What can Product Managers and Entrepreneurs learn from Machine Learning

Let me start with a brief introduction. I’m am a big technology enthusiast and have been playing with Artificial Intelligence a lot. I worked as a Software Engineer for a while, but recently I have turned to product management. And while I am not yet an expert PM, in my recent random thinking had an interesting thought that I wanted to share.

When someone is working on a problem that they are trying to tackle with the use of machine learning, they don’t just come up with a solution and then let algorithm run and train itself. All the number of ways in which you can set up, arrange and mix different approaches is vast and one has to try out a lot of them out before settling with a satisfactory one. The key is not to waste too much time and get the answer and a feel of a certain approach ASAP to see whether it is worth it to dive into it more or try something different. Furthermore, once one has decided to dive in deeper with an approach, it also requires readjusting the meta parameters many times before settling for one. The key again is to get the results fast and to try something else if some meta parameters are not working out. Eventually, the algorithm goes through the process of training, whether it’s any kind of neural network, a genetic algorithm, genetic programming or something else, where it takes many epochs/generations/evolutions to at some point end up with a great result. On the way through it will often get a way worse solution than the one it had before that, but it does not matter. The only thing that does matter is that small chance that with every try it will improve and start or continue moving in the right direction.

I believe you already see where I’m going with this. As an entrepreneur or a product manager or almost anything you are doing in your life, as long as the goal is to achieve something and get better, it is also very important to iterate fast. Chances are that no matter what you are doing, especially early in your life/career, it has the potential to be way, way better. While you can learn a lot by focusing on something and making it “perfect”, by your understanding of “perfect” at that time, you will probably get it to a more “perfect” state if you get feedback about your progress and the direction earlier rather then later. There is this famous saying how the majority of startups fail and that their chances to succeed at very low. While it certainly is not easy, I would say that a majority of those failing startups closed themselves in a garage (or wherever :D) and worked on their perfect solution, only to find out that nobody wants/needs it and/or does not want pay for it.

Humans have a tendency to be “shy” about their work and be afraid to show it to others. But since humans did direct those ML algorithms to iterate and fail fast until they get a good solution, it seems we do have a logical understanding that it makes sense and if you try to apply it in your life you’ll be amazed at the results. Nobody laughs at the machine (or at least the machine does not care) in comparison to a human, but the constant thing that remains is that in a day/week/month/year/at some point, the past failures won’t matter; nobody will remember or care about them. And if you do learn something and improve from a mountain of failures you will be more satisfied then not trying in the first place.

I want to also touch upon a difference between the machines and humans. At least, for now, I believe that humans are still way better at understanding the context and can use the failures better/more effectively to move in the right direction. Even more, a human is capable of getting totally unexpected insights from his/her failures and adjust what is measured to get better insights. We could say that a human can meta-adjust him/her self. While yes, machines can adjust their own meta parameters, but I think you realize that a human can still get a better insight into and understanding of complex situations and their surroundings then a machine.

That said, think a bit about what you can learn from your everyday experience. Make every experience a learning experience and you’ll be amazed at how fast you will progress.

Ok, none of this is new. All of the self-development books talk about it, and all the experienced entrepreneurs and product managers tell you “fail fast” etc. I did not write this post to share new wisdom, but to share an interesting parallel between how machines learn and what seems to be a great way for a human to grow as well. This also probably isn’t new, but I haven’t seen someone making this comparison. I’m fairly new to writing, product management, and entrepreneurship, but I do want to improve, so please share your thoughts and comments to help me grow :)

This story is published in The Startup, Medium’s largest entrepreneurship publication followed by +441,678 people.

Subscribe to receive our top stories here.