I coined the term “open allocation” here (I didn’t invent the concept; I just gave it a name) and I still believe that OA is the right way to run a tech company. However, “no bosses” is a bit too extreme and I don’t think it would work for most companies. Tony Hsieh deserves a lot of respect for betting big where others wouldn’t dare– holacracy, Las Vegas, an industry that most techie sheep would consider too “boring”– and I really hope that he succeeds.
The problem with “no managers” is that it purports to eliminate the coercion and control structures, and it probably does to some extent, but it’s very hard to make that stable. You can get rid of the middle management layer easily, but you still have owners and de facto executives, and you have people who want to be owners and executives and whose desire to elevate themselves is strong enough that, even if they don’t actually need to be monitored, tempered, or ejected, the perception that such people exist is very hard to wash away. If I have 0.05% of the company and the CEO has 20%, he’s going to be somewhat justified in suspecting that I want his job or, at the least, would be inclined to do things that aren’t necessarily in line with the company’s interest in order to increase my compensation or equity.
No matter what, in any organization there will be people who can offer protection, and those people will evolve into bosses. It’s not necessarily good or bad, just inevitable. You can dictate how they use that power and hope that, most of the time, people follow the rules and those who break them are taken down by those around them. For example, open allocation tells bosses that they shouldn’t use their power to determine what gets worked on (and one hopes that any who try will be chastised for abusing their positions). It separates the intense conflict of interest that exists between people and project management and allows the latter to be democratically selected. It doesn’t actually remove the need for management. You still need a last resort when it comes to conflict resolution, and you still need someone to keep the culture clean and prevent, say, bro culture and sexual harassment. Management can get in the way of open allocation, but the lack of it can also get in the way, as people with the power to sway the group and the executives (sometimes through fear) become the de facto new middle managers.
Open-plan offices are also a bad idea. You need to give people a private space for work and have open spaces available when people want to collaborate. You give everyone a 100 SF office and have about 100 SF per head of open space and private meeting rooms. You’re talking about $250-500/month per person, which is pennies for cognitively intense work. To be honest, the Zappos open-plan environment looks like hell. This stuff about “The Beach” and “lead links” sounds bizarre and humiliating as well. It seems like a “lead link” is a manager but you’re not allowed to say that he’s a manager, because that would challenge his claim of natural leadership. Which, if I’m correct, would make things not significantly different from the culture of mandatory consensus (i.e. having to pretend to actually like the boss’s ideas, instead of just implementing them; because to complain attracts future scrutiny and even if you do a great job anyway, the suspicion will lead to any shortfalls being attributed to negligence or, even worse, malicious sabotage) that exists in the modern corporate world.
I wish that bosses weren’t necessary, but I know what people are and, unfortunately, humans will create a hierarchy out of nothing. Creating a middle management layer is something companies do in order to control that leader-discovery process. Now, they often control that process in a bad way– choosing people whom executives believe they can control, rather than the people whose technical or interpersonal skills merit decision-making positions– but that’s another discussion.
To be honest, I think that the old corporate system did a better job of realizing meritocracy than these newfangled startup cultures. Sure, incompetents would get still get their seniority-based promotions and high salaries, and be “Vice President” after 15 years, and that burned people. On the other hand, people actually knew who the solid people were, and who the jokers were, because there was time for that institutional knowledge to grow. If Tom was an incompetent who’d made VP by being there for 20 years, he might get “20 reports” but the company would know better than to put any of its best people under him. If Bob was really good and had only been there for 7 years, he’d make less than Tom, but he’d get more important projects and his pick of the new talent. Eventually the Bobs decided that this was unfair and, fed up with their slow advancement, left en masse, and there were a variety of other huge problems with the “one job for life” corporate system, but I think it did a pretty good job (relative to its successors) of establishing meritocracy.
