Implementing Six Sigma in IT

\

Six Sigma combines a proven approach known as DMAIC (Define -> Measure -> Analyze -> Improve -> Control) with a 3.4 DPM (defects per million) process quality target.  The 3.4 DPM number comes from the standard normal distribution, and is much less important than the DMAIC improvement methodology itself.  In the information technology world, a “defect” is defined as any undesirable process outcome, such as a Level 1 help desk response time exceeding “x” number of minutes.

Accelerated improvement using Six Sigma is built on solid metrics, training, and management commitment.  Metrics come first, as a means for framing up the improvement potential and justifying the training and project work.

Metrics

DMAIC’s Define and Measure phases depend on existing metrics at either the process or business level.  Oftentimes the metrics development effort is the toughest part of a Six Sigma initiative, but the benefits of metrics go way beyond Six Sigma.   Here are three good links related to IT metrics development –

Info-Tech

Basic Guidelines on Metric Development from the University of California

Comprehensive list of potential IT metrics from Pink Elephant

Training

Once Metrics are in place, DMAIC is the winning formula for moving those metrics toward Six Sigma capability at an accelerated pace.    Training at the leadership and technical levels will be necessary.  There are very few IT-specific training programs available, but Villanova University does offer an IT-based program.  If an IT-specific program cannot be found, then a transactional Six Sigma program is the next best option.  One last note on training – programs that incorporate hands-on projects between classes are by far the most effective.

This site describes each phase of the DMAIC process, along with some specific examples of Six Sigma tools applied to IT projects.