On Y2K Project Cost Overruns...

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

One thing to keep your eye one will be the increasing number of 'gee, some things have come up we didn't count on. I guess we didn't ask for enough money. Looks like we're going to need more to fix this' requests.

We will see a huge number of these in the coming months. Examples are already very numerous. If you doubt this, just ask and I would be happy to point you to several recent articles.

I'm seeing more and more of these and they are a good indication of the status of the projects. These cost overruns are not at all surprising - every large software project I've ever been involved with has cost significantly more than it was initially budgeted. The large the project, the more serious the overrun.

Consider the IRS's recent 11 billion dollar debacle - and the project never did get completed. There was way too much wishful thinking that went on there before the bitter pill was swallowed. Good money was thrown after bad long after the evidence was clear that their 'centralization' project was not going to make it.

And Y2K is by far the largest project I've ever seen.

I would recommend that management look long and hard at these requests and make the tough decisions. Is it really mission critical? Would the money be better spent on contingency planning and implementation? Do you stand a realistic chance of completing the project (remember the 80/20 rule)? Don't accept the word of your direct reports, talk one-on-one with the folks in the trenches (without their managers present). MBWA? How long will you allow the project to keep dipping into the well? What is the real risk of failure. Would you be throwing good money after bad or can at least some key projects be completed? If you've been managing large software projects for any length of time, you know exactly what I'm talking about. If you don't have significant large software project management experience, get advice from at least two people who have. I'm not urging you to 'pull the plug' if you've got a very good chance. But don't kid yourself either or be misled by unsupported 'don't worry, we'll make it' statements from within your own ranks. Put your limited resources to the best use for both your company and your customers.

Arnie

-- Arnie Rimmer (arnie_rimmer@usa.net), October 20, 1998

Answers

Excellent advice. If you are a senior manger, throw a dinner/party at a nice restaraunt with free booze. Invite only grunts doing the actual coding and absolutely NO SUPERVISORS. After dinner, circulate and ask simple questions,"strictly between you and me what do you think of XXX Y2k status." You will find out more in one evening than in a month of meetings. Do NOT betray any confidences and start berating your supervisors. Just readjust your resources and priorities. Been there, done that!

-- R. D..Herring (drherr@erols.com), October 20, 1998.

Moderation questions? read the FAQ