DEV Community

Cover image for Know when to automate things (and when not to) ๐Ÿ‘พ
Adheeban Manoharan
Adheeban Manoharan

Posted on

Know when to automate things (and when not to) ๐Ÿ‘พ

According to a 2020 study on "shift in number working hours from 1870 to 2020", It was found that an average worker in a rich country today really does work many fewer hours than the average worker 50 years ago. Although, this steady decline in the number of working hours could have been the direct cause of multiple events like the end of the great depression, major global wars etc.., one major and inevitable reason is the rapid technological advancement that we've been experiencing in almost all the fields in the last few decades.

Every task in today's world is either already automated or people are trying to automate it. A McKinsey Report estimates that automating tasks can increase annual productivity by about 1.4% in industries. Yes, Automation plays out really well in several of today's fields.

Now, how about in the tech industry? Well, this is where modern day automation was born. As techies, we all have this constant urge to automate things. But don't you think that some of us overdo this a bit? Well, atleast I do. I try to automate almost everything that I do in a day, if I see the opportunity to do so. That too with a versatile and powerful scripting language like python, not even a second's thought. Adding to that, the thrill that gives us as developers while facing the challenges the task poses while trying to automate.

As cool as that sounds, while trying to automate a particular task, I sometimes end up spending more time on the automation in itself than it would've required to perform that original task, which technically defeats the whole purpose of automating in the first place. In short, I ended up wasting time trying to find the easier route.

So, that task clearly was not a suitable candidate for automating. Right, so when is a task automation worth your time?

Repetitive, manual tasks ๐Ÿ”

I heard this from a colleague whose work I admire a lot, ' If you are doing something more than twice, automate it ! ', He said. Tasks that need to be done frequently or on a set schedule are perfect for automation because the benefits are immediate. You'll spend less time each day on repetitive tasks that can quickly become annoying or irritating. Even it is a task you ought to perform once in a year, go for it. Time intervals don't matter, as long as it is repetitive and consumes a lot of your precious time and energy.

Boring and complex tasks ๐Ÿฅฑ

If a task doesn't excite you and if it is quite complex to perform, automate it. Doesn't matter if it takes quite some time to automate, because doing something that you don't like brings your whole spirit down, which in turn dials your excitement down in the other tasks resulting in the decline of your overall productivity. Also, if that task is complex, but we'll have to do it manually, it is by nature prone to many human errors. In this case, an automation framework to perform this particular task goes a long way in preventing these kind of human errors, provided that you get the framework right.

Distractive tasks ๐Ÿ˜ถโ€๐ŸŒซ๏ธ

When you start your day, you might have a lot of tasks on your plate and it would be very tempting to get started with the most easiest one, rather than the important ones. These kind of tasks will easily sidetrack you from the very important and impactful tasks. But, these are the kind of tasks that cannot be skipped, no matter how insignificant they are. If time allows you, you should automate these kind of tasks, to stay focused on your main tasks.

If a task checks all the boxes I have listed above, It is already begging to be automated, go for it. Don't spend your precious time on the insignificant ones //

Here's a thumb rule:

If you hate it, automate it !

Good luck on your automation journey, I would love to hear your thoughts in this, Please do comment. Cheers! See you in the next post.

References:

Top comments (0)