Hello World!
What is the most important organ of the human body? Is it the heart? Maybe the lungs? The Brain is a good choice, too. They are all vital, so the question seems pointless, but I will try answering anyways. Well, not exactly this question, I am not a Doctor, but I will answer the Data Science version of it.
A working Data Science solution also combines many working parts, all of which are vital for a Use-Case to produce value. My life may not depend on it, but which toothbrush you get recommended does.
In my experience, the most important part of a Data Science Use-Case is understanding the business question. There are 2 main reasons for this:
- An approximate answer to the right question is better than an exact answer to the wrong question. (Quote by Tukey)
- Getting Buy-In from stakeholders is one of the hardest and most important things in Data Science. Show them that you understand the business makes them trust you and your recommendations more.
The obvious solution to this is learning more about the business. So if I have the goal of improving a Use-Case and I have the choice of improving my understanding of one of the areas:
- business
- data sourcing and pipelines
- modeling
- software engineering and testing
- deployment
- presentation
I will focus on the business side.
All of this is a lot to ask of one person, which is why most companies nowadays split them up into roles like Product Manager, Data Engineer, Data Scientist and Machine Learning Engineer.
Of course, no one can be an expert in all of those things, but having a decent grasp on all of this helps in decision making and communication in your day-to-day if you work on Data products.
-> Data Science has a learning curve that stays steep for a long time. Knowing more after you already know a lot can still have a big effect.