Data in analytics

Have you ever gone to see your doctor but ended up not being treated correctly for your ailment? The fault might lie on the fact that you have been describing your symptoms incorrectly and, thus, the doctor prescribed you the wrong cure.

Similarly, in business, data provide value to the business. The accuracy of data input is of great importance as an analysis cannot be correctly made with incomplete data. Data in analytics is the science of examining data to be able to draw conclusions from the information. This helps organizations make better business decisions and improve business operations. 

There are three major types of analytics: descriptive, predictive and prescriptive. They coexist and are studied in a consecutive manner to obtain a complete overview of a situation to make better decisions.

Recently, I have been assigned to a new division and one of my main goals is to increase sales, be it by geographic expansion or through new product development.  Since our department is fairly new, I gave my staff the task of compiling the year to year sales data, date of order, quantity sold and product variants. Each staff handled different accounts, thus, I asked them to segregate the sales by account. 

After they submitted their reports, I was in a quandary since the data showed inconsistencies. Upon further inspection, I realized that in the reports, some of them gave me quantity sold as per purchase order date, while others gave me quantity sold as per delivery date, and another gave me quantity sold upon payment date.  

They all gave me correct information but the reports were not consistent since the quantity sold varied indicating date of order, or date of delivery, or date of payment. They gave data values that are both correct and unambiguous, and yet it became problematic.  Why? Because inconsistencies in reporting create inaccurate data.

We addressed this issue by agreeing upon a consistent format of reporting. I was now able to draw some analysis from the reported information. How much sales did we have for a particular account? What are the most salable product variants? Are there any particular sales trend? By answering these questions, I was able to technically know “what happened” in those years. This is what is known as descriptive analytics. It turns data into an actionable insight.

With this data, I try to understand why certain variants sells more than another.  Is it the price? Is this variant more appealing to the consumer? Is this packaging size more convenient to carry? After analyzing the historical data patterns and trends, I now try to make predictions about what and where the product variant will sell more. Predictive analytics is the ability to use descriptive data to predict “what might happen next?”  Predictive analytics become more accurate with more data. With accurate and consistent descriptive data, I can do a better sales forecast.  

By using the data from descriptive analytics and insights from predictive analytics, I proceeded to formulate strategies that will increase sales. “What should be done?” is the hallmark of prescriptive analytics. Prescriptive analytics suggests the best actions to take for improvements as well as the proper actions needed to avoid negative circumstances. Since our target market are the millennials, I recommend the implementation of digital campaigns. Our data show the relationship of increased digital exposure to higher consumer activation and purchase.  Yes, our sales have increased!

Prescriptive analytics needs a firm understanding of the past and the present to be able to realize the variations and combinations of circumstances that can lead to the desired outcome. This analytics approach recognizes that different scenarios can lead to different predictions. 

Just as doctors need to eliminate the cause of a fever, they also need to be aware of the patients’ allergies and other medications taken before giving a prescription. Thus, different patients might have the same sickness but the doctor may prescribe a different cure because of these variables. 

So, when you go see your doctor for any form of ailment, describe your symptoms accurately and correctly. A doctor will only be able to diagnose the sickness from your description and analyze those symptoms to diagnose your illness. Then and only then can the doctor prescribe the best cure for you.

Regina C. Dy, PhD. is a consultant at Euro-Med Laboratories Phils., Inc. She is a part-time faculty of the Ramon V. del Rosario College of Business, De La Salle University.  She can be contacted at [email protected] The views expressed above are the author’s and do not necessarily reflect the official position of DLSU, its faculty, and its administrators. 

Topics: Green Light , Data in analytics
COMMENT DISCLAIMER: Reader comments posted on this Web site are not in any way endorsed by Manila Standard. Comments are views by readers who exercise their right to free expression and they do not necessarily represent or reflect the position or viewpoint of While reserving this publication’s right to delete comments that are deemed offensive, indecent or inconsistent with Manila Standard editorial standards, Manila Standard may not be held liable for any false information posted by readers in this comments section.
AdvertisementSpeaker GMA