When do you take a holiday?

When you're running your own company, there is the temptation to be "output driven" in terms of taking breaks. Avoid that, and take care of yourself instead.

There are two schools of thought in terms of when one can afford to take a holiday - it is based on whether you look at input metrics or output metrics.

When you base it on output metrics, you decide you’ll take a break (or any other such indulgence) only when you have “achieved something”. This can also be thought of as the Ganesha-Subramanya model. In that movie, for example, the protagonist brothers have a vow that they will only marry (or otherwise engage with women) after they have achieved something.

The other model is based on “input metrics”. This is more normal and conventional. For example, typically after you’ve worked for five (or six) days, you decide you need a day or two of rest. The work week legitimises this period of rest. It doesn’t matter what you’ve achieved (or not) in the week. That the weekend is here means that it’s time for you to take a break from work.

Normally most people follow this input metrics model. However, sometimes, when you’re hyperfocussed on something or building your own thing, you can tend to get into the output metrics model. Because you are responsible for your own outcomes, you want to try and maximise your chances of getting your desired outputs. And sometimes you forego any rest or other indulgence until some measure of achievement has been attained.

Usually I follow this “input metrics” model as well and am keen on taking my time off during the weekends. However, ever since we launched our angel funding round in January, I’m finding that I’m working most weekends - what is work for me is “non-work” for a lot of my investors, and so they choose to talk to me during a weekend. And combined with that I do some other work as well.

This continuous work has accelerated after coming to SF as well - one part of the argument is that “I have only three weeks here so I need to maximise the output in this time”. Last Saturday I went out to meet two sets of friends, but on Sunday sat in my room all day working (it didn’t help that the weather was absolutely atrocious that day, with constant drizzles through the day).

Today is a good day in SF (nice and sunny), and after having recovered from a mild stomach ache in the morning (which forced me to convert an early morning meeting to a Zoom meeting), I decided to head out. I have work to do - in fact I have tonnes of work to do. However, I decided that this being my last weekend in SF on this trip (I shift base to the South Bay on Tuesday), and the weather being nice, I should take advantage of it. And that I can work in the evening after it gets dark (fairly early) - as it happens I got back an hour ago and will start work soon after I finish this blogpost.

I’ve also come to believe that the method in which you choose to take your holidays significantly affects mental health - being “output focussed” can work great for you if you’re doing so for a short period of time (a “sprint"), but if you do it for an extended period of time, then it leaves you with the risk of burn out. There is also the issue that when you hyperfocus on only one thing for a long period of time, your efficacy in terms of how well you can perform in that can also drop - there is no time for you to think of fresh ideas.

The other thing to consider is the correlation between input and output at work - if you work in a “linear job” (such as an assembly line worker), your output is approximately proportional to your input. Our current industrial model (5 days of work a week, largely fixed work hours) is based on this kind of a prior.

But if you work in a “high power law job” (where “achievement” is intermittent and lumpy, then input and output are not very highly correlated. And a lot of our modern “knowledge” work is this way.

And running your own startup can take this to an extreme - there are a small number of “breakthroughs” strung out through a long period of time. Suddenly you’ll get a customer. Suddenly you’ll get an idea that transforms your product. Suddenly you’ll have a highly impactful meeting. No one day is like another.

In such a situation, input and output are not correlated at all. And the temptation can be to do a Ganesha Subramanya and be output-driven.

The fewer the breaks one takes, the less the “average impact per unit time worked”

I’m writing this post as a note to my future self, and to other entrepreneurs, that this temptation is not likely to be good for our health in the medium to long run, and can increase the chance of burnout. Yes, you might want to maximise your chances of making an impact (and so work longer hours), but you also need to consider that the fewer the breaks you take, the less the “average chance of making an impact per unit time worked”. And this can get counterproductive very quickly.

So take care of yourself. Yes, there is always work to be done. You always think that postponing your break until some “achievement” is a good thing. You think people will think bad of you if you are seen taking too many breaks. What matters, though, is overall impact. And for that you need to take care of yourself.

Don’t be like Ganesha and Subramanya.