A bit away from recent Xark topics, but "there are no unrelated topics", right?
A friend of mine recently sent some commentary on the future of Google. This provoked more of a rant than I usually indulge so I thought I'd post it here:
I'm amazed at how many "business experts" who write for the papers seem to want cool little startups to become IBM. I've worked at IBM. "I" doesn't stand for "Innovation", despite the number of patents they file. "I" stands for "Inertia". Inertia can be useful, but don't mistake it for something else.
As an example of their business acumen, these commentators write things like:
As an example, the company spends at least $14 per employee per day on all that free food. At 19,000 employees, that works out to $67 million a year, or about 20 cents per share that would drop to the bottom line if Google were to have the temerity to ask its workers - shudder - to pay for their own meals. Twenty cents is a trifle; analysts expect 2008 earnings of nearly $20 per share. But in a pinch? No-brainer.
So, let's analyze that: Let's say an average employee costs a fully burdened $85/hr. (That's a substantial but not outrageous fully-burdened salary+bennies in a small company in MA.. I'm sure the numbers are slightly different in Silicon Valley.)
Therefore, Google pays employees an extra 9.8 minutes per day. Do you think they make that back? Somehow I suspect they do. They might, actually - shudder - recover 20 minutes per day in travel to and from lunch. No, of course travel time doesn't come out of the company time -- they get their 8 hours anyway. But would they like 8.15? How about 8.3? It also encourages employees to eat together. It's amazing what problems get solved over food. Was that time work time or lunch time? If the company benefits it's certainly worth paying for.
At 20 minutes "extra" per day for $14, Google would be achieving an ROI of 102%. And this Super Genius says, cutting that is an "no-brainer". I agree, that would indeed be a "no-brain" decision, and I'm ignoring the whole "work environment" type benefits.
The other thing these guys like to insult is the whole lack of central control. They view it as "It's time to grow up, kids" and give examples of all the waste lack of control engenders. Obviously, the guys who are supposed to be in charge don't know what they're doing -- otherwise they'd obviously be telling people exactly what to do and when.
"You can't manage what you can't control" is quite true. But perhaps there are other forms of control than "allocation of resources" as taught first thing in MBA schools. Perhaps the people in charge at Google aren't actually *ignorant* of business management, they just disagree.
Perhaps a leader might engage minds and say "we're moving in that direction". Is that control? I think so. In fact, I think it's *more* control than allocating resources. If you get people enrolled in an idea and a goal you don't *have* to spend your time micro-managing. Think JFK's space race. Think of the X-prize competition.
In reality, the central planning style of control has some benefits and some drawbacks (I've recently been working with a bunch of Russians who came of age under the Soviet Union -- you should hear *them* talk about central planning). With central planning, you theoretically don't waste resources competing with yourself. It's easier to set and uphold some standards. Of course, you can often waste resources getting it wrong or following a central planners personal pet project or political agenda. As for standards, sometimes they're a good idea, sometimes not.
Non-central planning results in waste of resources, failures of coordination and self-competition, but it means that when you get someone convinced to move in a direction, they are doing it because they think it's the right thing to do. Besides, people doing the work (e.g. the Gmail team) might just have a wee bit more knowledge about how things work and in which direction they're moving than the central planning committee. Perhaps, once upon a time the wee little worker peons couldn't see the big picture. Now they can (ahem) Google it up.
Ultimately, if you look at companies as an ecosystem and apply some "survival of the fittest", projects under a "consensus" decision model might be starved by other projects, or killed or out-competed by other projects. That doesn't happen in a central control company -- in a central control company, the whole company is starved or out-competed by another company. That is, surely, a more efficient use of resources, right?
One thing these commentators glimpse, however, is how difficult it is to attract and retain good people as a company grows. They like to attribute it to a company no longer being dynamic and attractive. If a company has been following their advice, I'm sure that's true. However, a great deal of the problem isn't about "already done that" -- it is about sheer size.
At 19,000 people, Google is a statistical approximation of the universe. Variance from average is harder and harder the larger you get. Of course, if Google acted like every other company they'd be, well, average, right?
Realistically, I don't think Google can make incremental changes towards being a "normal" company. "Normal" is a different equilibrium point. Google has found another, non-normal, equilibrium point. If they move in small ways towards "Normal" they could seriously hurt themselves.
What Google has been doing for years -- and quite successfully -- is to create a blazing bonfire of creativity and productivity and then find a way to harness some of it as profit. They use Internet scale harnesses and you don't fill that with regional scale energy. The *have* to have a bonfire to sustain themselves. They can fail 3 ways: not being able to harness enough of the generated energy, not being able to generate enough energy to pull their harness, and the firewood budget growing too large for the harness to pull.
All of these projects -- even projects that "don't pay for themselves" -- contribute to the bonfire. Do you take away the wood at the edges just because it's not contributing to the huge flames in the middle? If you do too much of that you'll have a mighty sorry bonfire.
Commentators such as this would have them place tight cost controls on their fire-wood budget. If Google does that then they are dead -- there might be enough inertia to raise a new company from the ashes, it *won't* be the same company.
Google, intelligently, was offering a "blank check" for firewood -- as long as it keeps coming.
Control doesn't scale. It's the lesson of our age, as well as the lesson of democracy. You give up control, but you get back initiative and creativity and buy-in.
You clamp down and cut off and assert your authority and pretty soon all you're controlling are your own excuses.
Posted by: Daniel | Monday, May 12, 2008 at 13:50
Daniel - Or maybe control does scale -- into super-sized -- but results don't? Or a curve that reverses, where up to some point there are "economies of scale" but past that point the economies start canabalizing on the results that took them to that point?
Dewey - Wonder how the "business schoolers" would respond to your analysis? And I wonder what Google's turnover rate is? With 19,000 employees, it wouldn't take a high percentage leaving to look like a lot. Could be a real tribute to Google's hiring and employee engagement practices that they actually engage some of their employees in what they're doing so much that they want to go do it for themselves. What a model for entrepreneurial success they are! Probably anyone with an entrepreneurial bent who winds up in that culture becomes even more entrepreneurially inclined. Gosh, darn! Google certainly should control that.
Posted by: Rocky | Monday, May 12, 2008 at 14:17
I think control doesn't scale. The bigger the company, the more managers and resources they have, the less power they have to react quickly to changing markets; to get things done rapidly; to change internal, inefficient processes; even to monitor and control their individual employees (ask KBR about that).
I read somewhere that in many major companies, it takes an average of SEVEN people to make decisions. Even if each only took a day, that's a more than a work week.
In a world where you are mostly selling the same thing in the same way you did 20 years ago, you can have big lumbering corporations.
In this world, you can, oinly if you change how they operate. Which Google is attempting to do. The fact that "business experts" don't see this shows how far behind they are.
It's too bad, but the cluetrain done left the station.
Posted by: Janet | Tuesday, May 13, 2008 at 09:11