Back to being a kid
Lately, I’ve been thinking a lot about how to effectively generate ideas that my co-workers might not have thought of because they don’t deal directly with the data owned by the company where I work.
I started googling to find out, but nothing could satisfy my curiosity about this. After hours of searching for articles, I decided to take a break, visiting Youtube from my phone.
As usual, Youtube will give me recommended videos based on search history, videos I’ve watched, watch later lists, and other things, there must be a lot of variables for the recommendation system. I was quite surprised when I got a video recommendation that I will show below and also curious at the same time, “Why was I recommended to watch this video?” away from the history of searches I did recently on Youtube.
In this post, I will try to provide a summary of what I made from this video and how it relates to my career in data analytics.
Think deeply about simple things
Everything in life comes from simple things in the beginning. Over time, simple problems can be resolved, but these things are far less than the new things that are popping up. I feel that I already know this, but for some reason when watching this video, I feel shocked because it turns out I forgot about it.
Let’s say my data analytics team has a problem with the churn rate of our website visitors that continue to increase in the past year.
When we do brainstorm to find out what we should do to find a solution to this problem or maybe other similar problems that often arise when working, we often immediately try to find a solution, for example making a predictive model from the data we have to know what variables most influence the churn rate, without thinking deeper about the simple things that we need to discuss first, such as “Why do we have to think about the churn rate?” After trying to ask this myself, of course, I have the answer, but I doubt myself, is that the right answer? Let’s say it’s the correct answer.
Based on many research results, acquiring new users is far more expensive, 5 to 25 times compared to reviving old users.
After getting that answer (which came from me), another simple question arises in my mind, “Why can the difference in costs be up to that much?” The answer from me in a simple format is “Because new users need time to adapt to a new environment, need to get a lot of experience (good and bad), need referrals from people in their lives, need more attraction so they want to be loyal to come back again, and other things that are not I can mention one by one here.” Over and over again I kept asking “Why?” in my mind, as well as I kept trying to answer all of that, it turned out the “Why?” never ends.
Without realizing it, when we did a brainstorm, we immediately decided to make a solution by creating data modeling. In fact, the answers to the question “Why?” it’s actually been a framework for figuring out what’s causing our site’s ever-increasing churn rate. The problem we faced can be resolved easily only by using exploratory data analysis (EDA), with no need for data modeling, for this time.
But in the long run, yes, we need some methods to do those things automatically, so we can work on other problems. That’s when we need to build data modeling and use machine learning.
Why? Why? Why? What if …?
In my humble opinion, “Why? Why? Why? What if …?” is the crucial moment. This is the part we need to back to being a kid. I am sure that we all know that kids are terrifying, their curiosity is beyond imagination, and all of that is purely based on their mindset, they just want to know more, about everything!
Currently, when I write this post, I have 9 years of experience in data analytics. With all that knowledge and experience, sometimes I felt that I know all the basic things about data analytics but at the same time, I don’t know about anything, because I felt my creativity is limited, my way of thinking is too stiff, my experiences make me think that the solution to the problem I’m facing now should have been found before. Then I remembered Einstein’s words, “If you can’t explain it to a six-year-old, then you don’t understand it yourself”.
Without realizing it, I had put myself into “you” in that sentence for too long, until I forgot that some times I could also put myself into “a six-year-old”.
I wrote this post mainly to be a reminder to myself if one day in the future I return to being like this so that I don’t forget that occasionally returning to being a child is a good thing. Of course, I would be more grateful if this could serve as a reminder too for those of you who read this article.
