A Guide To Predictive Analytics: Definition, Importance, and Common Techniques

Many effective business courses preach the benefits of being proactive and strategic. In today’s competitive environment, it’s not enough to react to every breakthrough and ad hoc setback. Instead, organizations need to be forward-thinking: anticipating outcomes, capitalizing on opportunities, and preventing losses. With growing volumes of data and easy-to-use software, predictive analytics is more accessible than ever, helping organizations become more proactive and increase their bottom line.

What is predictive analytics?

Predictive analytics determines the likelihood of future outcomes using techniques like data mining, statistics, data modeling, artificial intelligence, and machine learning. Put simply, predictive analytics interprets an organization’s historical data to make predictions about the future. Today’s predictive analytics techniques can discover patterns in the data to identify upcoming risks and opportunities for an organization.

What are some common predictive analytics techniques?

Potential applications for predictive analytics vary widely, as do the types of models used to power resulting insights. Determining what types of predictive analytics techniques are best for your organization starts with a clearly defined objective. Once you know what question you want to answer, you can choose the model that serves you best. Predictive analytics models can be roughly grouped into these four types:

Regression models

Regression models estimate the strength of a relationship between variables. The model tracks how actions (independent variables) impact outcomes (dependent variables) and uses that information to predict future impact. These statistical models can be simple, with one independent variable and one dependent variable or a multiple linear regression with two or more independent variables. A variety of regression techniques exist and can be employed depending on the application and types of variables involved. By defining the relationship between variables, organizations can perform scenario analysis, also colloquially known as ‘what-if’ analysis, to plug in new independent variables and see how they affect the outcome. Organizations might use a regression model to determine how a product’s qualities affect the likelihood of purchase. By analyzing the relationship between the color of the product and the likelihood of purchase, an organization might see a correlation between blue shirts and more sales. Because correlation doesn't equal causation, the organization might explore how other factors affect likelihood to purchase, such as size, seasonality, or product placement. They can use these insights to help with marketing efforts or product development to determine which products might perform well in the future.

Classification models

Classification models place data into categories based on historical knowledge. Classification begins with a training dataset where each piece of data has already been labeled. The classification algorithm learns the correlations between the data and labels and categorizes any new data. Some popular classification model techniques include decision trees, random forests, and text analytics. Because classification models can easily be retrained with new data, they are used in many industries. Banks often use classification models to identify fraudulent transactions. The algorithm can analyze millions of previous transactions to learn what future fraudulent transactions might look like and alert customers when activity on their account looks suspicious.

Clustering models

Clustering models place data into groups based on similar attributes. A clustering model uses a data matrix, which associates each item with relevant features. With this matrix, the algorithm will cluster together items that have the same features, identifying patterns in the data that might previously have been hidden. Organizations can use clustering models to group customers together and create more personalized targeting strategies. For example, a restaurant might cluster their customers based on location and only mail flyers to customers who live within a certain driving distance of their newest location.

Time-series models

Time series models capture data points in relation to time. Because so much of the world’s data can be modeled as a time series, time is one of the most common independent variables used in predictive analytics. A typical model might use the last year of data to analyze a metric and then predict that metric for the upcoming weeks. Tableau’s advanced analytics tools allow organizations to forecast and explore multiple scenarios without wasting time or effort. Because time is a common variable, organizations use time series analyses for a variety of applications. This model can be used for seasonality analysis, which predicts how assets are affected by certain times of the year, or trend analysis, which determines the movement of assets over time. Some practical applications include forecasting sales for the upcoming quarter, predicting the number of visitors to a store, or even determining when people are most likely to get the flu.

Other predictive analytics techniques

Often a combination of these models are used to mine the data for insights and opportunities. For example, neural networks are a set of algorithms designed to mimic the human brain and identify patterns within the data. Neural networks use a combination of regression, classification, clustering, and time series models, so they are capable of handling big data and modeling extremely complex relationships. In fact, neural networks can handle more than just text data. With deep learning techniques, they can also input images, audio, video, and more, and training on labeled datasets allows these networks to improve their accuracy. These deep learning techniques are currently being used for voice and facial recognition software, and networks can analyze facial movements to identify a person’s disposition. With information like this, organizations can potentially predict the emotions customers will feel when using certain products or services.

Bring analytics to life with AI and personalized insights.

Try Tableau for free

Graphic of visualizations

Why is predictive analytics important?

Predictive analytics allows organizations to be more proactive in the way they do business, detecting trends to guide informed decision-making. With the predictive models outlined above, organizations no longer have to rely on educated guesses because forecasts provide additional insight. The benefits of predictive analytics vary by industry, but here are some common reasons for forecasting.

  • Improve profit margins. Predictive analytics can be used to forecast inventory, create pricing strategies, predict the number of customers, and even configure store layouts to maximize sales.
  • Optimize marketing campaigns. Predictive analytics can unearth new customer insights and predict behaviors based on inputs, allowing organizations to tailor marketing strategies, retain valuable customers, and take advantage of cross-sell opportunities.
  • Reduce risk. Predictive analytics can detect activities that are out of the ordinary — such as fraudulent transactions, corporate spying, or cyberattacks — to reduce reaction time and negative consequences.


How do I get started with predictive analytics tools?

With so many types of predictive models and potential applications, it can be difficult to know where to get started. Follow these four general steps for implementing a predictive analytics practice in your organization:

  1. Identify the business objective. Before you do anything else, clearly define the question you want predictive analytics to answer. Generate a list of queries and prioritize the questions that mean the most to your organization.
  2. Determine the datasets. Once you outline a list of clear objectives, determine if you have the data available to answer those queries. Make sure that the datasets are relevant, complete, and large enough for predictive modeling.
  3. Create processes for sharing and using insights. Any opportunities or threats you uncover will be useless if there’s not a process in place to act on those findings. Ensure proper communication channels are in place so that valuable predictions end up in the right hands.
  4. Choose the right software solutions. Your organization needs a platform it can depend on and tools that empower people of all skill levels to ask deeper questions of their data. Tableau’s advanced analytics tools support time-series analysis, allowing you to run predictive analysis like forecasting within a visual analytics interface.

Use these predictive analytics examples, methods and first steps to create a forward-thinking organization that’s ready and willing to make informed decisions using data predictions.