Demystifying Predictive Analysis

The ability to predict future events and trends is important across industries. From weekly weather forecasts to algorithm-driven medical advances, predictive analytics is popping up more often than you might think. Here’s an overview of predictive analytics to help you start making data-driven strategies and decisions.

What is Predictive Analytics

Predictive analytics is the use of data to predict future trends and events. Use historical data to predict potential scenarios to help you make strategic decisions. Businesses use predictive analytics to find data patterns and identify risks and opportunities.

Predictive analytics is often associated with big data and data science. Businesses today are flooded with data that resides in transactional databases, device log files, images, videos, sensors, or other data sources. To gain insight, data scientists use deep learning and machine learning algorithms to find patterns and predict future events. These include linear and nonlinear regression, neural networks, support vector machines, and decision trees. Insights gained through predictive analytics can be further used within prescriptive analytics to drive actions based on predictive insights.

One predictive analysis tool is regression analysis, which allows you to determine the relationship between two variables (simple linear regression) or three or more variables (multiple regression). Relationships between variables are described as mathematical formulas that help predict outcomes when variables change.

“Regression allows us to gain insights into the structure of that relationship and provides measures of how well the data fit that relationship,” says Harvard Business School Professor Jan Hammond, who teaches the online course Business Analytics, one of the three courses that make up the Credential of Readiness (CORe) program. “Such insights can prove extremely valuable for analyzing historical trends and developing forecasts.”

Forecasts help you make better decisions and develop data-driven strategies. Here are some examples of predictive analytics in action to get you started in your business.

5 EXAMPLES OF PREDICTIVE ANALYTICS IN ACTION

Finance: Forecast Future Cash Flows

Every business needs to keep financial records and predictive analytics plays a big role in predicting the future health of the business. You can use historical data from previous financial statements and industry-wide data to forecast sales, income and expenses, to help you anticipate the future to make decisions.

HBS Professor V.G. Narayanan mentions the importance of forecasting in his Financial Accounting course, which is also part of CORe: “Managers need to look ahead to plan for the future health of their business,” he said. “No matter what field you work in, there is always a lot of uncertainty in this process.”

Entertainment & Hospitality: Identifying Talent Needs

An example being studied in Business Analytics is the use of predictive analytics by casino and hotel operator Caesars Entertainment to determine the number of people required at a venue at a given time.

In the entertainment and hospitality industry, customer inflows and outflows are influenced by many factors, all of which affect the number of staff a venue or hotel needs at any one time. Overstaffing is costly, and understaffing can lead to a poor customer experience, employee overwork, and costly mistakes.

To predict the number of hotel check-ins on a given day, the team developed a multiple regression model that considers multiple factors. This model allowed Caesars to do its best to staff its hotels and casinos and avoid overstaffing.

Marketing: Behavioral Targeting

In marketing, consumer data is abundant and used to create content, ads and strategies to better reach potential customers wherever they are. We do predictive analytics by examining past behavioral data and using it to predict what will happen in the future.

Predictive analytics can be used in marketing for forecasting sales trends at different times of the year and plan campaigns accordingly.

In addition, historical behavioral data helps predict the likelihood of potential customers moving down the funnel from awareness to purchase. For example, a single linear regression model can be used to find the number of engaging pieces of content that predict with statistically significant confidence a potential customer’s likelihood of becoming a customer later. With this knowledge, you can plan targeted advertising for different points in the customer lifecycle.

Manufacturing: Avoiding Malfunctions

Although the above example uses predictive analytics to act based on likely scenarios, predictive analytics can also be used to prevent unwanted or harmful situations from occurring. For example, on a manufacturing floor, algorithms can be trained on historical data to accurately predict when machines are likely to fail.

When criteria for imminent malfunction are met, algorithms are triggered to alert employees who can stop the machine. This could save the company thousands, if not millions, of dollars in damaged product and repair costs. This analysis predicts outage scenarios today, not months or years ago. Some algorithms even recommend fixes and tweaks to avoid future malfunctions, improve efficiency, and save time, money, and effort. This is an example of prescriptive analytics. Most often, one or more types of analysis are used together to solve a problem.

Healthcare: Early Detection of Allergic Reactions

Another example of using algorithms for fast predictive analytics for prevention is the healthcare industry. The Wyss Institute at Harvard University has partnered with the KeepSmilin4Abbie Foundation to develop a wearable His technology that predicts anaphylactic allergic reactions and automatically administers life-saving epinephrine. Called AbbieSense, the sensor recognizes early physiological signs of anaphylaxis as predictors of later reactions much faster than humans. Algorithmic responses are triggered when responses are predicted to occur. Algorithms can predict the severity of reactions, alert people and caregivers, and automatically inject adrenaline when the need occurs. The technology’s ability to predict responses more quickly than manual detection could be lifesaving.

Use Predictive Data for Future Strategies

No matter your industry, predictive analytics can provide the insights you need to take the next step. Whether you’re making financial decisions, developing marketing strategies, changing operations, or saving lives, building a foundation of analytical skills are critical to your future success.