Zappos says goodbye to bosses
washingtonpost.com
washingtonpost.com
You say tomato I say tomato. Seriously though, renaming a title does not equal removing it from your business structure.
Why not call them managers and simply take away the elements of their job which cause issues? Seems to be this is more about drinking the kool-aid than it is about actual structural changes to how things are accomplished.
But what are those elements? A good boss keeps a LOT of crap away from its employees, a bad one is the exact reverse. I want a manager, he does most of the crap political work for me while I grow professionally.
Is it a problem managers exists or is it a problem some companies can't manage to hire good ones?
"You guys are like a rock band. You're the stars. You need to show up, practice, and play gigs on schedule.
I'm like your band manager. It's my job to take care of all the random crap like booking tour buses, replacing blown amps, and protecting you from the annoying label guys so you can just focus and do your magic."
It's a good analogy. A good manager protects their team and lets them focus on their work and gives them what they need to succeed.
Yes, a manager is like a manager. :)
The difference is that the band manager is clear he works for the band. His goal is to help the band do what the band does best.
A bad SW manager often operates under the idea that he's in charge. He gets to set what the people do, when they do it, and how they do it. He doesn't really care what skills or experience his team has. He's the boss, and they better shut up and do what they're told.
A good manager plays with the team. A bad manager plays against the rest of the team.
Team manager a.k.a. office mom. Someone who is a lot like the band manager. He doesn't know C from HTML but knows that developers needs snacks, juice boxes, headphones, fast computers, and 4k monitors. He takes care of the "crap political work" and keeps developers busy with what they do best.
Project/product manager. This is someone who may or may not be the technical lead, but definitely needs to understand how software is put together, who on their team will be doing it, how long it will take, etc. The job of this person is to translate business requirements ("We need feature X shipped by March, and bug Y fixed immediately.") into technical tasks, timelines, and assignments. It's also possible that this is the person responsible for figuring out which features the product should have (no developers shouldn't always be the ones deciding this). This job is hard. You are responsible for setting realistic goals and not letting anything drop, but you are also responsible for moving the project fast, and not getting stuck.
The first job is more tactical. We know that developers constantly distracted by noise or going to be unproductive, etc. The second job is strategic. It requires a crystal ball to predict which bugs will come up and where the difficult bits of implementation will be. This isn't booking buses and switching out blown amps. Managers that can do this job alone, or even both at the same time are worth their weight in platinum.
Hypothetically let's assume you are a manager. You have a team of smart people and like any other team, everyone wants to work on the hardest problem. Particularly two guys want to do the same thing: How do you handle this?
The most important thing I advocate in a manager / employee relationship is trust. You should be able to trust that your manager has your best interests in mind and vice versa. If you do trust each other, I don't see why there will be a problem in doing what is told. In fact you won't even see it as being told to. The managers on the 'good end' of the spectrum will also make it appear as if it is a suggestion, but any smart person should be able to see through that for what it is: an order to do X or Y.
Most of the issues happen because of the lack of trust. This is the prime thing I tell new folks: Choosing a manager who is exactly on the same page is prime directive. Get the human element out and build trust. You miss out on that / settle, you are just settling for a life of misery.
The existence of people who are bad at their job does not mean that the job description is problematic.
I don't agree, just stating what they think they've figured out. I'm suggesting a manager (sorry lead-link) who can assign and remove people but not be involved in what they're actually doing to be highly problematic (I'm probably oversimplifying)
While developers may prefer to have their nose buried in code, there needs to be someone else who is coordinating the team and making sure the pieces are fitting together.
Anyway you phrase it, it always comes out looking like a manager.
You might say "why not require the developers to manage themselves" To that I'd say not everyone wants to manage themselves or others, and some folks don't want to be buried in work but would rather coordinate and lead a team.
Let the managers manage! If they suck, fire them. Don't feel the need to re-invent business just because your managers aren't doing their job.
I'd be really scared of working in an environment like this because it fosters politics and infighting.
Depending on your personal character and career objectives, this means different things. If you have an objective of making a splash and moving out of your management role, burning your team while achieving something that makes you look good will advance your fortunes, and some other schulb will bear the costs.
Those employees can have a hard time getting involved at higher levels of the company, exactly because their job is so cordoned off from everything. Which leaves people in the cliched position of needing to perform well enough Job A to get promoted out of it. (Which has always been a logical flaw in the corporate world.)
There certainly are other solutions to this problem -- this is just one solution, and is probably also targeted as a solution for other issues as well.
The normal way organizations try to decrease the amount of crap political work is to hire someone who just works on that crap all day. Managers.
Instinctively, you would think- this person produces nothing, adds nothing to the product, they should serve the actual workers. But this isnt usually the case.
