development team hawking its own G2G (geek-to-geek) product. That’s dramatically different from running, say, Microsoft or IBM–very different animals operating at massively different scales. Frankly, “marketing” at FogCreek consists of stroking the egos of its “rock star” developer customers.Don’t get me wrong, Joel’s blog about software development is IMO one of the best on the subject. This is one of the rare times when I disagree strongly with what he’s written. He should stick with managing development teams…it’s a model that’s not necessarily appropriate for other types of businesses, or at significantly larger scales.

  73. panterosa,

    The humor of the post is almost rivaled by the sparring back and forth in the comments!But why has no one mentioned the family model for companies??? I am mystified. The nurturing and then freeing to work method is so much of parenting.The bad fit of leadership and personality/style is so much a part of families that don’t work. The good fit is what we dream about, harmony, growth, and synchronicity. Leadership style is subjective – what you need to lead you well, motivate you, etc. Good dynamics are far better than the right set of rules to follow, or the supposed golden guru/leader.

    1. Emily Merkle

      maybe mentor – not nurture so much 

  74. Youssef Rahoui

    “The “management team” isn’t the “decision making” team. It’s a support function. “I like that.Reading this article, I came out with the idea that the CEO’s job is just that: set the vision, hire, empower ; repeat, rince.

    1. fredwilson


  75. Donna Brewington White

    I just can’t tell you all the negative connotations that use of the word administration (in place of leadership) dredges up. Makes me squirm.And at a startup?  Leadership as administration? You’re kidding, right?What’s wrong with having a team of leaders?  The right type of leaders who empower.You are going to have leaders anyway, no matter what you call them or what title you give them.  Just make sure you have the right ones.And when you talk about decision making, well there are decisions and then there are DECISIONS.  The latter create framework. Anyway, you must be doing something right and a lot of what you are presenting here is exciting.  It’s fun to shake things up a bit.

    1. fredwilson

      i said right upfront that Joel’s approach is unorthodoxbut it sure works for him

      1. Donna Brewington White

        You did say that.I am glad that you included this post in the lineup. There is a lot to glean from Joel’s approach.And look how much fun we had with this bone that you threw us.

  76. sigmaalgebra

    There are mixed messages here from Spolsky and some of the comments on the ‘efficiency’ or ‘efficacy’ of academic organizations.I confess:  I never for even a nano second had any desire to be a prof, but due to some circumstances beyond my control for a while I was a prof.  I didn’t like it.  But I’ve come to understand some things about such work.As a former MBA prof at a research university (and Ph.D. from a good research university), I’d like to clarify ‘efficiency’ in academic organizations:Some of the very best work in our civilization, country, and economy comes from academic research.  I gave some examples here in…Yes, such work looks like a mountain stream, full of cold water, mud, and gravel, but is also one of the best places to look for gold.I concede that a university administration can have a tough time deciding on what statue to put in the quad or the policy on student cars on campus.  And any questions that involve ‘political correctness’ can be more difficult, still.But, at a research university, the main point is, may I have the envelop, please?  Here it is:  There are three main points, “Research, research, and research.”.So, how does a research university ‘manage’ such research?  Does the president, academic dean, dean of the school, or department chair do it?  Not really!  At most only in a quite indirect sense.Instead, there is an axiom:  The core of great research is great thinking, and that happens within a single head.  Not a committee.  Not a team.  And not directed by a manager.  The axiom may not be 100% correct, but it’s close.So, in a research university, a research professor is mostly just left alone to pick the problems, find the solutions, and publish the solutions.Then the university president, deans, and department chair evaluate the research?  Typically not!  Mostly they won’t even be able to read the papers!  Instead, the evaluations are based on versions of ‘peer-review’, nearly all from outside the university:  Are the papers in highly regarded peer-reviewed journals?  Are some of the papers marked “exceptional”?  Did the prof get invitations to present the research at conferences?  Is the prof asked to be a journal reviewer, editor, or editor in chief of work in his field?  Is the work frequently referenced by others?  Did the prof receive prestigious prizes for the work?  Are other research universities inviting the prof to ‘visit’ (recruiting trips)?  Does the prof get asked by NSF/NIH to review grant proposals?  Does the prof get NSF/NIH grants?Research that looks good on such criteria is a good bet to be really good on nearly any related scale.So, if the prof is in an engineering school, do business people with tough engineering problems come to ask for help?  Is the prof a practicing engineering professional?  Do students in the prof’s courses get great job offers from companies who hire engineers?  Does the research the prof does result in patents and patent revenue for the university?  Does the prof serve on boards of directors of companies that do advanced engineering work related to the prof’s research?  Are people getting venture funding for startups where the ‘secret sauce’ is based on the prof’s work?  For each of these, typically not!  Yes, in some ways Dean Terman created some exceptions.  Still, so far at a research university a school of applied science or engineering is not a research-teaching hospital for addressing technical pains in the business world or a school for training engineering professionals; maybe it would, could, and should be, and at one time I so assumed, but it’s not.Still, net, for the style of research the research universities do pursue, that research is relatively well managed.  The best results are magnificent, just top, center crown jewels of civilization and, in particular, among the most valuable parts of our economy, health care, national security, etc.Could an information technology startup borrow from how universities ‘manage’ research?  Yes, a little.  But I doubt that many such CEOs would be at all comfortable with such a process.  So, there is a problem for businesses to ‘manage’ some of the best technical work, and as usual the flip side of this problem is an opportunity!

