The culture myth
brandonhays.com
brandonhays.com
Free food is not corporate culture. Scooters is not corporate culture. Beanbags is not corporate culture. They are the natural expressions of an already-working culture, and will not lead themselves to one just because you've decided to build airplanes out of straw and bamboo.
Shortly before I left Amazon I was given a plastic (code) ninja action figure, which was, supposedly, some kind of new corporate mascot. We were all encouraged to take pictures with this ninja in interesting places, as some kind of "fun" thing. But of course, you can't order fun into existence, and the whole thing smelled very obviously of stuffy suits failing to get it. A fun, exciting company invents traditions like this, not vice versa.
If you work in a team of four and your manager hands you a toy ninja and says "let's take pictures with it!", that is cool IFF there is no higher management?
Or, is it cool only if a peer thinks of it?
Or, the issue is that the idea came from more people at a distance more than X?
Or the issue is that they told you that you must take pics with it, rather than saying "you know what would be cool" ?
Or is it just a general "this doesn't feel right".
Because it would seem to me that your company _invented_ a tradition like that, it's just that you didn't like that it wasn't invented from the guy sitting next to you.
I understand and have experienced the feeling of looking at BigCo and thinking "boy, they are trying too hard to seem fun", but I am wondering how much of the judgement is the _existing_ notion that they are a boring corporate.
FROM: Corporate Fun Squad <dl-funsqad@exchange.bigcorp.com>
TO: All US-WEST Employees <dl-us-west@exchange.bigcorp.com>
SUBJECT: Put your picture on Sharepoint!
At my current job, if my manager hands me the toy ninja, I would happily try to take pictures with it. Because my manager is very much knee-deep in the trenches with the rest of the devs, designers, and QA folk day in day out. He has, without fail, and repeatedly, shown that he understands the situation "on the ground" and has always treated our concerns with the respect and gravity demanded.
At Amazon there was not a chance - my manager was the guy barking orders from up high, funneling the whims of even higher management down with little to no filtering or defense for our ability to get it done. He was a straight pass-through, and it sucked. Him handing me the ninja would have been a damned insult (and it was).
The question is not job titles, it's entirely about the working relationship, and this is also why "inventing culture" fails. When you have an incredibly segmented workforce, where management is entirely separated from the grunts, and neither side have any respect for the other, any such "culture" is naturally a decree enforced from the top, rather than a genuine expression of what employees already feel.
> Because it would seem to me that your company _invented_ a tradition like that, it's just that you didn't like that it wasn't invented from the guy sitting
> next to you.
1) Your boss goes around and hands everyone a plastic action figure and says "okay let's all take pictures of him doing funny things!" Nobody's going to do it. It's going to feel forced, and it's going to have the opposite effect.
2) You come in one morning and find your boss taking a picture of an action figure bent over and peering into the guts of a server. Then you see a few coworkers doing it the next day. Pretty soon you're trying to top them.
Management can demand that people have fun, but the only way it works is if you observe it in action.
Its NOT cool if the manager gives you this stuffed toy because someone else told him it would be a good idea and its the new corporate direction
In short, the gesture is meaningless irrespective of who it comes from unless the person means it and its in keeping with his/ her character. i.e. not FAKED or FORCED
The line between cool and douche is driven largely by intentions (being in character v/s being manipulative)
Hope this helps
If you're happy at work and then you're generally having a good time and more inclined to do goofy stuff like this. If work sucks and you're stressed out then there's nothing more irritating than someone trying to paper over real problems with a hollow gesture like this.
Yes, but even if the reason for bucking the new "tradition" is spite or envy (and I don't think it is in this case) that's reason enough. The whole point is that culture emerges; it is not prescribed. It doesn't matter why it didn't emerge. If it didn't, it can't be forced.
You don't seem to get it.
The whole thing about culture is that it must not be _invented_, it must be _organic_ (grassroots).
That "it wasn't invented from the guy sitting next to you" is the WHOLE bloody point which makes it not a culture but some lame pointy haired idea in the vein of "this culture thing sounds nice, build us one by Wednesday".
Whereas the genesis of the idea may have been the result of the exact same thought processes either way.
Well, that's the very definition of grassroots, isn't it? If it's coming from "up the chain", it's not grassroots anymore.
>Whereas the genesis of the idea may have been the result of the exact same thought processes either way.
Not really. When it comes from the bottom, i.e the people actually doing it, it is not an "idea", it's something you do _spontaneously_, for fun, etc.
One is something that just happens --which can then be codified, repeated, passed on to new team members, built into a culture--, the other is something that is designed.
Note: The basic logistic equation is 1/(1 + e^-t) which one can extend to be S = M / (1 + e ^ (-rt-y)) where M is the maximum growth capacity, r is growth rate, y is number of years to the 50% mark. Anyway, it produces nice S-shaped growth curves that generally approximate growth in a limited environment.
