Division of Labor, Stacking, Sequencing, and Cycle Time Reduction
By
Saul Swartout
Senior Consultant
It is almost impossible to read about, participate in, or lead a redesign or reengineering workshop or project and not hear about the reduction in cycle times and the quality improvements result from value-added task analysis. In a macro way, it is a simple activity. Figure out which tasks add value and keep them. Figure out which tasks do not add value (or enough value to justify the costs) and get rid of them. Figure out which tasks you are not doing but would add value and bring them into the process. Figure out how to do each task as efficiently as possible. Finally, we sequence all these tasks into sub-process and—voilà—we have our new process.
Through task elimination and more efficient task completion, we have reduced our cycle time. Let’s go implement the new process.
Not so fast. We have left a good number of cycle time improvements on the table. Two related and often overlooked activities will reduce our cycle times even further: maximizing division of labor and task stacking.
Division of labor? You bet. To complete each task requires a specific set of skills, and many tasks have identical or very similar sets of required skills. For instance, in an insurance company, the skills required to underwrite a risk are typically different from those required to rate a policy, and those skills can differ from the skills required to issue the policy. To maximize this division of labor, tasks requiring similar skills should be completed by the same person as often as possible. When viewed on a process chart, these divisions of labor take the form of “swim lanes” that represent an individual or a group with similar skill sets. This person or people are charged with completing the same set of tasks within a process.
Sounds simple enough, but mismatches frequently occur because not enough care is taken to analyze what skill sets are best suited to each task. Another reason is that it is often assumed that because the task has been historically completed by an employee of a certain skill level, it should continue being performed at that level in the future. Another culprit can be skill requirements within one person or group that are far too broad to be cycle-efficient. These mismatches can increase cycle times and lead to employee boredom, frustration, and ultimately, to unacceptable turnover rates. If all the above problems and associated expenses are not bad enough, these mismatches can severely underutilize salary and salary-related expenses. In short, they can institutionalize division-of-labor inefficiencies.
Let’s say that we have been very careful with our skill-and-division-of-labor matches and have further reduced our cycle times. Are we ready to implement the new process? Nope; not yet. Now we have to stack and sequence these tasks within each swim lane while minimizing the number of hand-offs. Stacking involves sequencing tasks between multiple swim lanes so the tasks can be worked on simultaneously.
Hand-offs typically increase cycle times and can lead to quality reductions, so balancing cycle time improvements produced by stacking against hand-off impacts always involves tradeoffs. But because cycle time improvements can be substantial, the rewards might be well worth the effort. On one recent project, stacking alone reduced cycle times by an additional 15% after all the other process improvements were considered. This saved over one full week per transaction!
Yes, now we are ready to implement the new process. Of course, getting the right people in the right place and creating the right environment can be equally important success factors, as is our new process and process metrics. But these are topics for another article.