    1. LE

      “So, in a research university, a research professor is mostly just left alone to pick the problems, find the solutions, and publish the solutions.””Then the university president, deans, and department chair evaluate the research?  Typically not!”My sister writes research grants for academic institutions and has co-published some things as well. She once had a job as Director of Research at a hospital system. She ran into trouble and got forced out when she tried to challenge some of the Cardiologists and the projects they were trying to get funding for. Bad move to challenge rain makers.

      1. sigmaalgebra

        Easy to believe.  What I described about a research university is quite narrow.  Biomedical clinical research in a hospital can be quite different.  Also biomedical research can make much larger uses of teams than what I described.  While apparently nearly all the best research in  theoretical physics is from a single researcher, much of research now in physics is part of ‘big science’ with many names on a publication.Again, what I described is narrow. My main point was to illustrate how some of the best academic research is managed by, really, being, as in the quotes you gave, left alone and not really ‘managed’ at all. Then the evaluation is heavily from outside the organization. These points are, for a business, surprising, and maybe some businesses could learn from them. A lot? I doubt it, but maybe a little.

    2. Emily Merkle


  77. Donna Brewington White

    In my first professional job, I wrote job descriptions at a $6B financial institution, from the mail room to the penthouse.One of the shortest job descriptions was for the CEO. That was around the same time that someone explained to me that our CEO made the highest salary in the company for making one or two decisions a year. 

  78. Scott Barnett

    Always hire people smarter than you and let them do what they do best.  I love the upside down org chart, I’ve always “talked the talk” in that regard but never shown it in exactly that way. It is the exact way to be running your business.

  79. James

    Steve Jobs on managing people:…At least in his own mind, Jobs seems to be in agreement with Joel’s concept of hiring smart people and giving them the freedom to operate with ideas not hierarchy.

  80. dkural

    If you want to build a ship, don’t drum up the men to gather wood, divide the work and give orders. Instead, teach them to yearn for the vast and endless sea.– Antoine de Saint Exupéry,  Citadelle, 1948, misquote



    1. fredwilson

      i’ll meet that with +10 for having the guts to take a shot at Steve Jobs while he’s being canonized



      2. Emily Merkle

        dude Steve would say “bring it”

  82. Michael Diamant

    “….As a service corps that helps the talented individuals that build and sell products do their jobs better.”Key responsibility: making sure everyone top to bottom has the tools to do their jobs well.  If you can’t allocate resources to give people the tools (intelligence, software, budgets) then you can’t expect them to produce the results you’re looking for.  

  83. Robert Thuston

    I disagree.  Steve Jobs was a good manager because he had an insane commitment to getting things right.  There are nicer people out there with this same quality;  Steve just happened to be a dick.  He demanded excellence of everyone in the organization, and that translated into raising the performance of the whole.  I don’t think you can just flip a chart upside down and say, wow, now we’re thinking different.  Although interesting, I think it’s too soft.Leadership is all about culture – you demand excellence, they demand excellence – you hire badasses, they hire badasses – you help others think through priorities, you hold people accountable, you debate with people to find the right answers, and they do the same.I think its a misnomer to say that you “hire the right people and get out of the way”.  Take a look at Nick Saban, the guy is the heart and soul of the team, he’s a hands on leader involved in everything.  Nick Saban scales.

