KNOW YOUR DATA 101: Predictive Data, Descriptive Data and Prescriptive Data

Overview

To understand the data, be it an analyst, scientist, or visualizer, it is important to know where it resides, what is in it, and what it means.

What does “knowing your data” mean?
There are many layers to this. They are – Know the data layout: where it lives, how is it organized? Data is useless when you do not know what to do with it. Knowing your data records is integral. A good way to start is playing with your data to understand it, performing some data density assessments etc. This will help you find the time needed to clean and prepare your data the right way. It also helps you to know your data quirks better as it affects downstream systems.

Data analytics refers to extracting, preparing, modeling, and drawing conclusions from the data to make better decisions. There are more than one type of data analytics and they are descriptive, predictive, and prescriptive. When you look at the different types of data analytics and how each kind can help you uncover various insights for the business, you notice how it paints a very conclusive picture of your business from the past and present. This enables you to make more informed decisions armed with analysis that will shape the future of your business.

Descriptive Analytics

Descriptive analytics is a field of statistics that focuses on gathering and summarizing raw data to be easily interpreted. Generally, descriptive analytics concentrate on historical data, providing the context that is vital for understanding information and numbers. Let us say you go to the doctor and check your vitals like your weight, height, blood pressure, resting heart rate etc. Descriptive analytics are essentially your vitals. It is the simplest form of analytics. For a business, these are the analytics that are essential and need to be collected and reported. They are designed to give you the very basics of who, what, where, when, how, and how many. It provides you with key metrics about your business and serve as a basis by which you can dive deeper. You will find this kind of analytics in all business intelligence (BI) tools.

How does it work?

Data aggregation and data mining are two techniques used in descriptive analytics to discover historical data. Data is first gathered and sorted by data aggregation that makes the datasets more manageable by analysts. Data mining is the next step of the analysis and involves a screening of the data to identify patterns and meanings. Identified patterns are analyzed to discover the specific ways that users interacted with the content and within the environment.

Use cases

Common use cases of descriptive analytics are reports that provide historical insights regarding the company’s production, financials, operations, sales, finance, inventory, customers etc.

Sentiment analysis is another widespread use case for descriptive analytics today. Analyzing customer sentiment on platforms either by involving or mimicking popular social media channels is crucial to understand how best to engage people.

By leveraging various forms of text analytics, natural language technologies and even machine learning, organizations can master sentiment analysis with descriptive analytics, demonstrating “what the most and least active areas of the business are, so that you can start targeting the areas that are the least active, and need an extra push for engagement.

Use Descriptive Analytics when you need to understand what is happening at the fundamental levels of your company, and when you want to summarize and describe different aspects of your business.

Predictive Analytics

Predictive analytics is the use of data, statistical algorithms and machine learning techniques to identify the likelihood of future outcomes based on historical data. The goal is to go beyond knowing what has happened to provide the best assessment of what will happen in the future. Using predictive analytics, you can find tendencies, clusters, and exceptions to predict future trends. Here is where our doctor analogy falls apart a little bit though. If your doctor took your vitals, test results, health data and other similar baseline data about you. After processing that information, she could very likely predict health risks that you may have in the future. 

Predictive analytics are great forecasting tools. Looking for trends? Planning for the future? Then, Predictive analytics is what you need. This also enables statistical modeling. They require the use of machine learning and artificial intelligence and can be a huge asset to your business. 

One of the truly great uses for any business operations tool set is the ability to generate predictive analytics, which enables businesses to identify potential events and opportunities, and either avoid or capitalize on them. 

How does it work?

Predictive Analytics is a statistical method that utilizes algorithms and machine learning to identify trends in data and predict future behaviors. 

The analytical models run one or more algorithms on the data set on which the prediction is going to be carried out. It is a repetitive process because it involves training the model. Sometimes, multiple models are used on the same data set before one that suits business objectives is identified. It is important to note that predictive analytics models work through an iterative process. It starts with pre-processing, then the data is mined to understand business objectives and this is followed by data preparation. Once preparation is complete, the data is modelled, evaluated, and finally deployed. Once the process is completed, it is iterated on again. 

Data algorithms play a huge role in this analysis because they are used in data mining and statistical analysis to help determine trends and patterns in data. There are several types of algorithms built into the analytics model incorporated to perform specific functions. Examples of these algorithms include time-series algorithms, association algorithms, regression algorithms, clustering algorithms, decision trees, outlier detection algorithms and neural network algorithms. Each algorithm performs a specific function. For example, outlier detection algorithms detect the anomalies in a dataset, while regression algorithms predict continuous variables based on other variables present in the dataset. 

Use Cases

Common use cases of predictive analytics are fraud detection, Risk management, Dynamic product pricing, Predictive maintenance, Patient utilization patterns, Churn prevention. 

Use Predictive Analytics any time you need to know something about the future or fill in the information that you do not have. 

Prescriptive Analytics

Prescriptive analytics focuses on finding the best course of action in a scenario, emphasizing on actionable insights instead of data monitoring. It represents a more advanced use of predictive analytics. It is a form of advanced analytics which examines data or content to answer the question “What should be done?” or “What can we do to make it happen? Say, after your doctor has taken vitals, run tests, and drilled down to the root of the problem, and prescribes a medication or change in diet and lifestyle. It is the same for prescriptive analytics. They prescribe what action to take to eliminate a future problem or make the most of a promising trend. 

How does it work?

Prescriptive analytics makes use of machine learning to help businesses decide a course of action based on a computer program’s predictions. Prescriptive analytics uses the latest technologies such as machine learning and artificial intelligence to understand what the impact is of future decisions and uses those scenarios to determine the best outcome. When used effectively, prescriptive analytics can help organizations make decisions based on facts and probability-weighted projections, rather than jump to under-informed conclusions based on instinct. 

Use Cases

Common use cases of prescriptive analytics are Optimization of Travel and Transportation, Product availability and price optimization, Improving Health Industry – offer better healthcare for less money, they will be able to improve future capital investments for new facilities or hospital equipment and improve the efficiency of hospitals. 

Finally

As you can see, these different types of data analytics can provide deep insights that will uncover trends and help you make better decisions. Businesses these days are utilizing data to discover insights that can aid them in creating business strategy, making decisions, and delivering better products, services, and personalized online experiences. While business analytics is a broad field, when we look at these three distinct methodologies – descriptive, predictive, and prescriptive – their potential usefulness is clearly vast. When used in combination, these different methods of analysis are extremely complementary and valuable to business success.