Guy Kewney

“I think,” said the senior IT support guy for a large high-tech manufacturer, “we techies are going to be largely immune to the effects of the credit crunch.”
It would have been really nice if he had then changed the subject. Instead, he turned to me and said: “What do you think?” ­ and insisted on getting the discussion going.

That was embarrassing, because I have been caught this way before. What I can tell you is that I will get it wrong. Whatever you say about the future will be wrong ­ that’s a given.

I know what he meant. Simply put, working in his office, helping customers support a new generation of network equipment, helping them install it then gathering feedback and passing it on to the design team, is a function that cannot be outsourced.

That is not to say that some people won’t try to outsource it. But I suspect they will fail.

The reason I think they will try has nothing to do with the business case. There is a strong strand of simple bullying involved in managing IT staff. There are plenty of people who could stand in for Dilbert’s pointy-haired boss (PHB) without the benefit of makeup. They would come up with the idea of bands around the head that administer shocks when the PHB presses a button. “To make us work better?” asks Dilbert. “That’s one theory, sure!” says PHB, who obviously does not care, as long as he gets to press the button.

They bully IT staff because they fear them ­because they do not understand their arguments. “We need three new staff,” says the systems administrator. “Maybe we do,” thinks the PHB, “and maybe he just has some friends who are out of work. I can’t tell so I’ll just say no because otherwise he will think that I respect his opinions.”

My answer to my high-tech friend was: “Maybe. I hope so.” Part of my reason for optimism is that I honestly think outsourcing mania has become a mini “bubble” ­ not on the same scale, perhaps, as the housing or dot-com bubbles, but a bubble, nonetheless. The problem I perceive is one of scaling.

Recently, asked to provide a soundbyte for an IT programme on TV, I found myself onscreen commenting that yes, cloud computing did raise issues about “what do we do if it fails?” and what I actually said, on television, was: “It shouldn’t happen, but neither should metro electricity fail; and you don’t half have to throw away a lot of rotten food when it does.” I said it, and it is a point worth considering. But it is not the main worry, which is capital expenditure.

I have seen several cloud-powered services on Web 2.0 sites. They follow a clearly discernible pattern: They start off brilliantly. Then they succeed beyond expectation. Then they slow to a crawl. Then there is a period where you can almost hear the wails of “But you’re our biggest customer! We can’t quadruple our equipment purchase budget just for you, unless you commit to at least a year’s business” as the cloud company discovers that it has just one client which uses 90 per cent of its resources.

The reason for the cock-up is that the cloud company does not have the insight needed to sensibly forecast growth. And when it comes to customer feedback, you have to know what questions to ask.

I think we have reached the point where it will be easier to hire out-of-work IT people than to get an outsourcing firm to do what is needed, when it is needed. We should not suggest that having IT staff on the premises is the same thing as knowing what questions to ask because I know plenty of management types who could keep them onsite, and still make them work in a dark cellar and tell them nothing. But in a competitive environment, maybe such idiots will be exposed rather more quickly than otherwise.

I hope so.

Like this story? Spread the news by clicking below:

 del.icio.us     Digg this     reddit!