  84. Devrim Yasar

    by your own reasoning, you could also deduce the fact that Steve Jobs was in fact a leader who did what you prescribe to others – because he created a culture that utilizes everybody’s brains..  just sayin.. (don’t get me wrong, i loved everything else in this article :))

  85. jjxtra

    Then why the frack do the people at the BOTTOM of the pyramid get paid / have stock options exponentially greater than those at the top of the pyramid?

    1. fredwilson

      it’s a tough job to have the entire org on your shoulders

  86. Chris Kenst

    Joel’s style of management seems like it’s more reproducible / sustainable than Steve Jobs’ style. Meaning if the CEO or VP of Engineering leaves you can move another administrator into the position without disrupting the team or products. With Steve Job’s autocratic top-down style, no one can replace him, leaving the company and teams vulnerable.

  87. rufwork

    “The chairperson of the department might call meetings and adjudicate who teaches what class, but she certainly doesn’t tell the other professors what research to do, or when to hold office hours, or what to write or think.”That’s quaint, even if we ignore for-profit first schools of higher learning for the time being. Sure, we thought that’s how things worked when we were in the academy, but spending a little (ha) recent time in graduate school has finally convinced me otherwise.  Now, in 2012, academic freedom has been differently, yet still fully, inhabited and exploited by the same C&C [and ostensibly market-based] mentality you’re lamenting.  There’s still plenty of ritualistic kowtowing to exhibit a lack of interest in social Power, but it is now largely display.  You now research primarily for funding (and the 50% kickback the department receives), even in the humanities, because research solely for knowledge is too easy to leave on the budget cutting floor.(As an aside, universities would actually do well to follow Picard’s lead over what they’ve got currently.  And thank you for the moral of the story, “But you? You are not Steve Jobs. ”  That’s well said.)

  88. Greg

    YES!Nice paradigm shift… Management books and company leaders bang on about “servant leadership”, but turning the org chart on its head really brings this concept home, cultivating a change in the company culture, changing the way employees see themselves, and effectively increasing employees’ perceived value!

  89. Michael Mullany

    Joel. I think for fuller context you need to add in what you wrote in a blog post long ago – which is that bottoms up management only works when “Everyone knows the business and everyone knows the technology”. If management doesn’t invest in teaching everyone about the business and the technology then empowerment can easily go off the rails (sometime hilariously).

    1. Emily Merkle

      lol – some say hilariously … others say in flames..

  90. caretakerbob

    Joel’s inverted pyramid is right for the wrong reason.   All the benefits he lists are indeed great side effects. But to figure out why to invert your org chart, ask the question: where on the typical corporate org chart would you put your customers?   Down below your most junior frontline staff members typically, ie at the bottom.  When your customers should always be put at the top.This inverted org chart actually works brilliantly for any company up to about 150 staff members.   In organizations larger than that all the traditional training your staff have had outweigh management’s ability to explain why your org writes the chart upsidedown.  😉

  91. W_public

    Intel operates a bit like Apple. A bit of emotion and passion would help a lot of companies. What would computing be like without Apple. It’s hard to assert that other companies have been as successful. That’s a different kind of fallacy.

  92. Hung_Dinh

    This is what people call Managment 3.0 ( you guys can search and read this great book! )we are living in the age of knowledge management, no longer labour management.

  93. new1234

    This article highlights the difference between entrepreneurs and engineers.  i feel Joel comes from an engineering background.  I am also an engineer and have worked in several high-tech startups.  I’ve heard many a talk from CEOs, supposedly creating vision and inspiration.  They believe this is motivating and visionary, showing projected growth curves, and how the world’s going to be changed, etc.  To the mind of an engineer, the opposite is true.  It tends to indicate that the CEO has very little clue of the technology and the product.  My motivation does not come from ‘buying’ into the vision.  My motivation comes from making something that is useful to someone else.  I like creating stuff.  I work hard for these reasons.  Frankly, very very few products change the world.  Most help someone do something and are willing to pay a little for it.  So I agree with Joel, cut the crap and let engineers make stuff of value.

