Data was never given the level of consideration that it is given now, and this thought often leaves me to wonder why?
The answer to it is as simple as time. It’s never too late to identify the underdogs and give them due respect.
In my four years of industry experience (yes, I’m a young professional!), I have analyzed data for several purposes such as identifying outliers, addressing them, creating frameworks, optimizing processes, and publishing dashboards, but my methods lacked a scientific approach. So, I had to upgrade. Despite achieving desired results and making an impact inside the organization, there were efficient ways to analyze the data that were hidden from me and had to be explored. I rolled up my sleeves and decided to dive into it. My recent graduate studies helped me hone my data analysis skills and approach problems with a much more structured approach. I still have my sleeves rolled up as every single day I learn something new and realize there are so many more ways to unlock business value from data. I’d like to share some thoughts based on my personal experiences and learning.
Intuition is good, but data is better
Many a time we rely on our gut instinct to make decisions. I, professionally, have taken some important decisions that drove business strategy and a few of them have been based purely on gut instinct. Human instinct is certainly a powerful tool, but decisions backed by data are often better. This is something I have learned over the years, and trust me, this makes me feel so much more confident in my analysis now.
Make data your best friend
As is rightly quoted by Kurt Bollacker “Data that is loved tends to survive.” How you decide to treat your data can vastly affect what conclusions you make. I have always been a cleanliness freak and that applies to my work style as well. I like my data visually appealing, so I allow pre-processing to consume 40% of my entire time dedicated to a project. Turning raw data into tidy and consistent data is one huge achievement I always enjoy.
Understanding where your data is coming from, what all sub-links it is connected to, why you selected that piece of data, it’s range of impact, really helps you not get overwhelmed, stay composed, and organize your actions.
There is no such thing as a ‘failed analysis’
Data has always been around. If you enjoy playing with data, studying trends, making predictions, creating dashboards, just go for it. You will never fail at obtaining insights. Knowing the basic high-level flow of the process (problem identification, data collection, data exploration/processing, model generation, data visualization) can make one extremely comfortable with any kind of data at hand. If a valid dataset yields results, you may come out knowing something you didn’t earlier.
Having a great story and charts added to your analysis may give answers to questions you didn’t know you had. I especially look forward to this particular section of the data analysis process because business intelligence and data visualization is something that has always psyched me up. This excitement has often led me to explore various tools/software available out there. The ones that I have worked on and continue being my favorites are TIBCO Spotfire, Tableau, QlikView, PowerBI, Knime, and Excel. There are many more tools getting launched every other day, and this only makes me more passionate about making my data talk the clearest language using the best visualizations possible. Data visualization is a quick and easy way to convey concepts in a universal manner.
After all, data is all around us and turning data into insights and those insights into action is not as difficult as it seems