Optimization Magic Wand #1 - Throttle the Work
The #1 most common sub-optimization is working on too many things at the same time. One of best ways to improve your outcomes and productivity is to throttle the work. Lean and Agile refer to this as “WIP (Work in Progress) Limits”.
Limiting WIP is probably the most counter-intuitive, culture challenging optimization out there. And it’s also the most effective.
What would happen if you only worked on one thing to completion before moving on to the next? Many people have described focusing on one thing and doing a good job as a “luxury”. Do you want "doing a good job" and "raising productivity" to be a luxury in your organization? Or should it be a responsibility?
Let’s be super-clear here: Throttling doesn’t mean that you do less. It means that you do less at a point in time. You will actually end up doing more over the quarter or year. The key to throttling is to increase focus and keep work moving. Just like a sewage pipe, put in too much at once and nothing moves! You need to slow it down, to get more through.
Scenario: Your team is asked to complete 12 projects this year, each one will make $1M/month after launch.
Option 1: Your first option is to start all the projects in January, work on them all year, and deliver them in December. Expectation met! And on Jan 1, you make $12M, and you keep making $12M every month.
Option 2: The team decides to work on only one project at a time. They would like to start generating revenue as soon as possible. In fact, they do, generating $66M before the year is over! Not only that, but they are able to complete an additional 4 projects with the time they’ve saved by eliminating context switching. In January, they are generating $16M/month.
But the Option 2 team has some problems. Although they have delivered the company over $66M in additional revenue, some people are not happy with them. Why? The owner of the orange project says “ can’t get any status on my stuff, no one is aren’t even working on it!” The team says “Yes, but you’ll get it sooner than you would otherwise, if we start work on it now you won’t get it until December, and we’re looking at getting it to you in September!” Will the Orange Project Owner trust the team enough to delay the start? If not, a stakeholder can cannibalize the interests of the whole company!
Why does throttling help?
Reduction in time spent coordinating. Think about how much time you spend trying to coordinate schedules for a meeting. And how much time elapses between when you identify a need to meet and when you actually meet. In option 1 above, if the orange project team encountered an obstacle, how long would it take to arrange a meeting? Everyone is working on other projects, and they all have different times committed. Maybe a week? And what is the impact to the other projects when we have to change meeting times? How about in Option 2? If we need a meeting in option 2, we can have it right now! If we are all working on the orange project and we have a critical obstacle with the orange project, we can all drop everything and meet.
Reduction in time spent context switching. Individuals lose a huge amount of productivity when context switching. The chart below show is based on the study published in by Quality Software Management by Gerald Weinberg and Roger Brown. You can see that when switching between 5 projects a single person loses a whopping 75% of their productivity to context switching, leaving only 5% for each project.
But wait, that’s just the beginning. What happens when you compound the loss from context switching across an entire team? What about an entire organization? You are well below 1% productive time, and the waste is somewhere about 99%.
Value realized early. The upside of throttling is that when you focus on the most important thing and get it launched, you can start realizing the financial benefit right away. If you factor in the time-value of money plus the additional time earning or saving, you can dramatically affect your financial gains.
Have you tried using throttling or WIP limits? What was the result? We’d love to hear from you in the comments.