Stories about Software


Defining The Corporate Hierarchy

Rites of Passage

Think back to being a kid, and you can probably remember a rather dubious rite of passage that occurred when you figured out that you weren’t going to be a sports player, lead singer, or Hollywood star.  You probably felt sad, but your parents and older siblings likely breathed sighs of relief that you’d never be explaining to people that a manual labor gig was your “day job.”  State lotteries notwithstanding, giving up on improbable dreams is considered by adults to be a sign of maturity in budding adults.

If you think about this, the easy message to hear is “you’re not going to be great, so give up.”  It’s depressing and oft-lamented by college kids having mini crises of identity, but it’s actually a more nuanced and pragmatic message, if a poorly communicated one.  It’s that the expected value of these vocations is horrendous.  For baseball players, actresses, and rock stars, there’s a one in a million chance that you’ll make ridiculous sums of money and a 999,999 in a million chance that you’ll make $4,000 per year and have half of it paid to you in beer nuts.  So the expected value of going into these positions is about a $4,200 per year salary and a handful of beer nuts.  So the message isn’t really “give up because you’ll never make it” but rather “steer clear because anything but meteoric success is impoverishing.”

The better play, we tell our children, is to head for the corporate world where the salaries range from minimum wage in the mailroom to tens of millions per year for CEOs of companies that create stock market volatility. Most importantly, you can find every salary in between.   So if you aim for the heights of CEO and fall short, mid-level manager making $140K per year isn’t a bad consolation prize. And so a funny thing happens. We consider it to be a rite of passage to abandon the delusion that you’ll be Michael Jordan, but we encourage the delusion that you’ll be Bill Gates until people are well into middle age.

That’s right, “the delusion that you’ll be Bill Gates.” You won’t be him. You won’t be a CEO, either, unless you pop for your state’s incorporation fee and give yourself that title. You’re about as likely to “work your way up” to the CEO’s office over the course of your career as any given child is to luck into being the next multi-platinum pop star. So, it’s a rather strange thing that we tsk-tsk children for indulging pie-in-the-sky fantasies past a certain age while we use nearly identical fantasies as the blueprint for modern industry. Kid wants to be Justin Bieber? Pff. Thirty-year-old wants to be Mark Zuckerburg? Keep working hard, kicking butt, and acing those performance reviews, and someday you’ll get there!


A Cynical Theory of Management

Let’s get really cynical for a moment – more cynical, in fact, than my actual views on corporate politics. We’re going to take a walk past some pretty snarky theories of management.

This is a drawing by a cartoonist named Hugh MacLeod that’s ingenious in both simplicity and cynicism. Drones at the bottom, ruthless manipulators at the top, and a creamy center of interesting folks in between. These are the ones that think they’re going to make it to the top. They won’t, and they’re clueless to that fact. They’re the kids that still think they’re going to grow up to be Halle Barry.

Venkatesh Rao took this cartoon, combined it with the television show The Office (American version), and created an ingenious and beautifully cynical theory of corporate management that cut to the core with its incisiveness. The “Gervais Principle,” as he coined it, refined the Dilbert Principle, which was itself a refinement of the Peter Principle. The two Gervais Principle predecessors were cynical to the point of comedy whereas the Gervais principle is cynical but accurate to the point of tragedy.

The Peter Principle holds that people are promoted until they prove incompetent in their role, and then they remain there; competence is rewarded with promotion and incompetence is rewarded with the status quo. The Dilbert Principle, with more of a knowledge-worker focus, rings true to those of us who have seen terrible programmers promoted to project managers. It states that bad employees are promoted into management to prevent them from doing damage with their incompetence. The Gervais Principle gives a lot more credit to those at the very top (which, in my opinion, makes it far more accurate in its reasoning about corporate leadership); it says that the sociopaths that run the organization knowingly over-promote dedicated but relatively inept people into middle management.

