Introduction to Predictive Scenarios in SAP Analytics Cloud

Smart Features solutions were recently featured on our blog – I highly recommend catching up to get to know more about predictive analytics capabilities. In the following article, I’ll share more details by presenting the types of predictive scenarios available in SAP Analytics Cloud.

It’s clear that predictive analytics has been around practically forever, but the boom for Smart Features is happening right now. We live in a time when access to analytical software is becoming widespread, and these tools don’t require the advanced skills of mathematicians or statisticians. That’s why the vast majority of companies – both small businesses and industry leaders – are turning to predictive analytics to increase their bottom line and gain a competitive advantage.

Currently, there are three types of predictive scenarios in SAC:

✓ Classification
✓ Regression
✓ Time Series

Classification

Classification is a type of statistical algorithm that assigns observations to classes based on those observations’ attributes (characteristics). Looking at the topic from a business perspective, classification can be helpful, for example, when analyzing the usability of online ads or to determine the area of best local sales for our future stores. Therefore, SAP Analytics Cloud can automatically generate insights that users will operate without manually working through large data sets.

Regression

Regression is a statistical method to describe co-variations by fitting functions to them. Simply put, it allows you to predict unknown values based on observations that are found in a data set. Regression can be incredibly useful for making the right business decision, meaning that you can generate insights that you wouldn’t otherwise observe for seemingly random data. To use examples: we can analyze how traffic on highways affects the reach of a new billboard campaign or track whether the implementation of new procedures will impact quality control.

Time Series

We can define a Time Series as a sequence of ordered observations made at different moments on a variable that characterizes a certain entity or community. The predictions associated with Time Series have a ton of applications for business decision-makers. Time Series scenarios are perfect for comparing predictive data with budgeted data, so users can analyze cost forecasts or learn how to plan the company budget for the next months. Another plus is that Time Series integrates seamlessly with SAC Planning. With SAC doing most of the work for us – business users can focus on their upcoming goals without wasting their precious time doing time-consuming activities manually.

I highly recommend watching the video showing step-by-step how to implement a predictive scenario.