Does size matter?

I had a great chat with a friend last night where I mused that maybe the size of a company is inversely related to it’s effectiveness. When a company gets too big, it gets bogged down. Is there a solution to the problem, or are all companies just destined to become big, useless hulks?

Big is not so beautiful

Google’s had a few losses to Facebook over the past few months. Why on earth would that happen to a company that’s been rated best company to work for by Fortune magazine (for several years running)? Google offers it’s employees perks that most of us can only dream about.

“Entrepreneurs want to work at the hottest place on earth…” is a quote from one Google employee. But why? Don’t people like security? It’s not likely that Facebook is a “risky” place to work, but they certainly aren’t profitable, and burning through piles of investment money just isn’t a sustainable model. Google is a far “safer” bet for the money.

Do people really like safety though, or are they willing to sacrifice safety in order to get something more; something like importance?

People like to make a difference

People need to feel needed, and they like to feel important. Job satisfaction comes out of feeling like you’re making a difference: be it on a global scale, or just for the company you work for. If you stopped being there, would anybody notice? Are there enough layers of cruft (also known as middle management) piled on top of you that nobody nobody can see you?

When companies reach a certain size, more time needs to be spent communicating within the company and ensuring that the right people are working on the right things. I think that this level of overhead just continues to grow as more people are added, eventually, pretty much grinding things to a halt. You eventually hit a point where everyone is spending so much time managing and communicating that it’s a major achievement just to have something to manage and communicate.

I’ve seen this happen at places where I’ve worked. As departments grow, all of a sudden one person can’t keep a handle on everyone else. A hierarchy develops and everyone has to spend time communicating and clarifying and reporting. Basically, everyone becomes a manager, with less and less to manage.

What’s the answer?

One approach is to take the 37signals stance and stay small. Only grow when you absolutely have to. I’m really curious to see where this position goes as 37signals continues to release products and has to continue to grow the team. I think they’re managing to delay the onset of organizational paralysis, by only adding people when they have to, but if they continue to do more, they will inevitably grow. What will 37signals be like when they have 50 people? 100 people? They’re a smart group, so I’m sure they’ve given it thought, and will be ready when the time comes. They may be an interesting company to keep an eye on, because of their slow growth, we’ll be able to identify when that paralysis point hits (if it hits).

That doesn’t help those places that are already big though. Could Google scale back it’s operations? Probably. In any large organization, there are people that aren’t really all that effective. I can’t for the life of me find the quote, but I once read that one CEO decided that when someone leaves, they wouldn’t replace them immediately and they would see how things operated without that position. Honestly though, the could never shrink back to the size where every employee would kick ass.

If there’s a large company out there willing to get experimental, let me know. I would love to know if what I’m about to propose would, in fact, make a difference.

It seems to me that companies used to divide a lot, and in some places still do. Why is it that an oil and gas company can recognize a clear separation in it’s business and split into two distinct companies; and yet a technology company like Microsoft, keeps rolling unrelated businesses together under the same umbrella? Related, what’s with this attitude of technology companies that they have to be everything to everyone? Microsoft excels in operating systems and office software (and to some degree) hardware and they’re just not positioned to become a serious player in online search, advertising or even online applications. I don’t need a Microsoft-powered car, nor do I want to wear MS branded shoes. Stick to what you (were) good at, and spend your resources recovering from the steaming pile of crap that is Vista.

But would splitting a big company into smaller ones help to reverse innovation paralysis? Even if you don’t want to split up a company, in order to be able to move employees around (for example), would subdividing into AUTONOMOUS units help the situation? I think the trick, no matter what you do, is to make each company/unit hungry and in control of their own destiny. Simply paying lip service to it is not enough.

I worked for a department within a university for a number of years. When that department first started, we kicked ass. We had to, we worked on an entirely cost-recovery basis (we were a separate business running within the University, charging for our services). The only thing that the university provided to us was space and access to their infrastructure (we didn’t need to create our own payroll system as an example). We did have to cover all of the cost of each of our employees as well as buy any hardware/software we needed. Gradually, as we began to show that there was a demand for our services (and we billed other departments directly enough that they complained to admin about us), we became more and more funded by the institution. It started out as a couple of positions, then it was a core team. Then it was a hardware and software budget. Then it was a training budget. The point is, that this department became just another department at the university, and I think it changed. There was less hunger and less drive to do everything faster and better. People started to burn out once they recognized that they didn’t have to stay on their toes constantly. Paralysis STARTED to set in. I left before it became just another bureaucratic machine, and it may never become that. It wasn’t the atmosphere of the early days though.

Splits in technology

There is a classic example of this at work, and the results, unfortunately aren’t very promising. Palm the makers of handhelds split into Palm Inc. and Palmsource with Palm continuing work on hardware and Palmsource focussing on the Palm operating system. Unfortunately I don’t think that the result was that great. There’s too many factors at work to know whether the split is a cause of Palm’s decline, or just an unrelated factor.

Dangers

Getting back to the conversation that sparked this post, my good friend raised a very good point: this approach can be dangerous if all of the top performers congregate in one company, and all of the under-performers form the other. I think that the answer here is that any split needs to be tightly managed. It could be that with the split, certain employees need to leave, and new individuals brought in. I don’t think that one team needs to be weakened in order to “prop up” another team, but rather these need to be treated as distinct (smaller) companies, that should have the best people working for them.