Why would they do this? In middle management, these clueless folks serve various ends for the sociopaths, but the most important ones are “foil” and “buffer.” As foils, they can be cannon fodder on projects with low chances of success and blamed when things go sour. The buffer play is a bit subtler. The sociopaths that run the company have power, influence and lives that make most people jealous, and the losers at the bottom rungs of the corporate ladder are line level employees resigned to a relatively powerless lot in life. They put in just enough effort to remain in good standing at work and remain aware that their employment is a pretty bad deal for them and a pretty good deal for the sociopaths at the top. A lot of direct interaction between the executives and the rank and file would quickly lead to resentment.  So the sociopaths over-promote a handful of losers who put forth disproportionate amounts of effort.  These former losers enter the clueless ranks of middle management to act as a buffer. The losers can’t really hate the clueless because the clueless aren’t calculatingly taking advantage of them. The clueless believe that they’re on track to be CEO while the losers and the sociopaths both know that’s absurd. In the show, Michael Scott represents this archetype – incompetent, fanatically loyal to his company, and clearly not headed for the C-suite, whatever he might think.

If you want to really conceptualize this and have it driven home, consider a hypothetical scenario as follows. The CEO of some large organization with thousands of developers decides to hold a mandatory contest to see who can write the ‘best’ web application, whatever ‘best’ means (left vague intentionally). The contest will be done on the developers’ own time, but the winner receives a $50,000 bonus and a promotion to CTO. Picture what happens next.

A solid majority of the developers roll their eyes, spend an hour implementing some hello world kind of thing because it’s required, submit it and forget about the contest. A relatively sizeable minority heads in the complete opposite direction and goes absolutely nuts with this thing, certain that they’re going to win that sweet, sweet prize. These developers pour hundreds of man-hours each into it over the course of months, completely losing sight of the fact that they’re each individually contributing tens of thousands in free labor. Who wins? Well naturally, it’s the developer who figured out that his sister is friends with the CEO’s favorite nephew, parlayed that relationship into favorable treatment, and then plagiarized some web app from GitHub.

What’s the fallout of this? The losers always understood that the contest was pretty hokey and probably too good to be true, so they roll their eyes at the CEO and his nepotism and figure it’s business as usual. The clueless are disappointed, but they know that the best, most qualified candidate won. They had their chance but just didn’t quite work hard enough. They vow to work even harder next time, and the company sells their free labor for millions, earning fat performance bonuses for the sociopaths at the top. The sociopath who cheated earns a seat in the executive room.

Do the losers resent the C-levels? Sure. Do they mutiny? Not if the clueless are promoted into a role above them. The clueless so genuinely believe in the organization and its wisdom that it’s impossible for the losers to hate them. What they feel for them is a mixture of pity, disgust, and occasional gratitude (if they happen to be nice or generally benevolent in application of power), but not hate. The losers satirize them in cartoons with pointy haired bosses and they gossip about them around the water cooler, but because of the clueless buffer, they don’t collectively revolt and go out in a blaze of spite.

In effect, the clueless create two different organizations within one organization. There is the organization of losers and clueless where putting in 60-hour weeks and being obsequious lets you claw your way up a few of the bottom steps of the pyramid. And then there is the organization of clueless and sociopaths, where putting in 60-hour weeks and being obsequious keeps you right where you are with that next level of advancement always being oh-so-close-but-better-luck-next time. Creating the bottom level organization, where tripping over yourself to provide free labor is rewarded with small stakes promotions, allows the top level organization to sustain a model where the losers and clueless get terrible economic deals and keep coming back for more. The clueless have no idea this is occurring and the losers understand it but have no appetite for rebelling against their clueless managers who are answering emails at 3 AM and working 60-hour weeks.

Ratcheting Down the Cynicism and Redefining the Terminology

Venkatesh’s analysis is wonderful, and you should read the posts and buy the e-book. The sections that describe how these archetypes deal with one another, in particular, are absolute goldmines of strategic understanding of corporations. But I have three main problems with using the archetypes, as described, to elaborate on my own theories of corporate politics: the names themselves, the assertion that over-performing middle managers are generally idiots, and the placing of corporate citizens into one of three buckets on the basis of assigning them serious shortcomings.

Specifically, the losers are some mix of lazy and cowardly, having given up on the idea of controlling their own destinies. The clueless are idiots that don’t understand the nature of their relationship with the organization, and the sociopaths are ruthless users and manipulators of other people. All three archetypes are mainly defined by their core flaws. I go into a lot more detail about this in the introduction to the book that I’m writing, but I prefer to think of them not in terms of their shortcomings. Instead, I think of them in terms of what the modern corporate structure has done to them – broken the losers, tricked the clueless, and forced the sociopaths into ethical conundrums. I don’t agree that the corporate structure is optimal or inevitable, and I think its deep flaws show themselves through the human beings that execute its various rituals.

