The M. State University is a large, public, metropolitan university located within 20 miles of four cities in Minnesota. (Scene: Kurt Wilson, Director of Administrative Information Management (AIM), is meeting with Paula Teague, Assistant Director of Applications Development.)
Kurt: Good morning, Paula. How's the cold?
Paula: Much better, thank you. It's good to be back. I assume this is the meeting I had to cancel when I got sick?
Kurt: Right. As you know, the administrative information systems master plan will be complete within the next three months. Assuming the executive committee approves the plan, the real work begins -delivering the new business processes and applications outlined in the plan.
Paula: I've been wondering when you were going to address that issue. We can't keep up with new systems development requests as it is. Am I going to get additional staff.
Kurt: I'm afraid not. In this era of staff downsizing, I suspect that we'll be lucky to hold on to what we have.
Paula: Well, I know we can increase productivity using CASE tools driven off the planning models your staff has recorded in the new repository. But there is a learning curve with CASE technology, as well as the new methodology. Also, these new applications call for a greater degree of adaptability and integration than we have historically expected. I just don't see how we can deliver more systems with the same or fewer people.
Kurt: I've got an idea. I've been running some numbers against the time accounting system. According to our own records, we are using 19.3 FTE (full time equivalent staff) to simply support existing systems maintenance.
Paula: That wouldn't surprise me. Legacy code is the anchor that inhibits new systems development in all shops. Don't tell me you are going to eliminate existing systems support? I think there would be an immediate and fatal backlash from the user community.
Kurt: True, but that's not exactly what I had in mind. Don't have a cardiac, but what would you say if I told you that I wanted you to consider reducing your maintenance effort to 8.5 FTE? [Paula does not respond] Your silence indicates that you are concerned.
Paula: And rightfully so, don't you think? The user community will scream for my head on a platter! You are talking about cutting support by more than 50 percent.
Kurt: Actually, Paula, I'm asking you to cut support by less than 25 percent, but to use 50 percent fewer people! Paula And how am I supposed to do that?
Kurt: I have a couple of ideas for you to consider. First, according to my analysis of change request forms, almost two-thirds of all requests fall into the category of enhancements. Half of those enhancements can be characterized as "desirable," not "essential." It seems to me that we could declare a temporary moratorium on such maintenance projects.
Paula: I can't confirm your numbers, but I'll agree that we are going to have to be pickier about what we choose to (do) & not do. I'd like to see your data after this meeting.
Kurt: No problem! Second, I'd like you to consider a SWAT team approach to maintenance.
Paula: SWAT? Like the police?
Kurt: SWAT stands for "Specialists With Automated Tools." With only 8.5 FTE for maintenance, it seems to me that it would be a mistake to assign one or two persons per existing development team. Instead, I see a maintenance SWAT team that takes over all maintenance.
Paula: It might work. I'd want to make sure the SWAT team had at least one member from each current development team to preserve application knowledge. But what's the "automated tools" angle?
Kurt: Glad you asked that. Our luncheon meeting will be with a sales representative from a company called VIASOFT. They sell what amounts to CASE tools for maintaining, enhancing, and re engineering our existing software programs. They promise to increase productivity of maintenance programmers who specialize in the tools.
Paula: Now we're talking. There would be the usual learning curve, but once the team is comfortable with the technology, we might just be able to do 75 percent of our existing maintenance with less than 50 percent of the existing staff.
Kurt: I think so! In fact, I'll make a bolder prediction. In time, I think you'll eventually be able to increase support over today's levels with less than 50 percent staff. Anyway, you're a pretty creative person. Give some more thought to ways we can make this work. We need to go to our luncheon date.
a. If "essential" enhancements are submitted, what process should be followed to select those that will be implemented?
b. Can you think of any aspects of systems support that have not been addressed by Kurt's SWAT and technology proposals?
c. What additional analysis would you recommend to Kurt before making a final decision on reducing systems support staff?