Analytics 101: A newbie-friendly sharing

Forewords

Last Saturday, July-24 2021, I was invited to be a guest speaker for a workshop organized by the Accounting Club of RMIT University. It was such a great honor for me to be there and share my (little) experience with the young data aficionados.

After over a month of working together with 3 rehearsal sessions, I can say that we rocked the workshop that day. Sorry for being humble bragging 😅

The workshop aside, I realized that there are so many gaps and misconceptions between the reality and expectations of the students. Therefore, I would like to reshare (a part of) the workshop content for the public audience so all the undergraduate or freshly graduated analytics enthusiasts can have a helpful guide to kickstart their future careers.

Disclaimer:

• I made mistakes all the time, please let me know so I can fix them 😁
• All my sharing is based on my personal experience, it might not be right for all companies out there. Your mileage may vary.
• The industry has been changing so fast. What is true today might be inaccurate tomorrow. Please consider my sharing as small and subjective guidance only. Trust but verify.
• My sharing includes some examples taken from the participants, anonymized. Should you are the owner and don't like it, let me know so I can remove that part.

Analytics 101: A Brief Summary

I will take some key points from my sharing and put them in this article. Hope they will be interesting enough for you. If you are keen to spend more time, feel free to refer to the slides at the end of the article.

Note: I put the reference in each slide note (when applicable). They are extra resources for you to explore.

Part 1: A broader picture of analytics

• Simply put, analytics is the process by which businesses use statistical methods and technologies for analyzing historical data in order to gain new insights and improve the strategic decision-making process.
• The Gartner Analytic Ascendancy Model is the most common explanation to illustrate the relationship between the 4 main levels of analytics, from the most basic to the most advanced one: Descriptive, Diagnostic, Predictive, and Prescriptive Analysis.

• "Data and analytics worlds collide" - quote from Gartner Top 10 Trends in Data and Analytics for 2020. This means big companies are going to "incorporate both data and analytics tools and capabilities into the analytics stack".
• To stay competitive in the industry, analysts are expected to expand their skillsets to include solid system understanding, data structure, tools, and other technology advancements.

Part 2: Opportunities & Myths

• A quick recap on Data Analytics, Business Analytics, Business Intelligence, and Data Science...

• ... and another one on Data Engineer, Data Analyst, and Data Scientist.

• Excel or spreadsheets apps are usually overlooked by the freshers since those apps are overshadowed by Python/R or PowerBI/Tableau. However, they are very powerful and good for first jobbers.
• You don't have to run from the very beginning. Take baby steps to master the basics first then go fast. A strong foundation gives you much more energy for acceleration in the long run.
• Watch out for the similar but so different roles: Business Analyst (Phân tích [Dữ liệu] Kinh doanh) vs. Business Analyst (Phân tích Nghiệp vụ).

• Last but not least, this is my personal, generic, and simple roadmap for the newbie to consider. And yes, not all ways lead to Data Scientists.

Note: I don't mean Data Analysts are more senior than their peers of Functional Analysts. It depends on the roles and companies. The chart just shows a possible way to join and grow in the industry for a fresher with little to no technical background.

Part 3: Case Study and Feedback

• You can look for the case study with the solution in the slide. It actually just focuses on some basic data manipulation and presentation skills using common spreadsheet apps.
• From the submissions of the participant, generally speaking, there are some improvements we can make to improve the effectiveness of data presentation:

Part 4: Common Pitfalls (the newbie version)

I once wrote a (messy) note on this. Here are some to remind you and all the junior analysts:

• #1. Confirmation Bias: You have a hypothesis in mind but you are only seeking data patterns that support it – ignoring all data points that reject it. 
• #2. Statistical Significance: Using data sets that are too small to suggest a trend or comparing results that are not different enough to have statistical significance.

My all-time favorite xkcd photo, haha.

• #3. Irrelevancy and Distraction: Focusing on data that is irrelevant to the problem you are trying to solve or being distracted by data that isn’t directly connected to your analysis goal. In the age of Big Data, this is doomed to happen more and more.
• #4. Correlation does not imply Causation: Mixing the cause of a phenomenon with correlation. If one action causes another, then they are most certainly correlated. But just because two things occur together doesn’t mean that one caused the other, even if it seems to make sense.

Bonuses:

• # Apples 🍎 and oranges 🍊: Comparing unrelated data sets or data points and inferring relationships or similarities.
• # Poor data hygiene 🧼: Analyzing incomplete or “dirty” data sets and making decisions based on the analysis of that data. Garbage in, garbage out!

And also a failure case of Amazon to remind everyone of the importance of input data in analytics and later machine learning/artificial intelligence.

Part 5: References

Nothing much but my favorite resources:

Part 6: Recap

Thank you for spending your precious time going through this boring and lengthy article. Just before you forget everything and close the tab, I got a recap for you:

Analytics 101: Slides for Your References

Still interested? Here is the slide for your reference. Let me know if I made any mistakes or drop me a message if you have any questions.

Good luck, future colleagues! 😁