Predictive Analytics

Predictive analytics focuses on creating actionable models to predict future behaviors and events
Predictive analytics employs mathematical and statistical algorithms, neural networks, artificial intelligence, and other advanced modeling tools to create actionable predictive models based on available data
Most predictive analytics models are used in areas such as healthcare services, customer relationships, customer service, customer retention, fraud detection, targeted marketing, and optimized pricing

he term analytics is used to characterize a vast array of methods that use data to help make better business decisions and there are many ways to organize them into subcategories. This framework distinguishes analytics into three large classes:

Descriptive analytics - what has happened
Predictive analytics - use data to predict what will happen
Prescriptive analytics - using optimized algorithms to optimize a desired objective given predictions of what will likely happen

Data Science / Data Analytics is a booming industry. If you feel confused and overwhelmed with all of these concepts, you are not alone. In this section we attempt to put a bit of order into this confusing landscape. When it comes to data, there are four key steps:

Store
Transform
Explore and Analyze 
Present/Visualize

Before trying to figure out what the data is telling you about your specific problem, it is helpful to gain an understanding of the role of data extraction  to understand the capabilities of data more broadly. Often data is used towards establishing a baseline against a particular metric, or KPI, but these macro-level statistics generally provide descriptive insights as to how well a certain business function is performing, and do not provide actionable guidance. 

The first key insight in is the concept of data hubris, or “a false dichotomy between data and intuition,” where the quantitative analyst sees the world through the lens of data and the executive just sees the world through experience and intuition. The key lessons are to use analogies, anecdotes, or abstraction to conceptualize findings within data, and (less critically) use a question-based approach to guide decision-makers to the right answer.  The bottom line question that the data translator/analyst should ask themselves is: “how does data help the person I’m speaking to?”

Descriptive Analytics

The core idea of Business Intelligence (BI) is the insightful visualization of aggregations of one or more measures across one or more dimensions. For example, in an organization, we would likely have sum of sales amount aggregated across years, continents, channels and product categories. In the simplest BI applications, measures of interest are directly stored as fields in one or more of the tables of the data model. However, as things become more complex, analysts often need to define measures that are not directly stored in the model. These are called computed or calculated measures.

Descriptive Analytics is the examination of data or content, usually manually performed, to answer the question “What happened?” (or What is happening?), characterized by traditional business intelligence (BI) and visualizations such as pie charts, bar charts, line graphs, tables, or generated narratives.

90% of organizations today use descriptive analytics, the most basic form of analytics. The simplest way to define descriptive analytics is that it answers the question “What has happened?”. This type of analytics analyses the data coming in real-time and historical data for insights on how to approach the future. The main objective of descriptive analytics is to find out the reasons behind precious success or failure in the past. The ‘Past’ here, refers to any particular time in which an event had occurred and this could be a month ago or even just a minute ago. The vast majority of big data analytics used by organizations falls into descriptive analytics.

Example:

Descriptive analytics is also useful in market research. When it comes time to glean insights from survey and focus group data, descriptive analytics can help identify relationships between variables and trends. For instance, you may conduct a survey and identify that as respondents’ age increases, so does their likelihood to purchase your product. If you’ve conducted this survey multiple times over several years, descriptive analytics can tell you if this age-purchase correlation has always existed or if it was something that only occurred this year.

Predictive Analytics

We are now going to take a closer look at the predictive analytics and distinguish it from the descriptive analytics. Whereas descriptive analytics interprets data already in hand, predictive analytics constructs mathematical models to make forecasts. Predictive analytics encompasses a variety of statistical techniques from data mining, predictive modeling, and machine learning that analyze current and historical facts to make predictions about future or otherwise unknown events

Predictive analytics helps predict the likelihood of a future outcome by using various statistical and machine learning algorithms. Still, the accuracy of predictions is not 100%, as it is based on probabilities. Algorithms take data and fill in the missing data with the best possible guesses to make predictions.

Analyzing historical trends and patterns can accurately inform a business about what could happen in the future. This facilitates the establishment of realistic goals for a company, strategic planning, and expectations handling. Predictive analytics is used by businesses to study the data and look into the crystal ball to find answers to the question, “What could be the future outcome based on previous trends and patterns?”

Example:

Every business needs to keep periodic financial records, and predictive analytics can play a big role in forecasting the organization’s future health. Using historical data from previous financial statements, as well as data from the broader industry, you can project sales, revenue, and expenses to craft a picture of the future and make decisions.

Prescriptive Analytics

Prescriptive analytics s 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 ____ happen?”, and is characterized by techniques such as graph analysis, simulation, complex event processing, neural networks, recommendation engines, heuristics, and machine learning.

Prescriptive analytics has been called “the future of data analytics,” and for good reason. This type of analysis goes beyond explanations and predictions to recommend the best course of action moving forward. It’s especially useful in driving data-informed decision-making. It is the process of using data to determine an optimal course of action. By considering all relevant factors, this type of analysis yields recommendations for next steps. Because of this, prescriptive analytics is a valuable tool for data-driven decision making

Prescriptive analytics is the next step of predictive analytics that adds the spice of manipulating the future. Prescriptive analytics advises on possible outcomes and results in actions that are likely to maximize key business metrics. It uses simulation and optimization to ask, “What should a business do?” Simulating the future, under various sets of assumptions, allows scenario analysis - which, when combined with different optimization techniques, allows prescriptive analysis to be performed. The prescriptive analysis explores several possible actions and suggests actions depending on the combined results of predictive and descriptive analytics of a given dataset.

Example:

An algorithmic use of prescriptive analytics is the detection and flagging of bank fraud. With the sheer volume of data stored in a bank’s system, it would be nearly impossible for a person to manually detect any suspicious activity in a single account. An algorithm - trained using customers’ historical transaction data can analyze and scan transactional data for anomalies. For instance, perhaps you typically spend $2,000 per month, but this month, there’s a $30,000 charge on your credit card. The algorithm analyzes patterns in your transactional data, alerts the bank, and provides a recommended course of action. In this example, the course of action may be to cancel the credit card, as it could have been stolen.

The Bottom Line - Data Analytics:

What’s The Difference?

The names themselves begin to give away the distinct differences between these two forms of analytics.

Descriptive analytics use data aggregation and data mining. Descriptive analytics lets businesses see what has already happened.
Predictive analytics use statistical models and forecasting. Predictive analytics helps businesses see what could happen in the future.
Prescriptive analytics allows you to control what is being molded.

Descriptive analytics takes data an organization already has and presents it to them in an easy-to-digest way. This is helpful for looking at a past year’s sales and other historical data, but when it comes to making marketing decisions and other key business moves, you need more than historical data points.

This is where predictive analytics come in. Rather than just showing you how many sales a business made last year or even how many clicks a website got last month, predictive analytics looks to take large amounts of data and dig deeper into its meaning. Based on the data, it will make insight-based predictions that can help guide critical moves.

With the use of advanced statistical models and forecasting techniques, predictive analysis helps a business see the future possibilities so that they can better consider their next potential moves. In this way, predictive analytics have an invaluable role in informing business decisions.

The goal of Data Analytics (big and small) is to get actionable insights resulting in smarter decisions and better business outcomes. How you architect business technologies and design data analytics processes to get valuable, actionable insights varies.