5 Steps to Analyst Success
All around us, we are surrounded by information and data. In fact, there is so much information bombarding us each day that there is even information about recovering from information overload. With terms like Big Data, Data Scientist, Machine Learning, and others dominating every corporate conversation, it is not difficult to see that, sometimes, there might be slipups in the analysis or understanding of this information. This can cause issues like what we have seen with recent interpretations of Uber data.
As a current social-research lead at Adobe and a former Godfather of the Adobe Digital Index team that produced thought leadership for outlets like the Wall Street Journal, CNBC, New York Times, and Forbes, I decided I would put together a short list of ways to ensure the best insights are created from the mountains of data that we now have access to.
- Have an intimate knowledge of the data you analyze. When I first started analyzing Adobe Marketing Cloud data, I attended every training possible about the products that were producing the data. I talked with engineers, salespeople, even customers to better understand how they used the products and implemented them. This knowledge helped me to realize what data was available to analyze, how to identify whether data was incorrect or needed cleaning, and most importantly, how to create a working relationship with the guardians of the internal data.
- Learn statistics. You do not need to be able to create complex models, write code, use R, or be able to predict the next economic downfall to be a great analyst. A basic understanding of means, averages, outliers, histograms, standard deviations, and other statistical terms will go a long way in helping you to create a true trend, free of outliers from a large dataset. You can go the formal route through an online degree such as a masters in stats from my alma matter, Utah, or through free routes on Coursera.
- Be curious. Adobe has an unmatched dataset with Adobe Marketing Cloud, so I have a great playground to learn about trends across all industries. However, I was only able to succeed in analysis because of my curiosity for knowledge through data. Over the last four years, among other things, I have predicted movie success with social buzz, analyzed Brazilian sentiment around the World Cup, shown that broadcasters still rule the content game in social, found the hot knew items in IoT, and determined which Super Bowl ad was the best based on six points of social data. These ideas were all cultivated by my curiosity for proving points with data.
- Always question results. Even after cleaning, averaging, and removing outliers, you should still always question your results. Make sure they pass the sniff test and compare them to other historical results, if available. Also, if possible, ask colleagues to do a peer review of your work to ensure that you feel confident and comfortable with what you share internally or externally.
- Find your passion. Social data is my passion, and it took me about two years to figure that out. I was able to analyze other data just fine, but social really got me excited. Everyone will be different, and it can take time to find what you truly enjoy. Once you find that passion, follow it and examine it as much as you can. You will find that, when you are working on something you have passion for, it will be the most natural path to success in becoming an analyst.
This list is, by no means, comprehensive but can at least provide a glimpse in to ways to produce meaningful insights from data. Curiosity and passion are the two traits that really cannot be taught. You will have to look inside yourself to gain an understanding of what truly drives you to find results in data.
This post taken from http://blogs.adobe.com
Kalyan Banga205 Posts
I am Kalyan Banga, a Post Graduate in Business Analytics from Indian Institute of Management (IIM) Calcutta, a premier management institute, ranked best B-School in Asia in FT Masters management global rankings. I have spent 6 years in field of Analytics.