1. What happened?
Welcome! In this video, we will deep dive into questions that focus on “What happened” and can be answered using descriptive analytics.
2. Understanding descriptive analytics solutions
Descriptive analytics solutions that can tell us what has happened can be invaluable for tracking trends, figuring out what works and what doesn’t, and providing a general overview of the business performance.
We will go through the workflow in detail from the step of receiving the business question up to delivering the descriptive analytical solutions. We will see applications of descriptive analytics from different industries such as healthcare, manufacturing, and fashion.
3. Healthcare industry example
Let's start with a practical application that applies to the healthcare industry. The patient experience team wants to improve patient satisfaction. In this case, the analytical question can be "What are the most common complaints of patients based on the last 3 months' complaints data?". Since the customer complaints feedback data is qualitative, we can use visualization techniques such as a word cloud to show the results. A word cloud displays a collection of words in a graphical representation, where the size of each word is proportional to its frequency within the collection.
4. Manufacturing industry example
Another application of descriptive analytics can be in manufacturing. For example, the operations team wants to improve supplier delivery efficiency. The analytical question can be, "Which suppliers have the most delayed deliveries based on the last six months' data?" To answer this question, a Pareto chart can be used to identify which suppliers have the greatest impact on delayed deliveries. This can be done by calculating the percentage of the total delayed deliveries caused by each supplier and then sorting the suppliers in descending order based on their impact on delayed deliveries. From the red line in our example, we can see that three suppliers, suppliers, 11,12, and 13, are causing more than 70% of the delays.
5. Fashion industry example
Now let’s consider another example in the fashion industry. A company that designs and sells clothing for men and women wants to develop customized products that meet each customer's unique needs and preferences. To help develop customized products, the analytics team can group the customers into clusters. The analytical question can be: “What are the distinct groups of customers based on their purchasing behavior and style preferences considering this year’s customer sales data?”. By applying a clustering analysis, the customers can be grouped into distinct clusters based on their similarities and differences in purchasing behavior and style preferences. This may also involve visualizing the clusters using scatter plots and interpreting the characteristics of each cluster.
6. Scenario - PureHealth (1/2)
Great! Now let’s deep dive further with a scenario. Imagine an online store called “PureHealth” that sells health and wellness products such as supplements and organic foods. A business question from the sales team is the following: “How can PureHealth improve website engagement to increase sales?”. Let’s assume that website engagement is measured by the average session duration, which is essentially the amount of time a user spends on a website during a single visit. To address this question, we can start by finding out which pages currently drive higher session duration than others.
7. Scenario - PureHealth (2/2)
Let’s now form the SMART analytical questions. “Which product pages on the website have the highest and lowest session duration based on the last six-month traffic data?” We can then go a step further and ask: “What is the demographic profile of users who visit high session duration pages compared to those who visit for less duration during the same period?” This way, we can check if high or low traffic is caused by specific customers for those pages. Data aggregation and visualization would be the descriptive analytical solution in this case to present the results.
8. Let's practice!
Time to practice what you've learned!