Anyway working with these a few things occurred to me. Near the bottom of the growth curve, growth is very slow. It then picks up suddenly, grows to a certain point, then growth rate declines and the market size tapers off. Lack of immediate success along these curves may be due to a growth rate that is too low or a market size that is too small (duh) but when we vary both maximum size (M) and growth rate (r) something really interesting happens.
If we have a curve that successfully predicts two or three years of growth but then overperforms what we observe, this has exactly two causes: Either M is much bigger than expected and r is much smaller (meaning smaller growth now but much more growth later), or M is much smaller and y is much bigger (meaning the market is already saturated). You can't differentiate these just looking at how a project is performing.
A small business can bet against the short term and for the future. A larger, publicly held business is often less able to do so, so getting over the slow growth in the lead-in of the logistic sigmoid is really tough in a larger corporate environment. This is probably a good part of the reason why big companies have a much harder time innovating.
I was involved in "Agile" before the term had been coined, and there was a ton of meat there. But year by year the substance was leached away, until most of the consulting money was in, as you say, adopting symptoms. In talking with manufacturing people, the same thing happened with Lean Manufacturing. And I see it starting to happen with the Lean Startup thing. It kills me.
I originally thought this was driven by cynical assholes. Which I could have solved as soon as I figured out how to dispose of the bodies. But the problem is worse: it's all well-meaning people who just aren't paying enough attention to what's really going on.
They are, culturally, very much a big-giant-corp pretending like they are a small nimble startup. Everything from the "two pizza team" concept down to "frugal like a startup" (read: this is code to mean that your dev machine won't have been built in the last 5 years) is meant to evoke it, but really just elicits frustration and anger instead. Even in the "good" teams it will constantly feel like your boss is waging an epic battle against Evil just so you can have the peace to get your work done.
Fun note: circa 2011 my dev machine was a Celeron 1.5GHz. My codebase took a whopping 15 minutes to build even though a $1000 machine could have built it in a tenth the time. I knew incrediby talented, amazing engineers who were given 2 15" monitors dating circa 2002 and told to live with it. I'm not one for exorbitant gear, but this is entirely indicative of Amazon's internal culture.
I'm glad I cut my teeth at Amazon out of college, but I am extremely glad to have graduated out of both Amazon and Seattle, and have found much happier employment and residence elsewhere.
Of all the negative anecdotes I've read about Amazon, this is the most effective.
> glad to have graduated out of both Amazon and Seattle,
care to elaborate on Seattle?
It was a city that was wholly incompatible with me, and many others. I'm not going to come out and strictly say that Seattle is a bad place to be - because I know some people who are very happily situated there.
There were a few things that greatly bothered myself and many of my peers at Amazon/Microsoft:
- weather. I grew up in the Northwest, so this wasn't a big deal for me, but I know a lot of people were particularly annoyed at the constant overcast drizzliness. Note that Seattle doesn't actually rain very much, it's mostly just constantly damp, cold, and cloudy instead. Some time in September the sun will disappear behind clouds, and be completely absent until the following April. People in Seattle are extremely active and outdoorsy during the summers, but that's sort of like handing a starving man a bucket full of cheeseburgers.
- really limited social scene. Seattle isn't a convergent city - i.e., the vast majority of its population are locals, born and raised. Their social groups do not at all intersect with that of transplant workers who arrived later. There's a thinly veiled contempt by the locals for perceived transplant invaders driving up their cost of living (a belief that is IMO mostly true). The transplant demographic is extremely limited, and is almost entirely software people. These two factors put together means that 99% of your friends will be software people, and every party you go to will be full of software people. I had some crazier hobbies (read: hipster) that mitigated this to some extent, but even that wasn't enough. The social isolation and the homogeneous transplant population really gets to me.
Compare with where I am now (SF), where despite the dominance of tech, I still end up meeting tons of people from different walks of life just in daily life.
- transportation. I'm a pretty staunch urbanist, and while many neighorhoods in Seattle boast of the urban, carless experience, it doesn't actually pan out. Downtown, Belltown, Capitol Hill all try to execute on this lifestyle, but it's pretty half-assed - a lot of resources you will need are not accessible by transit or walking, and you are depriving yourself of the region's best food and sights if you go carless. Cars are not optional for a high quality of life in Seattle, even if you live centrally. Oh, your friends are talking about this amazing K-BBQ place in Federal Way? Welp, get a car. There's awesome dumplings up in Shoreline? Get a car. Want to do something other than drink on a weekend? Welp, car or you're hosed. Overall, King County Metro is the second-worst transit system I've ever used. It is somewhere between "ass-tastic" and "complete clusterfuck".
http://www.everywhereist.com/the-seattle-freeze-making-frien...
And a bonus http://www.everywhereist.com/15-things-you-should-know-about...
Identify what you think is attractive and a positive value and pursue that in 5 years time.
I had this experience with different big corporations, where department A is a really nice place to work and department B is awful... There are some shared things, of course, but it could be different enough to make a huge difference...
The trick is to figure out how to land yourself in the former.
I think a better and less link-baity title would have been "What culture should and shouldnt be". He didn't really disprove any specific "myth" and only really stated an opinion on how to implement culture. His two unsupported illustrations of bad culture ("Nerf guns and toy helicopters" leading to it being "hollow") were IMO not really myths at all just a setup for the rest of the article.
