Why Analytics is not cool anymore.
I recently had a big change in mindset in the way I look at analytics which has led me to value it less, let me explain.
Before we delve into it, a little background, analytics has always been my first love. Writing SQL queries and watching data flow through them has been my passion. However, on the contrary, traditionally I've never been a fan of the "engineering" that goes hand-in-hand with analytics. I've always seen myself as more of a 'thinker' than a 'doer' if you will. However, recently I've realized that there are other things that are more valuable than doing analytics.
Having the ability to do "analytics" is indeed a privilege. You're assuming that the data you're working with is clean, well-structured and the problem you're working on matters. In many cases when you do analytics, you're somewhat in a fantasy bubble, distant from worrying about things like the architecture that got the data to you or the business impact that accompanies your analytics.
If you're someone who can look at a problem, build a SQL query for it, wrap it in a dashboard, and present your findings, it's undoubtedly a valuable skill that's not easy. Yet I'd argue that it's not hard enough. If you're just starting your data journey, sure it's a challenge, but if you're 1000 queries in, merely transforming and aggregating data won't suffice. This is when expanding your analytical skillset becomes crucial to making further progress. Based on my experience, this is where two distinct paths emerge.
The first thing that takes you beyond analytics is developing a "product-sense". This entails moving away from writing SQL for data problems to formulating the problem in the first place. You see, in the realm of data and analytics, having a knack for identifying the right problem often trumps having a perfect solution. Once you have a good problem to work on, writing the query becomes the easy part. If you can be the person who not only understands the data intricately but also comprehends the broader business objectives, you possess a unique skill set. This skill set empowers you to identify and articulate problems that can be effectively addressed using the available data. In this scenario, you become more valuable than someone who can only solve a problem once it's well-defined. Consider this HBR article that underscores the importance of problem formulation especially when working with AI.
The second realm to venture into beyond analytics is what one may call the "engineering" realm. If an org's data journey can be compared to the movie "Oppenheimer", analytics is just the scene where the bomb goes off, however, the rest of the story is what carries the movie, and that in this case is the engineering involved to build the data platform for others to leverage for analytics. Understanding the intricate data architecture that accompanies analytics enables you to do more than simply 'build on top of' existing structures – it empowers you to 'build' them from the ground up. This shift in perspective brings scalability to the forefront. Instead of crafting an aggregate table solely for personal use, you start contemplating how to transform it into a production-ready asset for others to leverage.
Finally, here are two more reasons to think beyond analytics. First is the AI revolution we have going on, AI is rapidly evolving to the point where it can automate many of the routine SQL queries that analysts traditionally handle. To still be relevant you need to develop a muscle beyond just writing SQL queries. Another reason to venture beyond analytics is to amplify the impact you drive with your work, which in turn makes an impact on your bank account. It's a win-win situation.
PS: If you're someone who is just starting on your data journey this is not meant to discourage you in any way. This is meant to make you aware of where your career is headed so you make the leap when the time is right.
If you found this helpful, share it. If you’re into this, see below.
While in school, my understanding of where #data lives was only limited to CSV files. However, I soon realized the data landscape in organizations is far more complex.
Check out my latest blog, I delve into the two primary realms where data resides. https://t.co/kQgbE4Yze8— Abhishek Singh (@abhishek27297) September 17, 2023