Compare this to our current, job-hopping climate. You don’t move up the chain through hard work over time, but by constantly renegotiating your compensation and position, and by representing your work experiences (regardless of what they were) as a track record of increasing successes. This is probably good for people like me, so I won’t complain, but it does mean that there’s less of a chance for true meritocracy. In the old system, organizations “knew” that they had to give the semi-incompetent but well-liked senior people high salaries, and just limited their damage and developed an in-house knowledge of who was solid and who was a joker. In the new system, you get a better job by finding the company that’s so desperate to hire (often, due to expansionary pressures from VCs) that it will give you a 30% bump in pay. And, while organizations claim to hire and position people on merit and compensate accordingly, they actually compensate you based on what you can negotiate (based on your CV, your negotiation skills, and market conditions) and then back-fit your title, role, and placement in the management organization accordingly.
It’s easy to look at the picture and decide that, since managerial authority is just another token to be negotiated, and that it’s often divorced from merit, that the whole concept is worthless. And in software, there’s so much “Agile” junk science (90% of what “project managers” do, in this industry, is harmful) that it’s tempting to throw the whole thing out. I would support that, if I thought that there was a way to keep management and management pathologies out for good. I don’t really buy it, though. “Flat organization” usually means the transition from oligarchy to “benevolent” dictatorship, but the problem with “benevolent” dictators is that they still end up relying on lieutenants, and the people who succeed in establishing themselves as such a person’s lieutenants are often human garbage. The reason why this happens is that people respond more strongly to fear than so-called “vision”. Vision becomes a nice-to-have under threat of existential risk, and creating a phony sense of existential risk (“we won’t be able to deliver software unless <what I want>”) in order to get what one wants is Black-Hat Corporate Politics 101. Not surprisingly, when it comes to exploiting the fears of “benevolent dictator” figures, the people who are best at it tend to be the worst human beings.
In order to make full bosslessness work, we’d need to be in a post-scarcity state: basic income, and low economic inequality in the society as a whole. The old “Theory Y”, paternalistic companies worked from 1940 to 1980, but got stripped in the 1980s because it was now possible for the bad apples to sell out their assume-good-faith companies to private equity firms (in exchange for 6- and 7-figure jobs at said firms). That threat still exists. Socially, we as a species are just not “there” yet. In the mean time, we have to figure out how to make middle management work better because we have no hope of eliminating in, nor the forces and suspicions that create it no matter what we try to do.
I coined the term “open allocation” here (I didn’t invent the concept; I just gave it a name) and I still believe that OA is the right way to run a tech company. However, “no bosses” is a bit too extreme and I don’t think it would work for most companies. Tony Hsieh deserves a lot of respect for betting big where others wouldn’t dare– holacracy, Las Vegas, an industry that most techie sheep would consider too “boring”– and I really hope that he succeeds.
The problem with “no managers” is that it purports to eliminate the coercion and control structures, and it probably does to some extent, but it’s very hard to make that stable. You can get rid of the middle management layer easily, but you still have owners and de facto executives, and you have people who want to be owners and executives and whose desire to elevate themselves is strong enough that, even if they don’t actually need to be monitored, tempered, or ejected, the perception that such people exist is very hard to wash away. If I have 0.05% of the company and the CEO has 20%, he’s going to be somewhat justified in suspecting that I want his job or, at the least, would be inclined to do things that aren’t necessarily in line with the company’s interest in order to increase my compensation or equity.
No matter what, in any organization there will be people who can offer protection, and those people will evolve into bosses. It’s not necessarily good or bad, just inevitable. You can dictate how they use that power and hope that, most of the time, people follow the rules and those who break them are taken down by those around them. For example, open allocation tells bosses that they shouldn’t use their power to determine what gets worked on (and one hopes that any who try will be chastised for abusing their positions). It separates the intense conflict of interest that exists between people and project management and allows the latter to be democratically selected. It doesn’t actually remove the need for management. You still need a last resort when it comes to conflict resolution, and you still need someone to keep the culture clean and prevent, say, bro culture and sexual harassment. Management can get in the way of open allocation, but the lack of it can also get in the way, as people with the power to sway the group and the executives (sometimes through fear) become the de facto new middle managers.