I can’t solve all of that in a blog post (and, of course, in the book, I just offer ideas for solutions), but what I can do here is rename the archetypes in more relatable terms. I’ve spent a lot of time in a lot of different organizations of different sizes and domains, and I just don’t run across cartoonish people like Michael Scott and Dwight Schrute. By and large, the people in these organizations, at all levels, are relatively well-intentioned, reasonably intelligent and doing the best that they can on the micro, day-to-day level. Corporate structures are, however, substantially less than the sum of their parts, so good faith efforts in the small are perverted into rampant dysfunction writ large across the face of industry. Organizations are pathological, as Venkatesh points out, and they are pathological in a way that corrupts their components.

To be specific, I propose that we name the losers, clueless and sociopaths to “pragmatists, idealists and opportunists.” Their roles, relationships and dynamics remain the same, aptly described by Venkatesh. But this softens the blow a bit, particularly in how I think of them. Pragmatists are line-level employees who find value in life outside of work, mainly because the hope of any meaningful advancement and enjoyment of their profession has been taken from them. Idealists believe heartily in the meritocratic company (and organizational superiors) as a benevolent steward of their careers because perspective has been taken from them. Opportunists refuse to yield hope or perspective and recognize that the only way to win the corporate game is to play by their own rules. In this realization, they give up ethical certainty and human connection – opportunists play a lonely, sad game to get what they get.


But, as I mentioned, the dynamics are not altered in the least. Pragmatists contribute as little as possible to preserve stability, getting a bad economic deal and recognizing it (indeed, my salary negotiation hacks post was basically addressed to pragmatists as a guide for how to start being opportunists). Idealists, believing in the company, work even harder and make their economic deal even worse. Don’t be fooled by a slightly higher salary and meaningless perks like offices and parking spaces – working 50% more your entire career to eventually get paid 15K more per year is an abysmal deal, compared not only with opportunists’ deals but also with minimum effort, lower-wage pragmatists’ deals. And the opportunists feeding the grist to the mill certainly over-promote idealists because of strategic necessity rather than any kind of merit. Even with different labels and humanized, sympathetic consideration, the show must go on. It just might be easier for you to see what role you have when they aren’t all described so negatively.

So What, Erik?

This is sort of a depressing note on which to end, I suppose, but I was really just trying to establish my preferred taxonomy of corporate citizens so that I can talk further about them in subsequent posts and refer you back here to help you understand the backstory. For instance, look soon for a post about LinkedIn buffoonery about how to get ahead that really should read “An Idealist’s Guide to Meaningless, Token Promotions” and includes retch-inducing advice like “Dress For the Position You Want.” It will become a lot easier to understand why this type of thing is stupid when you understand the true anatomy of the corporate beast and we have a common, shared vocabulary about it.

As an example of how this vocabulary would help, let’s briefly reimagine the salary negotiation hacks post. It’s an opportunist play to hire you to do work that is worth $150 per hour and to pay you $50 per hour for that work. Putting in 60 hours per week instead of 40 with the hope that, in a few years, you’ll be making $60 per hour is the epitome of an idealist (clueless) move. Don’t do that. Think like an opportunist. Figure out how to work less for your 100K per year, not more.

If you’re still with me and reading at this point, I thank you. I have the somewhat romantic goal of defining, via this blog and my upcoming book, a way for us to create organizations that do not require the concessions currently required of the pragmatists and idealists and to do away with these Gervais Principle distinctions. I think we can form organizations that consist of nothing but benevolent opportunists if we more closely align our work with actual, monetary (or other) value provided to clients/customers. So buckle up, because the posts and book in this vein are going to make for a weird, bumpy ride.

  • dave falkner

    I will look forward to reading the series.

  • Aaron Duty

    This corporate population breakdown reminds me of my previous employer and the insanity I witnessed there. I don’t miss it at all.

    • Once you step away from being in the middle of games like this, the different is pretty notable. Makes me think of coming in from frigid weather and only realizing as you’re warming up how cold you were.

  • Noam

    very interesting and I feel the same way

  • yogamini

    I might not be able to put it as eloquently as you did, but I have this very faint idea of how the clueless can be completely eliminated from a corporation changing the composition of the corporation to comprise solely of skilled opportunists who reap the benefits of their effort in proportion to the value they provide to the end product or service that the corporation survives and thrives upon. At this point I don’t know how this value could be measured but I know for sure that this value is not going to be a function of time, but will probably depend upon unbiased customer feedback, peer to peer feedback, etc. Also, decomposing the market into smaller segments is one solution where a large corporation can be converted into multiple smaller businesses comprising solely of skilled opportunists who are not necessarily sociopaths, who, instead of exploiting each other, focus their energy in bringing in more business, thus building a high quality product or providing a high quality of service.

