Effective CIOs know when a project needs to be marked 'not for resuscitation'.

A young man lies on the trolley, deathly pale with suspected internal injuries. An elderly lady sits quietly in a wheelchair with a clearly broken ankle. A pretty young woman rests her bald head on her husband's shoulder - the chemotherapy has left her prone to infection and she's running a suspiciously high fever. An abusive drunk with a deep gash on the forehead lurches at a porter.

It is just another Friday night in accident and emergency. The dilemma the doctors and nurses face each shift is, which to treat first?

Performing triage on information systems rarely has the drama of an A&E department, but for many CIOs it can be an equally challenging task. For some, triage is the first thing they have to perform on accepting a new position. They have been called in to fix up the systems - pull the plug on projects that are flailing or not aligned with strategy, prioritize those that need extra urgent attention, and keep the others ticking over until more resources are available to divert to them.

David Miller joined Investa Property as the manager of business planning - a role that has oversight of the information systems - a couple of years ago. "It was a fairly young organization which had grown quickly since inception. There was a need to do triage then," he explains.

Miller's route to the CIO seat at Investa is typical of what happens at many organizations when their systems are sick and need attention: He was hired initially as a consultant. "I was at Jones Lang La Salle and was hired to come in and do a review of the systems on a contract basis." He believes that the benefit of coming in as a consultant when there are problems to be tackled is that, "it gives you a broader depth of experience and the scope in geeing things up or winding down processes".

It was exactly this approach that fresh produce company Moraitis took when it hired consultant Con Colovos to perform triage on its sick information systems and replace the inadequate TIMS system (see CIO, June 2004). Like Colovos, Miller performed the initial triage as a consultant then was appointed permanent IT head. Having rescued the valuable IT systems and euthanased those without merit, Miller went on to develop a detailed strategic plan for Investa for the following two to three years.

Now to a CIO a strategic plan is like a doctor's syringe of penicillin; you need it but it may not always successfully treat the patient. No use injecting penicillin in a kid with a broken arm. As Miller notes, having the strategic plan was one thing, "but you need to be reactive to the changing business environment. If you're not then you will find the business units develop their own systems." The strategic plan may need regular tweaking or even wholesale rewriting as business conditions change. When there is no time to do that the CIO simply has to ignore what it says in the plan and apply common sense, shutting off resources to one project and diverting them to one that is more pressing.

And what about the business units that yell loudest? Isn't there a chance that like the abusive drunk in casualty you will treat them first just to get them off your back? "There's a degree of that," agrees Miller. "The squeaky wheel does escalate their issue to the forefront of the agenda."

What mitigates against the squeaky wheel syndrome are rigorous governance processes, which mature IS organizations are adopting.