Tips for managers on on outsmarting the system when you require IT support Dear Bob …I’m probably not your usual reader. I’m not in IT. I’m one of IT’s “customers,” which is a joke because I get no service no matter how good my business case is.[ Also on InfoWorld: “How to balance centralized and decentralized planning” | Get sage advice on IT careers and management from Bob Lewis in InfoWorld’s Advice Line newsletter. ] Our company consists of a number of semi-autonomous business units. The company centralized IT a couple of years ago, “in order to achieve economies of scale” or so they said. I don’t know if they’re saving any money or not. All I know is that since IT was centralized only a few projects get approved and everyone else is out of luck.I’m a middle manager, two layers from the top of our business unit. How can I persuade the execs at headquarters that the decision to centralize IT was a mistake?– Middling Dear Middling … Why do you want to commit career suicide? That’s what it means to be a middle managers who tries to persuade top execs they made a mistake. Don’t even think about it.Here’s the situation: Companies centralize IT to cut costs. They often pretend they’ll be able to increase their agility at the same time, but they’re fooling themselves. It doesn’t happen. Companies increase agility by decentralizing, to put control closer to the action. That’s for your benefit, because the execs don’t want to hear this. And if you turn the situation into a complaint about IT, all that will happen is that you’ll create resentment, which will make the barrier separating you from what you need even bigger.Here’s what you can do: (1) Play the game to win; (2) volunteer to be a guinea pig; and (3) roll your own instead. In order:Play to win: From your account of the situation, you’re following the official project proposal process. That makes you a good corporate citizen, which is ManagementSpeak for “chump.” You have to discover the real process — the one that precedes the official one, during which the decision is made that’s rubber-stamped by the official one. Most likely, the real process involves schmoozing the key decision-makers to get their support one by one before they convene as a group. I’d also advise figuring out who among the decision-makers is most likely to want what you want, and engage their help in figuring out how to sell your idea to the rest.Guinea pig: Often IT has pet initiatives you can use to your advantage. For example, you might need reports and analytics IT isn’t delivering, while at the same time the CIO has been promoting a big, expensive, showy business intelligence initiative. Volunteer to be the test case, where you’ll be the one to showcase just how great it’s going to be — in exchange for being first in line, getting lots of training and support, and otherwise getting your analytics.Roll your own: No matter how good you are at the corporate game, you won’t always win. If you can’t, become self-sufficient. Devote a staff member or two to building solutions internally, using Access, SharePoint, or even Excel macros if IT won’t let you have anything else. It will be ugly, but it will be better than doing without.I’ve yet to see an organization that’s made fully centralized IT work well with semi-autonomous business units. Which simply means you have company, which supposedly is something misery loves. – Bob Careers