Long term planning usually gets lumped into a manager's work load because its not something that needs to be done to get today's work completed. Because they are involved with planning future work, they end up owning that work later, hiring the people who do it, etc. There are also fewer of them in number and are more outspoken, so they end up being the one guy the bosses talk to.
Here is an alternate reality. An engineer comes up with a good idea for a new feature. He talks to his buddy DBA and SysAdmin. They think its a good idea so they agree to help out when they can. Engineer asks fresh out of college designer to help. Designer respects the reputation of the engineer and joins the team. A sales guy hears about the new feature and it sounds similar to something customers have been asking for so he tells them he can sell it. The bosses see a cluster forming in the corner and walk over and ask what they are up to. The feature is actually already implemented and being tested. Boss says it looks good, asks an analyst to see what its affect on factor x would be, walks on to a different cluster he sees forming across the room.
As their work on the feature is wrapping up people keep their ears open for the next project they'd like to work on. Everyone moves on to other ideas. People with good ideas become the nucleus for new clusters. People with good skills become members. From time to time the bosses look at the islands- the people always sitting by themselves. Either they are genius uber productive people or they are people who will probably be gone soon.
Communication is complex, and no matter how careful at scale there will be misunderstandings, hurt feelings, and bruised egos.
Resources are limited, and reaching a decision, if not a consensus, is an important part of successful engineering.
Time is valuable, and at times for survival a feature must be cut and a product direction abandoned. The handling of the removal of responsibilities and resources is necessary, but never something that is happily accepted by everyone.
In a system like this, who is going to do maintenance tasks? How do you keep people from just working on the fun things? Why would you expect an developer to have any insight into future market conditions? What happens when a customer says "I need this feature" and the account manager comes back and says "guys we need this feature" and no one wants to implement it?
This structure is great for a very specific type of company (B2C, creating cutting edge products that are fun to work on and interesting enough that they can hire the best talent [aka Valve]).
In most types of companies this would just result in the crappy work never getting done.
IMO you can look at this from an evolutionary standpoint. If this were such a great way to run a company don't you think you'd see more successful companies doing this besides Valve?
Believe it or not there are people who like doing the maintenance tasks.
>How do you keep people from just working on the fun things?
Why would you?
>Why would you expect an developer to have any insight into future market conditions?
I wouldnt. Although, if they are familiar with a product and its users they probably have as good of an idea as say, random account manager.
> What happens when a customer says "I need this feature" and the account manager comes back and says "guys we need this feature" and no one wants to implement it?
In my example the cluster formed around a developer. A cluster could just as easily form around a sales guy, an exec, the guy who makes sure the vending machine is always full.
The customer says I need feature x, the sales guy starts building a team while doing whatever work he can- taking requirements, etc;
It is the job of the bosses to build a good HR process and provide leadership and vision. If they are good at what they do and respected, when they say, "now we will move in this direction", the new ideas will start to gravitate in that direction. If not, there has to be a process to get execs who can do that.
> If this were such a great way to run a company don't you think you'd see more successful companies doing this besides Valve?
Heirarchy is something that has been firmly established in human culture for a long time. New ideas about how to organize humans are emerging and old ones are being revisited.
Having said that, the Gore company makes products for use by industry and has had a nontraditional organization for the last 50 years.
In the U.S. most non-hierarchical businesses are small local coops in the form of bakeries, coffee shops, etc- people experimenting. But there's no reason it couldnt work for a company that sells software or information.
As much as the world needs ditch diggers there are people who need jobs digging ditches. You dont have to force them to do it.
Sorry, but that's a recipe for shitty products.
A boss tells people what to do. A manager manages them, leads them, makes sure their needs are fulfilled, and protects them from unnecessary distractions.
Because they're not actually empowered to "manage" people? It sounds like what they're doing is deconstructing the role of "manager" and distributing certain elements of that role to different people. It doesn't appear to be a simple s/manager/"Lead Links".
Most people have heard of Valve, 37signals, Zappos, Google etc. not just for their products but also because of their attractions as employers. This is good marketing for what most of us deem as an extremely competitive talent squeeze.
You can read the details of holacracy here: http://holacracy.org/constitution
There are signifigant differences between what a lead link does and a traditional manager. They act more as a spokesperson for the accountabilities and priorities delegated to them by a larger democratic process.
This is yet another example of a flippant and largely content free criticism on hackernews made from a position of ignorance on the topic. I'm ambivalent on whether holacracy is a good system as I have no personal experience with it, but lets at least talk about what the article is actually about rather than just pattern matching whatever random crap is in people's heads.
I had a manager like that, set you a task, and how you did it was up to you. As long as it was progressing, he didn't care if I never came in except for the meetings.
