Reality IT: Good Intentions vs. Good Decisions

Over the past few months at ACME, we've engaged in some head-spinning analysis: whether to outsource one function and insource another. This might seem like organizational schizophrenia but, in fact, the two initiatives shared the same goals: to save money and use resources more efficiently.

However, the executives driving these initiatives often didn't complete their homework. My crew did further research into the IT costs. From our business analysis, we concluded that neither plan would save money. Here's a real-world view of how two executives looking to overturn the status quo learned that good intentions don't necessarily lead to good decisions.

Would-Be Projects

On the outsourcing side, CIO Steve Fox had been pushing to move our mission-critical servers to a collocation site. He believed collocation would provide better system uptime and growth options.

Our data center is about two years old and we recently completed a server consolidation effort, moving many applications to several 32-way big boxes. But we've been adding to those systems and provisioning new one-off boxes for various projects and, as a result, we're pushing the limits on physical space, power and cooling.The insourcing initiative, meanwhile, was the brainchild of CFO Xavier Beane. He wanted some billing operations handled by a service provider to be brought into the enterprise.

Xavier believed we needed better control of the processes the service provider currently performs, such as printing batched invoices, processing related notices and contacting clients who are late or have problems or questions. His idea was to bring the services in-house using our own finance and accounting staff, while IT managed the servers and applications required for these operations.

Making the Case

When we built our data center, the manager in charge of the project, Marvin Mips, researched other options, including collocation. When Steve proposed this new collocation plan, Marvin updated his previous research and also looked into ways to alleviate our current data center growth. Marvin determined that making improvements in the data center would prove more cost-effective than collocation. One reason was that we would have to support both our data center and management of the collocation site. We also negotiated with vendors to get credit toward an upgrade for our current cooling and power conditioning/backup systems. The upgrades will require some downtime, but will give us room for much more growth and further upgrades later. We also planned expanded server consolidation.

My manager of finance systems, Josh Benson, looked at the costs and complexity of insourcing the billing operations from an IT perspective. To run these functions, our finance application would need a major upgrade with an added module. Our call center's workload would also increase because the staffers would have to contact clients. Another issue was the high-speed line printers, which are expensive and require frequent and intense maintenance. Even if we brought other billing functions in house, we'd still outsource printing operations.Case Closed

Both projects died once IT costs were more fully analyzed, and neither executive was pleased. Steve was adamant about outsourcing--it had worked for him before at another firm--but he realized it would not be economical in this case.

Xavier went as far as to ask us to lower our estimates to meet his department's cost calculations. But Josh stood firm on his research and I backed him up. However, we did compromise by agreeing to revisit the insourcing proposal during the next upgrade of our finance system.

We try to get into the planning process early for projects like these, so that our analysis doesn't come as a surprise to the champions and we don't look like project-killers. It's never easy disputing executives; if you ever find yourself in this situation, make sure your research is as close to perfect as possible.

Hunter Metatek is an enterprise IT director with 15 years' experience in network engineering and management. The events chronicled in this column are based in fact--only the names are fiction. Write to the author at [email protected]. 0