Five Patterns of Data Insights in Excel

If you went through some of the answers that your peers have submitted in the previous segment, you would have realised that there are a variety of insights that can be generated from a dataset.

You might also have noticed that these insights have a certain underlying pattern to them as well. What we mean here is that irrespective of the business problem that any industry is trying to solve, the insight that is generated has some common patterns. Let’s take a look at Anand's lecture to understand what those patterns are.

As explained above, any business insight can be categorised to five specific patterns. These patterns are as follows.

1. Unknown Result
2. Surprising Extreme
3. Surprising  Comparison
4. Significant Outliers
5. Abnormal Distribution

What you just learnt is a robust classification through which all the insights that you generate can be categorised into. This framework creates a ground for you to ask questions, but you also need to know the area well for asking better questions.

For example, if you have the sales revenue data for a list of products manufactured by a company, you can ask specific questions beforehand like: 

• Is there a product that is showing an abnormal distribution in its sales revenue trend? 
• Is there a product that is selling unusually better than other products?
• Is there a product that sold far better than the expected historical standards?

In this way, you can tailor your analysis process effectively so that you are on the right track in extracting insights from the data that you have.

A normal question that you might have at this point would be  - “ Ok, I know what and where to look for insights, but how exactly should I look?” which translates to this question - “ What type of procedures should I follow on the dataset so that I can generate insights?”. You’ll get to know about this in the next two segments.