    • This is all very relevant to the path that I plan to pursue to some extent with the blog and to a great extent my book. There’s nothing here you’ve listed that I disagree with, from focus on value to seeking out the right kind of feedback to decomposing juggernaut organizations.

      I think that an enormous amount of the nonsense that we put up with in the corporate theater is the result of organizations that have no idea whatsoever how to value employee contributions. “How much more profitable are we thanks to employee X? Oh, no one has any ideas? Welp, then let’s just do a performance review focused around how some random manager thinks they’re measuring up agains the ‘company values…’ Yeah, that’ll probably mean something. We hope.”

  • MrTonyD

    I started reading this thinking it was “tongue in cheek” – but the resonances with my experiences in Silicon Valley are just too real.

    Although I’m not sure I’d give up on the “sociopath” terminology right away – I’ve been hired by three multi-billionaires in my career, and I would say that they really are sociopaths. They played by their own rules because they could – wealth does that.

    • Personality-wise, those people you’re describing sound like psychopaths (in the clinical, non-empathetic sense of the word). In my dime-store psychologist cartoon of the world, sociopaths are those who feel no obligation to conform to social norms while psychopaths are a narrower set that also lack empathy and the basis for normal human interaction. From that framework, sociopaths might be iconoclasts that are otherwise decent people.

      In any event, I’d say that opportunists tend to be sociopathic in the context of the rules of normal corporate work — they reject the notion that the path to the top is, “pay your dues, put your head down, work hard, wait for recognition.” Some who do that might be societal sociopaths or even psychopaths (as sounds like the case with these multi-billionaires), but not all, I don’t think.

  • I’m glad to see somebody working to make Venkat Rao’s work more consumable. He’s very, very smart, but “popular appeal” isn’t his strong point.

    • I love his stuff, myself, but it did always strike me that, when I’d try to explain how awesome it was to other people, they often kind of looked at me blankly. He once described that as “a feature, not a bug.” 🙂

      • That sounds like him. He’s tremendously elitist. Which is also presumably, to him, a feature and not a bug 🙂

  • Kurt Guntheroth

    This is so clear and lucid. It explains so much. Only problem is, now I have to fight the urge to open a vein…

    • Glad it makes sense to you, and I understand the initial bleakness. But, for what it’s worth, I think there are better options ahead. The rise of the so-called “gig economy,” for one, should be a help.

      • Kurt Guntheroth

        Wish I felt that way. The gig economy puts labor at a tremendous disadvantage. It allows employers to put off the cost of slack time onto labor. So far, there has been no compensating increase in wages. Labor has no power to negotiate pay or working conditions if all labor is fungible. Sounds like a recipe for 21st century serfdom to me.

        The gig economy will be (actually already has been) great to rock stars and home-run hitters. It might be tolerably ok for sufficiently skilled workers. It’s gonna really suck for most workers, plus anyone who fails to maintain cutting-edge skills over a 45 year working career.

        • I mean that the gig economy will help in the sense that the pyramid organizational structure I’ve described will start to dissolve in the face of this new reality. In the software industry, though, I don’t think gigs will be taken on by individuals fending for themselves, but rather by teams (firms) of professionals that defray risk, cost, and slack as a cohesive unit. That’s the short-form, anyway — long form is what I’m working on in my book 🙂

          • Kurt Guntheroth

            I’m a recent arrival to your blog, so forgive me, but I don’t see anything in the gig economy that is likely to disrupt hierarchical organizations with opportunists at the top and serfs/pragmatists at the bottom.

            I think you expect software folks to become like lawyers, civil engineers, and accountants, working for firms that pimp them out on projects. Unexplained is why these firms won’t have the same hierarchical nature as the enterprises that consume their services. In fact, if you know anything about these firms, you must realize that the partners are particularly exploitive of the junior workers, holding out the implicit promise of making partner in exchange for tremendous amounts of uncompensated overtime. Few junior staff ever actually make partner.

            An improvement would be a model that allowed the pragmatists to disintermediate the opportunists. It would have to be something like a collectively owned job dispatching web portal where the workers set the agenda. Something like a professional union perhaps. It would have to be something we haven’t seen yet. This would be nothing like the highly exploitive Uber model, which is firmly in the hands of the opportunists.