Open-plan offices are also a bad idea. You need to give people a private space for work and have open spaces available when people want to collaborate. You give everyone a 100 SF office and have about 100 SF per head of open space and private meeting rooms. You’re talking about $250-500/month per person, which is pennies for cognitively intense work. To be honest, the Zappos open-plan environment looks like hell. This stuff about “The Beach” and “lead links” sounds bizarre and humiliating as well. It seems like a “lead link” is a manager but you’re not allowed to say that he’s a manager, because that would challenge his claim of natural leadership. Which, if I’m correct, would make things not significantly different from the culture of mandatory consensus (i.e. having to pretend to actually like the boss’s ideas, instead of just implementing them; because to complain attracts future scrutiny and even if you do a great job anyway, the suspicion will lead to any shortfalls being attributed to negligence or, even worse, malicious sabotage) that exists in the modern corporate world.
I wish that bosses weren’t necessary, but I know what people are and, unfortunately, humans will create a hierarchy out of nothing. Creating a middle management layer is something companies do in order to control that leader-discovery process. Now, they often control that process in a bad way– choosing people whom executives believe they can control, rather than the people whose technical or interpersonal skills merit decision-making positions– but that’s another discussion.
To be honest, I think that the old corporate system did a better job of realizing meritocracy than these newfangled startup cultures. Sure, incompetents would get still get their seniority-based promotions and high salaries, and be “Vice President” after 15 years, and that burned people. On the other hand, people actually knew who the solid people were, and who the jokers were, because there was time for that institutional knowledge to grow. If Tom was an incompetent who’d made VP by being there for 20 years, he might get “20 reports” but the company would know better than to put any of its best people under him. If Bob was really good and had only been there for 7 years, he’d make less than Tom, but he’d get more important projects and his pick of the new talent. Eventually the Bobs decided that this was unfair and, fed up with their slow advancement, left en masse, and there were a variety of other huge problems with the “one job for life” corporate system, but I think it did a pretty good job (relative to its successors) of establishing meritocracy.
Compare this to our current, job-hopping climate. You don’t move up the chain through hard work over time, but by constantly renegotiating your compensation and position, and by representing your work experiences (regardless of what they were) as a track record of increasing successes. This is probably good for people like me, so I won’t complain, but it does mean that there’s less of a chance for true meritocracy. In the old system, organizations “knew” that they had to give the semi-incompetent but well-liked senior people high salaries, and just limited their damage and developed an in-house knowledge of who was solid and who was a joker. In the new system, you get a better job by finding the company that’s so desperate to hire (often, due to expansionary pressures from VCs) that it will give you a 30% bump in pay. And, while organizations claim to hire and position people on merit and compensate accordingly, they actually compensate you based on what you can negotiate (based on your CV, your negotiation skills, and market conditions) and then back-fit your title, role, and placement in the management organization accordingly.
It’s easy to look at the picture and decide that, since managerial authority is just another token to be negotiated, and that it’s often divorced from merit, that the whole concept is worthless. And in software, there’s so much “Agile” junk science (90% of what “project managers” do, in this industry, is harmful) that it’s tempting to throw the whole thing out. I would support that, if I thought that there was a way to keep management and management pathologies out for good. I don’t really buy it, though. “Flat organization” usually means the transition from oligarchy to “benevolent” dictatorship, but the problem with “benevolent” dictators is that they still end up relying on lieutenants, and the people who succeed in establishing themselves as such a person’s lieutenants are often human garbage. The reason why this happens is that people respond more strongly to fear than so-called “vision”. Vision becomes a nice-to-have under threat of existential risk, and creating a phony sense of existential risk (“we won’t be able to deliver software unless <what I want>”) in order to get what one wants is Black-Hat Corporate Politics 101. Not surprisingly, when it comes to exploiting the fears of “benevolent dictator” figures, the people who are best at it tend to be the worst human beings.
In order to make full bosslessness work, we’d need to be in a post-scarcity state: basic income, and low economic inequality in the society as a whole. The old “Theory Y”, paternalistic companies worked from 1940 to 1980, but got stripped in the 1980s because it was now possible for the bad apples to sell out their assume-good-faith companies to private equity firms (in exchange for 6- and 7-figure jobs at said firms). That threat still exists. Socially, we as a species are just not “there” yet. In the mean time, we have to figure out how to make middle management work better because we have no hope of eliminating in, nor the forces and suspicions that create it no matter what we try to do.
So… like Valve?