Keep Focus – Eliminate Bad Multitasking!

It is amazing how hard it is to get everyone to stop doing bad multitasking and focusing on a single task or project at the time. This is despite people understanding and seeing how negatively multitasking affects the outcomes - somehow people put pride in demonstrating they can handle many things at the same time ("keeping many balls in the air") even if this is so much less effective than keeping focus and delivering each task or project with excellence, one at the time.

Multitasking increases Lead Times and Average Process Time

To illustrate the severity of breaking several tasks or projects up into smaller parts and jumping from part to part, I will walk through a simple thought experiment.

Imagine you have only two tasks or projects that each take 2 days to do - there is 2 days of work in each. To keep it simple, you are only given two options about doing the work: 1) do one after the other, or 2) do half of the first, then half of the second, then the second half of the first, and finally the second half of the second. Assuming that there is no switching cost that will make the duration of the tasks take longer (e.g. having to get back into the task again, setup equipment, etc.) - the actual work time, then we can ask ourselves: "When will the first task or project be delivered, when will the second be delivered, and what is the average process time?". If we do the tasks or projects one at the time, the first is delivered after 2 days and the second after 4 days - the average process time is therefore 3 days. If we do them in parts by multitasking we can see that the first is completed after 3 days and the second still after 4 days - the average process time is now 3,5 days.

There is clearly no benefit by doing multitasking, even when there is no switching costs as we would expect in real life. It is like everybody trying to get out of the door at the same time in a burning building, so everybody blocks for each other - if only they just would walk out in a line they would get out faster, but everyone selfishly tries to be first out. This is what happens when every project or task is "high priority" – then the people doing the work jumps from task to task to adjust to the latest re-adjustment (project managers screaming) to what is most important and every project and task is completed much later.

If we do more detailed experiments with up to 20 projects that can be done one after the other or done in bites, where each project is divided into 20 parts, we will see the percent of the lead time before the first project is delivered goes towards 95% (19/20) when we do multitasking, while it goes towards 5% (1/20) when we do one project at the time.

The average process time for the projects when multitasking goes towards 100% (all projects are completed almost at the same time at the end), while it goes towards 50% when done one project after the other (it will always take at least half the time to do half the projects assuming equal time needed for each).

So, what is the real impact of doing bad multitasking? If you are delivering to several projects, someone are often waiting for you to deliver your part or parts, so if you and they are multitasking the impact on the project is on average that the time is doubled as the average process time is twice as big. However, the lead time for the most important projects could be multiplied many times.

"Bad Multitasking –

Nobody gains, Everybody suffers!"

- Eliyahu M. Goldratt

As demonstrated, there is absolutely no benefit of multitasking - the first project or task is delivered much later and in reality we would expect the switching cost of jumping from project to project or from task to task will have a significant negative impact on the time it takes to complete each project or task. So not only will it take much longer time before the first project or task is delivered, but all projects or tasks will require more work than necessary delaying all projects.

The causes of Bad Multitasking

In Theory of Constraint (1+2) it has been explained that some of the key causes for Bad Multitasking are related to:

• Too many concurrent projects - switching costs prolong each task to take longer than neither necessary nor planned for
• Over-utilization of resources - with no idle time any delay will impact all other projects or tasks due to dependencies
• Student syndrome - when work is postponed there will never be tasks completed ahead of time, but many may take longer than expected
• Parkinson's Law - when workers are punished or blamed for late deliveries, they will always ensure they are never late again by adding more buffer time than needed and they will never deliver early
• Integration points - when different task-legs of a project convene at a point where all must be completed to move forward, any leg that is late will delay the whole project, even if any or all of the other legs are delivered ahead of time

What we see is that time savings are most often wasted, while delays are accumulated!

Solutions to Bad Multitasking

• Freeze 25% of the projects or tasks - this will reduce Bad Multitasking and increase flow, more will actually be done
• Choke projects or tasks - when no new projects or tasks are coming in for a while, then the current overload will get out of the "work pipeline" so work can be done focusing on one project or task at the time
• Stagger projects or tasks - only add a new project or tasks when one is completed and focus project planning around the key resources (people) needed in projects that are constraints for deliveries
• Drop workload to 80% - this will ensure resources are not overloaded so queue time extends (tasks are idle, waiting for the resource to have time)
• Implement Relay Runner mentality - people must embrace the mentality of relay runners, where you sprint when you have the baton as it reduces Bad Multitasking and increases focus
• Limit personal queues - specialist should not accumulate lots of tasks for later as this will prevent others from working on them if they are available and it is harder to squeeze in emergency tasks with minimal impact

So, in concluding this post, you should consider how big a problem Bad Multitasking is for you and your organization, and to what degree the solutions suggested above could help solve this problem. How will you ensure focus? /LAX

References:

1) Goldratt, Eliyahu M., (1999) Goldratt Satellite Program - Sessions 3: Project Management and Engineering, (Video) Broadcast from Brummen, The Netherlands: Goldratt Satellite Program.

2) Goldratt, Eliyahu M., (2008) The Goldratt Webcast Program on Project Management: Sessions 1–5. (Video series: 5 sessions) Roelofarendsveen, The Netherlands: Goldratt Marketing Group