          • For existing companies, arrival of “gig labor” in force removes pragmatists from the hierarchy (which obviates idealists). If WidgetCo can only hire labor by paying freelancers, there are no pragmatists in the organization, ipso facto. (I’m not saying this necessarily solves any problems — it just removes the dynamic from the company)

            As for the problem of firms creating up or out pyramid structures (ala “high powered, tough culture” law firms), you avoid this by not trying to scale big enough that exploitative dynamics are required. This is, again, pretty short form, but the picture I’m painting (and the thing for which I’m advocating) is that developers form teams with which they’re comfortable and then work on gigs as a unit. This allows them to defray the risk and jack of all trades problem of being a solo free agent, but it doesn’t invite the toxic, rent seeking behaviors that inevitably emerge as things scale past the simple and intimate.

            (I realize that this paragraph isn’t sufficient to be a well defended argument — it’s just a teaser for what I perceive as a decent way forward)

          • Kurt Guntheroth

            I am drawn to the idea of the client enterprise being stripped of pragmatists and idealists, leaving the wolves to eat one another. Gig-economy firms servicing such clients might be able to price their services based on the value added rather than the cost of goods, sucking value out of the pockets of the opportunists instead of the other way around. Bwah-ha-ha.

            An engineering firm consisting of a single team is critically weak; it will either all be working or all be idle, making the best-case assumption that every team member can do every job. This is essentially the same problem as the solo free agent, but with higher costs. Economy of scale will pressure this firm to grow until it can take on multiple projects, so to efficiently allocate all its workers. It will grow efficient, specialist marketing, accounting, and HR staff, and come to resemble every existing enterprise. It is reckless to assume that even a small firm won’t grow an opportunist to reap the lion’s share of the reward, human nature being what it is.

            The solution is to design out the opportunists. I can’t say exactly what approach would be most effective, but I imagine an organization that starts out with altruistic goals building a system for providing workers, allowing the workers to form teams within the organization, and with a structure that directs compensation to the workers, rather than to an opportunist class. Perhaps each team agrees how to distribute its compensation, possibly using guidelines established by the organization. Teams would rate workers as managers, contributors, etc, creating a reputation-based economy. Opportunistic workers would ultimately be identified and cast out, unless it turns out that opportunistic sociopathy is a required leadership skill, in which case they most effective or most humane ones would still be identified.

            The organization I’m postulating might be a not-for-profit. It might arise from a union or guild. It would require altruism or mutual-self-interest at the start, but once the structure was in place, its freedom from opportunists would make its cost structure lower than a firm run by wolves. In this way it might be self-sustaining even in the face of human nature. Open sourcing its operating software would make it harder to subvert the organization, because a competitor could be quickly constituted if that ever happened.

            The same operating software could be used to run any specialist organization; an accounting or law firm, design shop, HR outsourcer, etc.

            In the best case, the existence of such firms would cause a phase-change in the whole business landscape, efficiently allocating resources as a free-market economy, while giving more of the profit back to the people who produce the work. Traditional opportunist-led companies would have to work very hard to prove they added any value.

            I’m laughing at my desk as I try not to use phrases like “overthrow the ruling class” and “workers control the means of production”. There is no Marxist motivation behind this idea, just a way to get more money to the productive resources, and give less to the drones.

            Now, how to get initial development funded…

          • Mike Rippon

            I’m amazed that co-operatives aren’t more common in the software industry. The values of providing long-term sustainable employment (worker cooperatives) & value to society (consumer cooperatives) seem to be much more closely aligned with the objective of providing good quality maintainable software that benefits it’s users, than those of a for-profit company, aka “ship this thing next week and we (*I*) can make a quick tidy profit”

            Are developers just so caught up in the “build a career” game that they’ve never stopped to consider the alternatives? I know I was until a year or so ago, and it’s only after 6 years in the industry and reading this blog that I’ve begun to seriously consider the alternatives (thank you for that by the way!)

          • Kurt Guntheroth

            I think the degree of altruism required to start up a cooperative is a high barrier. You have to work just as hard as if you started a for-profit business, but there’s no big payoff at the end.

  • SuperHappyCow

    Eh? This is really fascinating. Need to read more of your stuff.

    • Thanks — glad you like, and feel free to look around!