- security detail
- private jet
- yacht
- professional sports franchise
- foundation
- once respected media outlet as personal mouthpiece
(Jeff Bezos, CEO of Amazon, parent company of Zappos, owns the Washington Post.)
Note: Amazon founder and CEO Jeff Bezos also owns the Washington Post
That should be given more focus IMO
There was a recent HBR article about Google's management structure. They tried eliminating managers in 2002, but too many people just came directly to Larry when they had a problem (1). Holacracy doesn't scale.
1) http://hbr.org/2013/12/how-google-sold-its-engineers-on-mana...
Note that Zappos is owned my Amazon.
Just because it's not the primary reason does not mean it's not happening.
It is possible to remove the management role, at least at a project level, by using technology. For instance, the first time I saw Trac our team effectively did this.
To be practical however, a technology solution really does require the right group of people. You can’t have people who are uncomfortable logging into web sites, or too lazy to write good descriptions for things, and so forth. You need people who are (frankly) professionals, who look at something like a giant task list and do a fair job of distributing things sensibly amongst themselves. A solution like Trac really starts to shine when people learn the little niceties, like additional formatting in wiki entries or the magic sequences that set up links between things. Again though: you need a team that is willing to work with such a tool in the “right way”. If they do then you don’t really need a manager at all.
Data is facts, it's not opinionated.
Metrics, on the other hands are built upon data but are often carrying an agenda with them, they're political (whether it's office or nation politics) tools.
It's linked from the article, but it deserves to be more prominently noted.
What is the secret sauce in Holacracy that makes it different from hierarchies?
To answer your question, the secret sauce of Holacracy is the Holacracy Constitution – it replaces the CEO in deciding how decisions get made and how we organize. See: www.youtube.com/watch?v=31MljhiyxZ8
Also, you might be interested in two blog posts published in the past couple days in reaction to the often-inaccurrate press: - "Five Misconceptions About Holacracy" https://medium.com/about-holacracy/da84d8ba15e1 - "Holacracy at Zappos" by a Zappos' employee: http://rianschmidt.wordpress.com/2014/01/07/holacracy-at-zap...
By work I mean, inevitably it will settle into a more informal hierarchy but still there in practice (just not in title).
That said, I think this can go either way. Not having defined roles and relationships might cause more friction that it relieves. And to quote Douglas Adams, "It is a well-known fact that those people who want to rule are, ipso facto, those least suited to do it". This might just cause a lot of uncertainty with those willing to yell the loudest trampling those around them.
I'll be watching curiously.
These "lead links" sound like managers with less authority to me.
If they are hoping for natural leaders to step forward then that seems how you become a manager in that company, you ask for it. Which, in some cases, is probably a better system than what many companies have in place that places people in roles they shouldn't be in.
The roles are still there, some of the requirements and responsibilities are shifted a bit, but they just removed the titles.
But the nagging feeling is what this Valve employee (who was fired) described: The company is led by the bullies just like in highschool.
One point that stood out to me, is that all firms face a constant, internal tension between the need to do "operational stuff" that actually works very well with a traditional hierarchical management approach, and the need to do "exploratory work" which does NOT map well to hierarchy. Failing to understand and manage this tension may be why many firms feel so dysfunctional.
The other point that stood out to me, is the idea that instead of a strict hierarchy - or complete undirected chaos - the best way to combine efforts towards both goals is by having individuals with a high degree of autonomy and empowerment coupled with a strong shared vision and common goals.
In this model, the primary purpose of leadership is to imbue the members of the organization with that "strong shared vision and common goals". Or to put it more simply, you tell people what needs to be accomplished, not how to accomplish it, and trust them to use their judgment.
On a related note, the book Adaptive Enterprise[2] makes a strong case for the idea of high decentralized teams, connected to each other through what the author calls a "Commitment Management Protocol".[3]
I would say that both of these books have some useful ideas that could be applied to construct a better management structure than what most present-day firms use.
[1]: http://www.amazon.com/The-Origin-Wealth-Remaking-Economics/d...
[2]: http://www.amazon.com/Adaptive-Enterprise-Creating-Sense-And...
[3]: http://coevolving.com/blogs/index.php/archive/conversations-...
A hub however will simply arise as the result of variance.
This. The need for management arose once we stopped being nomads and started to organize ourselves into collective groups. Look at Native Americans, even they had Chieftains.
What does that mean?
Working right now so I was in a rush and Indians were the first to come to mind.
"Yes, when a young man kills much meat he comes to think of himself as a chief or a big man, and he thinks of the rest of us as his servants or inferiors. We can't accept this... So we always speak of his meat as worthless. This way we cool his heart and make him gentle."
So the point is not to get rid of status difference (which would mean replacing human nature and thus be foolish) but to find organizational forms that are better aligned with how creative work gets done.