Please stop multitasking. I’m begging you. Please.
The greatest trick we ever pulled on ourselves as knowledge workers was convincing ourselves we could juggle multiple projects with no consequences.
You know how sometimes you hear an idea that is so obviously and demonstrably incorrect that you can't really understand how it became popular? For me, multitasking is that idea. The lie that we can get more work done by working on multiple things at once is so absurd, so incorrect, and so utterly destructive that I can't believe it's a near-universal practice.
And yet, we all get suckered into believing it. I mean, I absolutely know that it's a terrible idea, and I still end up multitasking.
But I understand how it happens — it's a seductive idea: "I have three things I need to finish. I'll spend a little time working on each one, and then all three will be done faster!"
But in reality, multitasking is a productivity disaster. It not only fails to get projects done faster — it makes them significantly slower to complete.
Multitasking is how we manage to work hard all day and still get nothing done.
We've all had days where we know we were working but we can't for the life of us figure out what we actually got done.
For me, those days are always because I spent the entire day bouncing between tasks instead of focusing on a single thing.
Rather than actually working, I spent the entire day context switching.
Context switching is the silent killer of progress.
Gerald Weinberg’s book Quality Software Management: Systems Thinking offers up a heuristic for the impact of context switching: for every additional parallel task, 20% of working time is lost to context switching.
If we plot that time loss on a bar chart, it's pretty chilling.
That means for every parallel task you work on, you're losing about one full working day each week.
And it's worse than you might think: context switching time loss is compounding — five parallel tasks means you can expect about 80% time loss to context switching.
If that sounds hyperbolic, think back to a day where you worked hard on a few projects all day and yet nothing was crossed off your todo list. That's context switching time loss chompin' on your time; 5⨉ the time spent to get the same tasks done.
Multitasking feels productive, but sequential work *is* productive.
When you're staring down a sizable backlog with lots of stakeholders who feel very strongly that their project needs to be done right fucking now, the temptation is to start everything at once and work on things in parallel to "keep everything moving".
Lots of in-flight projects feels productive. It feels good. Look how busy we are! We're killing it! Go us!
The unfortunate reality, however, is that we're moving significantly slower than we would if we just did these projects sequentially. Working on one project at a time leads to both faster initial delivery and faster completion of all projects.
How to become as productive as you thought you were before you realized how much time you were wasting with context switching.
Something that still perplexes me about multitasking is that knowing how awful multitasking is doesn't actually prevent people from doing it. I've been up on a soapbox arguing against multitasking for years, and I still find myself losing days to context switching.
It requires deliberate effort and good habits to combat the siren song of multitasking. Here are a few of the techniques I use to keep myself productive and focused on a single task.
1. There can only be one priority.
The idea of having multiple priorities was invented by management consultants. It's the original corporate nothing word — it's a thing you say to make people feel good without actually accomplishing anything.
The word priority came into the English language in the 1400s. It was singular. It meant the very first or prior thing. It stayed singular for the next five hundred years. Only in the 1900s did we pluralize the term and start talking about priorities.— Greg McKeown (@GregoryMcKeown) November 30, 2018
Pretending we can have more than one thing at the top of the list just means we're not actually prioritizing anything.
When you prioritize, choose the one thing that needs to be done first, and only work on that one thing.
2. Close everything and use timers.
One of the sneakiest forms of multitasking is distraction. Bouncing between apps breaks concentration — that's a context switch.
Combat this by choosing an interval that works for you, set a timer, and close everything unrelated to your task for that full working block.
There are lots of strategies and apps to help with this, but don't overthink it. Just close stuff and set a timer.
3. Break down large tasks into smaller milestones.
With large projects, it can feel unrealistic to only work on one thing for months. But that doesn't mean you should just give up and multitask.
Instead, break large projects into milestones and treat each milestone as a project that you can focus on exclusively. This allows you to make progress in a productive way, get to a logical stopping point, and then work on the next project — whether that's the next milestone or something unrelated.
Each of these milestones is reasonably self-contained and allows me to hit a logical stopping point, work on something else, and then come back without needing to rebuild all the context from the previous phase — I'm starting something related, but contextually different.
Seriously. Please stop multitasking.
Multitasking has no redeeming qualities. It's the high fructose corn syrup of work habits; it's bad for us, makes us feel worse, and only exists because someone thought it would help them make more money.
It's not solely responsible, but multitasking and context switching time loss are heavy contributors to feeling overwhelmed, frazzled, and burned out at work.
And again — I truly cannot stress this enough — it has zero benefit. Multitasking is choosing pain and making a wish that we know will never come true.
Just do one thing at a time. It's better. Faster. Less stressful. More effective.
Literally everything about it is better.