  94. Emily Merkle

    No doubt. Ego that impedes forward progress or taints culture is not ok.All relative. If ego is petulance and contribution is there – Keep ego away from all humans and limit public representation if necessary :)Calculate ROI.

  95. Tgbwilson

    Interesting concept, well described; however, I would distinguish between strategic and operative management. The creative programming geniuses may be the right folks to build the product right, but somebody with an awareness of the industry and market needs to decide *what* to build.

  96. Asdf

    In some ways, the US military command structure is similar to what is described here.  I’ve heard officers described as “support staff” for the enlisted forces.  The officers make broad policy and make sure the bullets and bombs get to the field, but ground level tactical decisions are often made by experienced NCOs.

  97. Lucian Teo

    When your experts can’t agree, management needs to step in to take responsibility and make decisions. Highlights the importance of getting the team working together, empowering them, and making sure they are able to communicate their ideas very clearly so that less expert brains can ascertain risks accurately. 

  98. rustlem

    From the Tao Te Ching:The best rulers are scarcely known by their subjects; The next best are loved and praised; The next are feared; The next despised: They have no faith in their subjects, So their subjects become unfaithful to them. When the best rulers acheive their purpose Their subjects claim the acheivement as their own.

    1. Emily Merkle

      and why can this not only be possible but probable?

  99. Vivek Gani

    Regarding the steve jobs-isms: there’s a story about how steve almost didn’t have the auto-insert floppy disk drive on the mac if it wasn’t for an engineer hiding a sony engineer in the closet away from steve to finish working on the design. The engineers knew it just couldn’t be done steve’s way:…”When Bob Belleville revealed that he and George had kept the Sony alternative alive, Steve swallowed his pride and thanked them for disobeying him and doing the right thing. The Sony drives eventually worked out great, and it’s hard to imagine what the Mac would have been like without them today. “


    Maybe you’re just looking at dysfunctional organizations?Top management is suppose to be big brains that lead the company. Middle management is suppose to optimize daily functions. Labor is suppose to work in the trenches.Information from the trenches is suppose to be presented to the leaders so they can make informed decisions. Middle management is suppose to work on collating and presenting information to leaders and labor.If people don’t use a system properly then it won’t work as intended. Make sure the system is being used properly before abandoning it.

  101. Gabriel Trovão

    “hiring smart people who get things done—and then getting the hell out of the way”. Great post

  102. systemovich

    Spolsky for president!

  103. Rajasekar KS

    Good entrepreneurs recognize their shortcomings and hire talent (and give freedom) to take care of the missing pieces. If you want to control and command, you won’t be able to trust your own shadow.

  104. andyswan

    I thought this was pretty anti-leadership:”The “management team” isn’t the “decision making” team. It’s a support function.”I never equated being an asshole to being a good leader. I don’t believe that at all. In fact, my guess is that it’s fairly irrelevant. Definitely not my style haha.

  105. Aaron Klein

    There is a huge difference between “lead without being a jerk” (which I believe in very strongly) and “lead using a university model.”Read my description of the university model (which I’ve seen firsthand) and I think you’ll see what I mean by that.

  106. andyswan

    I am considering the context.  Are you?He VERY specifically gives the label of “administrator” to the CEO, COO and others primarily responsible for vision and leadership.  The entire essence of this post is that they are one and the same.  He very specifically states that he does not want the management team to be seen as making the final decision…which strikes at a core principle of leadership.

  107. FlavioGomes

     Some people mistake decisiveness for being an asshole.

  108. andyswan

    Who cares what those people think anyway.

  109. JamesHRH

    Experience says that most people want to be lead.That’s different from micromanaging and not at all administration.The admin theme is extremely engineer-centric.

  110. leigh

    I see how you got there.  I don’t like the administrative label on the leadership team much either.  Truth is great leaders are often creators as much as enablers.  Do agree with @ccrystle:disqus below though – probably not the point he was trying to make 🙂

  111. andyswan

    ya we all agree but what’s the fun in that?

  112. leigh

    sorry i disagree

  113. FlavioGomes

     Sure that goes without saying…but its interesting to note that the definition of an asshole is somewhat of a moving target for some.