Pros and cons of an overtime

15 July 2021

Okay, okay, it's a trap. I will not speak about pros of an overtime. They are very subjective and controversial, and I am trying to pass some valid solid point.

Almost every developer knows what overtime is. It sometimes has different nature, but it is always about spending your
own time to do a job.

So, what is wrong with overtime?
Overtime is a symptom. It signalizes that there is something wrong with planning.
Planning is one of the most important processes in every stable company because it determines finances, obligations, roadmap. Pretty much everything that forms company's future.

Overtime may and will happen naturally. Highly correlated with planning fallacy it
pushes us to restore balance by putting more efforts than we actually planned.
This experience helps us to understand the mistakes and to avoid them in the future.
Systematic overtime simply means that we do not learn from our mistakes, leading us to the same problems all the time.
So, here is a list of cons coming out of above:

  • company loses money or deadlines (often both);
  • company hardly predicts how much time or money will cost their new features or updates;
  • overtime is about doing urgent things in a great hurry. Quality will be affected a lot.
  • nobody likes overtime. It may (likely will) worsen team relationships and their mental health;

Back in the days I used overtime to improve my skills just by doing more job.
It was okay (for me) until I had to lead small team, and I faced the consequences.
I've found that stable planning is more efficient than doing job in curves.
Management receives reasonable estimates; developers did their job with all stages (architecture, task decomposition, development, test, deploy); clients received their features and fixes exactly at time that we promised.
Worked like a charm. Everyone was happy.

One may ask about exceptional situations when we have to do something out of work hours.
Bear rushed your datacenter?
Client have received painted duck instead of a rack?
Non-optimized query downtimes website?
I will have to say that this is a part of a planning to. Companies usually have duty schedule for such incidents.
Even if not that, we have to remember, doing overtime may be okay in some situations (be agile, do agile).
Whilst, to think that it is okay - is not.

Don't take everything plain: we have to challenge and prove the information we face.

Here is what really helps me to do it.