Predictive Analytics Software Buying Guide

by Drew Robb
Predictive Analytics Software Buying Guide

Shopping for predictive analytics software? Here are 10 vendors you should know.

While many vendors sell analytics software, not all of them provide predictive analytics, a far more specialized field. Unlike more traditional analytics, predictive analytics typically involves sophisticated techniques such as predictive modeling, data mining, machine learning and statistical algorithms. The focus is on predicting future outcomes or trends. 

So who are the key players? Forrester Research includes SAS, RapidMiner, FICO, Alpine Data Labs, Alteryx and Microsoft among its predictive analytics leaders and important challengers.

The only vendor with serious market share is SAS, with 33 percent of the predictive analytics software market. Next comes IBM, which commands 15 percent of the market despite the fact it doesn’t bill itself as a predictive analytics specialist.

Outside those two, no vendor holds more than a 2 percent share of the market. Clearly, there is room for advancement for vendors that provide predictive analytics software. The market is growing; more companies are becoming interested in predictive analytics as vendors introduce more user-friendly technologies.

Selecting a predictive analytics provider is only part of a successful initiative. We recently shared tips for getting your predictive analytics project started as well as tips on data preparation for predictive analytics.

In this software buying guide, we'll look at product offerings from five vendors:

  • Market leader SAS
  • Information Builders
  • Alpine Data Labs
  • MicroStrategy
  • RapidMiner


SAS offers a portfolio of predictive analytics software for different user roles. Its flagship is SAS Enterprise Miner, an integrated workbench for data mining and machine learning. It streamlines the predictive analytics process to create analytical models from vast amounts of data.

This is accomplished via a collection of statistical, data mining and machine learning algorithms. Decision trees, bagging and boosting, time series data mining, neural networks, memory-based reasoning, hierarchical clustering, linear and logistic regression, associations, sequence and Web path analysis and more are involved. There are even industry-specific algorithms such as credit scoring.

In 2015 SAS unveiled an add-on to SAS Enterprise Miner to help automate massive model development projects in analytics-driven organizations. SAS Factory Miner provides a Web based, automated, customizable, environment for building, comparing and retraining predictive models at scale across multiple business segments. The software tests multiple models simultaneously and identifies the best performing model for each segment based on predefined performance statistics. If desired, it allows manually fine-tuning predictive analytics models.

SAS users can take advantage of workflow templates provided with the product, or they can interactively build custom templates that include the various steps of a predictive analytics workflow, such as data transformation, data filtering, variable selection and model building. All assets are created automatically to deploy selected models into SAS or third-party production environments such as Hadoop, Web services or a database like Oracle, DB2, Greenplum or SAP HANA.

Information Builders

Part of the Information Builders (IBI) suite of business analytics solutions, WebFOCUS RStat is an integrated BI and data mining environment that is said to bridge the gap between backward and forward-facing views of business operations.

"With RStat, companies can access and cost effectively deploy predictive models as intuitive scoring applications, so business users at all levels can make decisions based on accurate, validated future predictions instead of relying on gut instinct alone," said Bruce Kolodziej, Predictive Analytics sales manager, IBI.

WebFOCUS RStat provides a single platform for business intelligence, data modeling and scoring. It is integrated with App Studio and WebFOCUS Reporting Servers, with access to more than300 data sources for BI developers and data miners. It offers data exploration, descriptive statistics and interactive graphs, hypothesis testing, clustering and correlation analysis. It has the ability to build and export models for prediction and classification, evaluating those models and incorporating their findings into reports, Kolodziej said.

Alpine Data Labs

Alpine's predictive analytics software aims to take a full stack approach with the Alpine Data Platform comprised of three layers: Alpine Core, Alpine Connect and Alpine Touchpoints. Alpine Core is the central analytics engine that allows users to run analytics in-database and in parallel, and build predictive models using a visual development interface. Alpine Connect is the model management and collaboration layer.

"Data scientists need to interface with business stakeholders in order to deliver relevant results," said Steven Hillion, co-founder and chief product officer. "Connect provides a way for them to collaborate on analytics projects inside of the platform in a Facebook-like manner."

The final part of the solution is Alpine Touchpoints, which provides an interface for business users to create predictive analytics applications that serve up the intelligence from underlying models in a business context.


