What You Don’t See Is More Impt!
There is a common weakness that I notice in most freshly started analysts. This weakness while makes them very excited on the first and subsequent pieces of analysis actually stunts their career growth trajectory.
What weakness is that?
It start with a strong focus on just looking at the data and nothing else. They jump into the data immediately, crunching them, writing up the codes needed to munge them into something they can analyse on, looking at the summary statistics or the parameters of the final chosen machine learning model. They then proceed to write their analysis based on those number crunched, moving to present the numbers as it is and got elated with the insights they are providing…till the Q&A comes into the picture. They get torn apart…or shredded by the audience.
I hope to keep the newsletter open to as many as possible. :) If you have benefited much from my posts and will like to join my cause, consider making a “book" donation. Link at the bottom of the newsletter or here. :)
Why is that the case? Because what is more important than the data is actually the context of the data. What about context? These are a few points:
Metadata - Data about data. For instance, “How is the data collected?”, “When in the customer process was it collected?”, “What is the time period of the data collected?”
Business process and policies - How business processes, market accessibility, customer characteristics impact the data collected? Does it produce any foreseeable biases? What is the economic climate when the data is collected? Is the insights something we can act upon? Do we have the resources to execute the policy based on the insights? How shall we benchmark the results to determine the impact attributed by executing the insights?
These are what I called the “unseen” since they are not explicit made known in the data but should come very naturally to seasoned analyst. A seasoned analyst will know how to bring in the context, to add more “oomph” to their analysis and in turn building up the trust capital with relevant stakeholders.
In the Knowledge Economy, all of us will have to do data analysis for more informed decision making. So if you are a fresh analyst looking at the dataset, please remember to bring context to your data. The more you learn how to add context to your analysis of the numbers, the better you will get and as time goes, you will be an invaluable asset to your organization.
What are your thoughts? Please share them in the comments below! Looking forward to hearing from you!
I hope you enjoy this issue! Please support by Liking, Sharing and perhaps buy me a “book” or two, to let me know I’ve made a positive. :)
Hi Koo Ping Shung,
Enjoyed reading this post.
In addition to the questions you have posed - to avoid getting shredded by the audience, I was thinking it would be prudent for new analysts to consider or research - well - their audience. What is it that the stakeholders or the business are actually interested in, and to what extent can they appreciate a technical analysis?
A deeper, and prior-to-presentation consultation or review with the experienced analysts (and stakeholders) and keeping an eye on past in-house research could also help. I was also reflecting that the entire team (including the future version of the current analyst) could benefit from a good documentation of the exploration.
Beyond all this - it felt interesting to reflect : how does one recover from a shredding? In a way, it may even be good to get such a shredding early on (both in their career, as well as in a project), since it may induce the analyst to think about context, and any other mistakes. What do you think?