Making hard decisions

DESPITE ALL THE attention given to CTOs who design and build new and innovative systems for their companies, a lot of what a CTO does these days is look at existing infrastructure and decide what can be removed from service without harming the business. During the IT boom of the past few years, a lot of hardware and software was thrown at a lot of different problems, some much-needed and others somewhat frivolous. But as staffs became leaner and resources became increasingly tight in the past 18 months, none of this infrastructure got up and left the building with the staff and the budgets (at least for companies still in business today). It had to be dealt with somehow.

Free IT resource Hear how top CIOs turn change into a competitive advantage. Sponsored by HP Free IT resource Try Sun servers, workstations and storage products free for 60-days. Sponsored by Sun Microsystems

Many CTOs find themselves in the tough position of having far fewer people to maintain this infrastructure, yet demand for core services such as bandwidth, storage, and others continue to rise. Traditional wisdom might dictate that you outsource these functions to a third-party service provider, but outsourcing is no panacea these days either because, guess what? Chances are that the folks working at the service provider are short-handed and just as stressed out as you are. Lower revenues lead to leaner budgets and fewer people, but IT demands continue to defy the laws of gravity -- they never go down. If you are a CTO working with a budget that is 50 percent of last year's, it's a sure bet the CEO of your company didn't agree to let you shut the network down 50 percent of the time and cut back the hours of Web site operation to 9 a.m through 9 p.m.

The issue becomes how to determine what services to suspend. Making that determination is often more art than science. At InfoWorld, I have made the decision to pull the plug on some services we were running simply because no one could give me a sound business justification for why we were running them and draining valuable resources. Despite the fact that these systems were not mission-critical, we were still getting bills for maintenance contracts and our smaller IT staff still spent valuable time maintaining these systems. After walking through the office and talking with as many people as possible about what systems they were using, I found that some systems didn't seem to make sense, so I used the "pull-the-plug-and-see-if-anyone-screams" strategy on some things. Generally there were no screams, and when there were, I learned something valuable about the way people work at InfoWorld.

When you need to cut costs, sometimes it's as simple as this: Follow the money. If you don't already, sit down with the people in your department who process bills and look at every bill above $500 in a given month. Chances are you will at least find questionable phone and telco services in use -- cancel them ASAP. Not only do these bills drag on your budget, they bog down your accounts payable department with more work.

One note of caution: Make sure you trim your infrastructure in such a way that you do not hurt your ability to scale it back up again later. Recently at InfoWorld, we were able to consolidate a group of 23 servers into 11 servers. What were these 23 servers doing? New servers had been purchased in the past to deal with what were essentially storage deficiencies. When we reduced the number of servers, we made sure that increasing storage demands could be accommodated in the new configuration without adding support overhead. If you must cut, remember that the economy will turn around at some point, and you will need to scale even more quickly than you needed to cut back.