The Math Behind Context Switching

Let’s imagine a situation. You are tasked with two projects – each will require about 15 minutes worth of work. You could choose to complete the first project, then move on to the second task. OR you could choose to multitask – jumping back and forth between each task.

Here’s what that might look like:

First Example – Sequential

Project 1                                           Project 2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Second Example – Multitasking

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

A lot of people believe that multitasking is the better option. Those people are wrong. Here’s why:

Even if context switching consumes absolutely zero time (in what world…), the successive method is still faster than the multitasking system. In both examples, it takes 30 minutes to finish both projects, but think about how long it takes to get the results of each project.

In both cases, Project 2 (blue) is completed at the 30-minute mark. But look at Project 1. With multitasking, results take 29 minutes to arrive… yet with the successive projects – Project 1 is completed in 15 minutes. Your team can iterate more quickly this way.

Let’s look at the average time it takes before Project A and Project B are completed. The average time will always be less with successive projects than with multitasking projects.  In the first example, the average time needed to complete projects is only 15 minutes. In the second example, the average time passed before completion of both projects is 29.5 seconds.

System Project A lasts Project B lasts Average Time Successive 153022.5 minutes Multitasking 293029.5 minutes

The above calculations assume that no time passes when context switching. But as we alternate back and forth between tasks, we do lose time. We can’t effortlessly glide between functions without having to turn off our thoughts on one subject while starting up the cognitive engines related to something different.

Let’s assume that each time we switch context, it requires 1 minute before we really get settled, understand the scope, and find a groove to achieve meaningful + productive work.

System Project A lasts Project B lasts Average Time Successive 1530 + 1 context switch = 3123 minutesMultitasking 29 + 28 context switches = 5730 + 29 context switches = 5958 minutes

Now let’s get realistic – it takes more like 5 minutes to really get back into the groove after clearing out our email or getting off the phone with a client. We can use this more realistic number to see how drastically context switching impacts us on a given day. Our attention spans last longer than a minute so let’s slow the context switch to every 5 minutes in this example.

SystemProject A lastsProject B lastsAverage TimeSuccessive1530 + 1 context switch (@ 5min) = 3525 minutesMultitasking25 + 4 context switches (@ 5min) = 4530 + 5 context switches (@ 5min) = 5550 minutes

People who multitask frequently should recognize that these two 15-minute long projects have turned into a workload that takes twice as long to complete as it should.

We strongly urge companies to remove these potentials for productivity drain. Let’s use Jira status updates as an example. Coding and explaining your code to a manager via written word work two different parts of our brains. Each time we alternate back and forth between these two different concepts, we lose something – as is demonstrated by the examples above.

At Nextup, we decrease the need for context switching here. We allow every Jira update, assignment, ticket, etc. to be accessed and updated within Slack. No need to have 10 Jira tabs open if you can understand the entire scope directly from Slack.

The longer it takes to successfully switch contexts, the more exponential this average time difference becomes. Let this be a testament to the concept that multitasking is not real. It’s a myth created by procrastinators.

Nextup.ai – we help short projects stay short.