Skip to main content

The Financial Cost of Task Switching

March 7, 2021

In the Peoples And Teams section of the Professional Scrum Master course, we discuss the impact of Task Switching. We discuss that task switching destroys efficiency and quality. 

Using the work of  Gerald Weinberg's book, Quality Software Management: Systems Thinking, the course shows that if someone is working on 2 projects simultaneously the effort spent on both isn't 50% but 40%. The individual who is switching loses 20% effort to context switching. 

Task Switching


That may not sounds a lot and people are often asked to work on more than one thing - this is a common question I asked in the class. It seems that the organisations believe this price a price worth paying, otherwise, why would so many people be asked to work this way?  Maybe the true cost of it isn't made visible to the organisation.

A technique to make this visible is to show the financial cost of task switching.
 
Let's assume that a developer is being paid £$€600 per day for their effort. If the developer is working on two projects simultaneously then 20% of the £$€600 is lost to context switching - £$€120 - every day they are working on two projects simultaneously.

Extrapolate that out and over the year, assuming the number of weeks worked is 48, it is a loss of £$€28,800.00 all due to not being able to start work and finish it.

 

Task Switching Cost

If you want to save money, potentially deliver early and deliver value sooner - help your colleagues to focus on one thing and one thing only. Then move onto the next


What did you think about this post?

Comments (5)


Tim Dickey
11:19 am March 9, 2021

Thank you for this post Steve. I've been researching this topic recently and I found your cost analysis to be enlightening. Have you considered a fully loaded rate for your numbers? While an hourly rate is a good starting point, including retirement contributions, insurance and other "total benefits" might make a more compelling story for this topic.


Adam Griffiths
12:59 pm July 1, 2021

Nice post and helpful idea - I'll have to read Weinberg's book myself.


Serge Huybrechts
09:10 pm February 1, 2022

I have believed and used this thinking quite some time, but the heuristic of Weinberg is not correct. Following his thinking, someone working on 6 tasks wouldn't be able to complete anything. That is not the case.

Here's a link to more accurate, research based information:
https://www.researchgate.ne....

TLDR; People working on 2-3 task at once lose an average of 17% on context switching. Increasing the number of tasks doesn't have a linear cause on the time loss.


KRM
05:34 pm October 24, 2022

I'd be curious how this applies to the trend of eliminating test roles and having developers handle coding and testing? Is the mindset change between coding and testing a big enough context-switch to reduce efficiency too?


edoardo capulli
08:37 pm March 19, 2023

Very simple, realistic and useful. If you actually read the original table on page 284 of Weinberg's famous book, you find than she is simply exposed, without proof. With just one task, W believes there is 100% productivity. Is that a reasonable assumption? I suppose that there is always a initial productivity non more than 90%. https://uploads.disquscdn.c...