Predictive analytics is a branch within data analytics that uses known data to make predictions about future outcomes. Combining statistics, data mining and machine learning, we can use predictive analytics within Alteryx to:

  1. Identify patterns and trends through Data Investigation tools
  2. Build models to make predictions through the Predictive tools

Data Investigation Tools

Predictive Tools


Utilising the Data Investigation and Predictive tools within Alteryx, we can calculate a probability of an event occurring (predictive score) and generate future insights to a significant degree of precision. We’ll explore these tools further in the next blog.

Let’s look at some definitions:

Term Definition
Predictive analytics The practice of extracting information from existing data to determine patterns trends that could potentially predict future outcomes
Statistics A branch of mathematics dealing with the collection, analysis, interpretation, and presentation of masses of numerical data
Machine Learning A branch of artificial intelligence (ai) where computers learn to act and adapt to new data without being programmed to do so. The computer can run independently of human interaction
Data Science The study of big data that seeks extract meaningful knowledge and insights from large amounts of complex data in various forms
Data Mining The process of discovering patterns in large data sets

Within the business world, some outcomes from using predictive analytics could save money, identify risks, or improve business functions for the organisation.

Industry insight: Predictive analytics within the global market is projected to reach approximately $10.95 billion by 2022, growing at around 21% annually between 2016 and 2022, according to a 2017 report issued by Zion Market Research.

To get a better understanding of predictive analytics, lets first explore some real-world applications:

Aviation Industry

Aircraft manufacturers use predictive analytics to predict when airplanes need maintenance. This helps them take a proactive approach to maintenance vs reacting to equipment failure. Thanks to predictive analytics, manufactures can reduce unexpected equipment failure, maintenance costs and downtime.


Marketing departments use data to analyse performance of a range of business activities. For example, marketing departments collect data on customers, campaigns, products, and events to name a few.

Using predictive analytics, marketing departments can predict:

  • forecast customer lifetime value,
  • future customer purchases – think online shopping when they suggest other products based on what you are buying, and
  • future income based on a broad range of variables.

These examples are only a few of many in the marketing space where predictive analytics can provide a high degree of precision.

It is worth noting that forecasting is different from predictive analytics. Traditional forecasting looks at the numbers, trend, and seasonality observations to predict outcomes. In Comparison predictive analytics is more about consumer behaviour and may uses explanatory variables to predict outcomes.

Weather forecasts

Weather forecasters analyse historical weather patterns and satellite imagery to predict the weather for several days in advance with high precision – most of the time. As an added benefit, we use this data to further understand the impacts of global warming.


Introducing the concept of predictive analytics, we have explored use cases and real-world examples. The proceeding blogs in this series will expand on the Alteryx tool palettes shown above as we breakdown and work through a hypothetical problem to make a prediction on a dataset.

Note: Alteryx does provide Machine Learning specific tools to analyse and make predictions on data. These tools are additional add-ons to the standard Alteryx Designer license. Touching on these tools briefly, they simplify the approach to predictive analytics and to my knowledge provide similar outcomes to the way we will be exploring in this blog series.


Scott Johnston
Author: Scott Johnston