But if there's any meat to the "myth" moniker, it's in the growing notion that culture can be cargo-culted by bringing in no-track vacation policies or nerf guns.
The first thing is, you can't build culture from the top-down and if you try hard enough you will look like Kim Jong Il. That there is a corporate culture is automatic though and it arises from the everyday way people interact in a business.
In the first days of a business, the founders tend to be very flexible, willing to take on eachothers' roles as needed to some degree. There aren't turf wars. Just everyone working together to make the company great.
Then you hire more people and a problem develops. People feel a need to quantify and control everything. And so management develops. What management really is there fore is to be a political filter which exists in between the employee and the executives and ensures that nobody takes proper risks.
My approach is going to be somewhat different. First we are going with a relatively flat model inspired by WL Gore, Valve and Github, and where we have to add management and escalation, we make these as non-management-like as possible.
Our key values are likely to be openness, collaboration, collegiality, and entrepreneurship.
We will hire quality employees and get out of their way.
Executives will work side-by-side with employees on projects deemed to be of critical importance. Leadership --- and followership --- will be distributed throughout the organization.
This all sounds great until one comes back to the culture issue. Is it even possible to manage corporate culture as one climbs the logistic sigmoid? Or does maturity bring habits? In other words, is it true you can't teach an old dog new tricks?
People don't feel this at all, inherently. Organizations say, we will evaluate you by these criteria, and upon that evaluation will depend your promotion/bonus/opportunity to work on projects of your choosing/whatever.
As soon as you do that, people make damn sure that what they do is counted, and that is a perfectly rational response. And unless everyone knows everyone (and their work) personally and in detail, there's no other way to structure or scale an organization that wants to attract ambitious people.
I am not saying that quantification isn't important. It is even necessary. What I am saying is I am not so sure that top-down structures don't get in the way more than they help.
It can lead to important team members becoming subservient and afraid of speaking up. It can make other members change what they say to try and win points. And it can engender a culture of distrust.
But you are likely to still have those new and exciting initiatives, where a small group of people needs to work together to make it happen. So you might have an executive and four employees working together, hands-on in this.
But you are right to worry about employees being afraid to speak up. This is why it is important to have a culture of openness, collaboration, and collegiality, and where leadership and followership are distributed.
One other caveat: because of the market we are in we are likely to be hiring older developers with at least some self-employment experience. At some point if we start hiring younger folks I would expect the transition will be a slower one with a lot more mentoring. I would worry a lot less about all this with older developers who have been their own bosses and are clearly told we want them to continue to be their own bosses to the extent possible than I would folks without that experience.
Of these three, only "trust in its people" is relevant to the development of culture, in so far as culture means "a productive and fun environment that I/others would want to work at".
Organisational vision is important as it pertains to preventing myopia; tight feedback loops keep people on track - but a great culture is best achieved by hiring highly skilled people with a good sense of humour, tolerance towards the views of others, strong interpersonal skills, and perhaps a little bit of wackiness - and letting them do their jobs, whilst trusting that when they're goofing around having a laugh it will be positive for the energy of the company and not come at too great a cost to their work.
Edit: Removed emphasis on culture solely for attracting employees
Tight feedback loops not only keep people on track, they provide amazing feedback about what you are doing. That security in knowing that what you are working on matters and that you are doing the job well creates security. Nothing is more important to the morale of employees, in my experience, than feeling appreciated and secure in their work.
The same thing applies to vision. I've met very few people in my life who are ok with picking up a pay-check while working on projects that are failing for companies that are adrift. A company that lacks vision, no matter how amazing the team behind it, will inevitably be a terrible place to work.
At least in my experience. After all, it's not enough for me to really like everyone I work with. I need the satisfaction of knowing that I'm doing a good job (feedback) contributing to something of real importance (vision).
We treat our culture as something akin to a shippable technology product, complete with user stories, implementation, measurement, learning, reiterating, etc.
Anyhow: http://www.blackrimglasses.com/2012/02/16/culture-is-a-produ...
Elements of healthy corporate culture such as trust are easily traded off by the mob for the seemingly superior "accountability" of micro-management, for example.
And you see the same patterns in everything at a big company. Ever notice how internal tools in a big company are typically of very shoddy quality? That's because the pain of poor tooling is diffuse and spread out over time while the effort required to create better tools would be immediate and concrete (representing some opportunity cost to develop some shipped feature or such-like). And the same calculus applies to refactoring and re-engineering efforts.
The same dynamics play out in society as large as well. It can be difficult to stand up and fight for individual rights such as freedom of speech or the right to own firearms in the face of strong, highly focused criticism. And if such rights hadn't been enshrined in the US constitution they would be that much harder to retain. Drug use being a perfect example. It can be a serious uphill battle to argue for the right to, say, use heroin and that such a right will lead to a better society. The same thing applies to, say, "letting" employees come into work at noon and work a 4 hour day sometimes.