The MicroStrategy platform offers three interfaces for the development and consumption of predictive analytics: Desktop (for offline analysis on a PC and Mac), Web (across any browser) and Mobile (with native apps for iOS and Android). The products offer a library of over 350 native statistical and predictive functions, and they can integrate with other open source and third-party advanced analytical models.

With native support for regression and time series analysis, logistic regression and decision trees, clustering and association, MicroStrategy users can build and deploy predictive analytics applications that range from forecasting and predictive classification, to segmentation and market basket analysis.  It supports PMML (predictive model markup language), as well as function plug-ins written in C++ and the open source statistical R language.

"Users can add R analytics that execute within the MicroStrategy engine just like our native analytics, without requiring the overhead of additional R Servers or unnecessary data movements," said Stefan Schmitz, vice president of Product Management at MicroStrategy.


According to Forrester Research, RapidMiner offers solid enterprise predictive analytics software that also includes cloud capabilities. It can deal with hundreds of data loading, data transformation, data modeling and data visualization methods and has access to data sources such as Excel, Access, Oracle, IBM DB2, Microsoft SQL, Netezza, Teradata, MySQL, Postgres, SPSS and Salesforce.com.

Integrating specialized predictive analytics algorithms into RapidMiner is said to be simple, and it can run on every major platform, operating system and device.


FICO is best known in the credit report business. That alone gives it the credentials to succeed in predictive analytics. After all, how difficult must it be to score every person in the U.S. on whether they will pay their bills?

"FICO has incredibly deep knowledge about what it takes to make predictive models actionable," said Mike Gualtieri, an analyst at Forrester Research. "This is apparent in their solution that is geared toward data scientists who are continuously building and deploying models."


Adding predictive analytics can be challenging for those without the requisite skill sets. Angoss focuses on bridging that gap, with an emphasis on support services as well as an easy-to-use interface for predictive model creation. It uses what it calls Strategy Trees to combine customer segments, scores, business rules and calculations to formulate business strategies that minimize loss and maximize profits. As well as forecasting, its predictive analytics software offers fast creation of multiple what-if scenarios.


The goal of Alteryx is to help people quickly get their feet wet in the field of predictive analytics. It does this by concentrating on data preparation, as well as offering an analytical apps gallery that lets users share their data preparation actions and their modeling workflows with other users. This is effectively an exchange and marketplace for pre-built predictive analytics applications.

The aim is to reduce the effort required for analysts in business groups such as marketing, sales and finance to prepare, blend and analyze data. Alteryx Designer delivers a repeatable workflow for self-service data analytics.

Azure Machine Learning

Microsoft acquired Revolution Analytics, a predictive analytics startup, last year. Forrester Research views a Microsoft Azure offering that integrates the Revolution technology as an impressive new predictive analytics entrant with a lot of potential. While it is far from a mature offering, the analyst firm believes Microsoft could well become a major force in predictive analytics software within a short time. Delivered exclusively as a cloud service, the Azure platform is gearing up for sophisticated model building and analysis.

"The Azure Marketplace offers a distinctive single source for data and analytics services built with Azure Machine Learning," said Gualtieri.


Wise.io has developed a machine learning engine that learns from past patterns to predict future behavior. Its customer service applications, for example, help business users regain control over the volume of customer inquiries. Applications such as Wise Response and Wise Routing are built on this engine, which then offers a predictive analytics model to automate various stages of customer support and improve the customer experience.

The self-learning predictive analytics software integrates with common CRM and customer support systems such as Salesforce Service Cloud and Zendesk to automatically detect and mimic the decisions and actions -- from routing, to responses, to resolutions -- that employees take to serve their customers.

"Traditionally, customer support organizations were forced to scale by adding people or by attempting to write static, complex and often ineffective business rules," said Jeff Erhardt, the company's CEO. "Wise.io machine learning applications analyze past support tickets and cases, intelligently inferring the criteria by which an organization's best employees make decisions."

The system then generates a predictive analytics model that can be applied to incoming customer inquiries to facilitate human decision making and automate key processes, continuously improving and adapting over time.

Open Source Predictive Analytics Software

Another option is open source predictive analytics tools, five of which Enterprise Apps Today featured in this article.

Drew Robb is a freelance writer specializing in technology and engineering. Currently living in Florida, he is originally from Scotland, where he received a degree in geology and geography from the University of Strathclyde. He is the author of Server Disk Management in a Windows Environment (CRC Press).

  This article was originally published on Tuesday Feb 16th 2016
Mobile Site | Full Site