Skip to content

Tough Love

About a year ago, in a post about custard apples, I was talking about experimentation. The idea being that you have to assume failure and conduct a lot of experiments to really be able to use the method effectively. I also argued that this is not a new concept. In fact, we understand this quite intuitively, especially when it comes to nature. However, we seem to apply it very selectively in our lives.

For example a colleague of mine pointed out in the comments that this concept is no different than a VC firm having a portfolio of startups, most of which might not succeed but a small percentage will compensate more than enough for the losses. It is also no different than a hedge fund. In fact the risk mitigation strategy itself is called hedging. So suffice to say that we already apply the concept of responsible experimentation in some areas of business. However, we seem to falter quite predictably in other situations and I want to explore that a little bit. Not just academically but also in being able to apply it in our lives and work.

The inevitable result of conducting many experiments is that some of them will fail. But that is not your biggest problem. Your biggest problem is that when something fails it doesn’t just go away. Sticking to the gardening metaphor, a failed experiment often turns out to be a stunted plant rather than a dead one. Your first instinct, naturally, is to pay it a little more attention and take better care of the plant. But while it throws up a few new shoots now and then, in a few weeks’ time you know that this one is not going to grow well. It’s holding up the limited resources you have at your disposal. There’s opportunity cost involved. You know what you need to do… but if you are like me, it’s not an easy step to take.

One of the biggest issues that I’ve had to deal with, in my (admittedly limited) gardening experience, is the act of “killing”

There are several things at play here, both emotional and practical.

1. Attachment aka Sunk Cost Fallacy

Firstly, it is natural for you to be emotionally invested in something that you have put so much effort and resources in. So there’s a natural hesitation in killing it.

2. And what of “hope”?

One of the big problems with “ending” anything is that it’s perceived as ruthless and crude by the world and in this analogy we are talking about life, no less! To kill something, even if it’s just an experiment, means giving up hope. And in a world where we are constantly bombarded by stories of success against all odds taking a practical decision feels shallow and selfish!

The problem isn’t that we make heroes out of optimists but that we make villains out of realists

3. The problem of “when”?

Even if you keep those emotions at bay, there’s another practical issue. Everything said and done, this decision is matter of judgment. A plant might start slow and whether it’s just going to remain stunted or whether it will turn out to be the one that bears fruit is anyone’s guess.

The truth of the matter is that there’s a chance that you’d be making a wrong decision while shutting down an experiment and you just have to be OK with that.

So is this an intractable problem?

The important thing to keep in mind is what you are optimizing for. Just like the hedge fund that optimizes for overall profit and doesn’t care about letting go of individual investments. Or the VC firm that cares about it’s overall portfolio and doesn’t worry about the individual startups that are not doing well.

Trouble brews when you realise that while you are running multiple experiments and optimising for your results, you are probably an experiment for someone else. While the team optimising for a specific feature is running experiments to understand usability, the product team is experimenting with features, the ecommerce team is experimenting with products, the company is experimenting with digital strategy, the market is experimenting with companies and so on… perhaps the human race itself is an experiment… perhaps the earth just wanted plastic! 😉

Leave a Reply

Your email address will not be published. Required fields are marked *