deFacto Global Blog

The Machine Learning Revolution Has Begun, Bringing Predictive Forecasting to Businesses of All Sizes

A revolution that has been quietly brewing in the sphere of machine learning and predictive analytics is taking flight. For customers, the analytics revolution is putting powerful new analytical tools in the hands of financial planners and business managers in every area of the business, enabling more comprehensive, effective, and reliable planning.

While the value of machine learning has been known for some time, the advances needed to make the practice accessible to a mainstream audience have been made only recently. As KDnuggets reports,[1] the convergence of three key trends is breaking down the barriers that have impeded the growth and employment of machine learning:[2]

1) Data “Flywheels”—This is where Big Data meets Moore’s Law, a confluence of ever increasing amounts of data and ever decreasing costs of processing data. The abundance of data enables more features and better machine learning models to be created.

2) The Algorithm Economy—Online algorithm marketplaces enable researchers, engineers, and organizations to create, share, and remix algorithmic intelligence at scale.

3) Cloud-Hosted Intelligence—As machine learning moves to the cloud, scalable web services are obtainable easily and at low cost. Data scientists no longer are needed to manage the machine learning infrastructure or implement custom code. Cloud-based systems can scale, generating new models on the fly and delivering faster, more accurate results.

Easier and More Affordable

A big takeaway from the recent Artificial Intelligence and Machine Learning Summit, as KDnuggets reports, is that cloud-based services have made machine learning systems easier to deploy and more affordable.[3] “With hosted machine learning models,” says KDnuggets, “companies can now quickly analyze large, complex data, and deliver faster, more accurate insights without the high cost of deploying and maintaining machine learning systems.”

A major factor driving down the complexity and cost of machine learning has been the introduction of cloud-based machine learning services by the major cloud providers over the past 12 to 18 months—IBM Watson Analytics, Microsoft Azure Machine Learning, Amazon Machine Learning, and Google Cloud Machine Learning .

The integration of machine learning into platforms and applications such as Microsoft Dynamics CRM by vendors and ISVs will further the trend towards making predictive analytics easier, more affordable, and an integral part of business processes.

Smart Machine Age

Experts see the infusion of machine learning into systems and applications as a major disruption that will have a revolutionary effect across all industries and areas of information processing. Gartner analysts define this new era of intelligent applications as the “Smart Machine Age.”  The Big Bang that is creating the Smart Machine Age, says Gartner, is a combination of radical new hardware, massive amounts of data, and unprecedented advances in deep neural networks. [4]

As Gartner relates, “The journey through the smart-machine age will be as transformative (and disruptive) as travelling through the industrial revolution.”[5]  By 2018, says Gartner, more than one-half of large organizations globally will compete using advanced analytics and proprietary algorithms, causing the disruption of entire industries.[6]

By 2020, Gartner analysts predict that, through the use of intelligent business analytics, information will be used to reinvent, digitalize, or eliminate 80% of business processes and products from a decade earlier.[7] Algorithm marketplaces, says Gartner, will disrupt the analytics ecosystem and likely even the whole software ecosystem.[8]

Algorithms Rule

That we are on the cusp of a major smart machine inflexion point was seen by the experts gathered at the recent Artificial Intelligence and Machine Learning Summit in Seattle.[9] Soma Somasegar, a venture partner at Madrona Venture Group, said machine learning intelligence was being built into the entire next generation of applications. “Every successful new application built today will be an intelligent application,” he said, adding that “intelligent building blocks and learning services will be the brains behind apps.”

Similarly, Joseph Sirosh, who heads Microsoft’s Data Group and Azure Machine Learning, said he believes that “everything at scale in this world is going to be managed by algorithms and data.” In the near-future, he said, “every business is an algorithmic business.”

Gartner also see algorithms playing the major role in defining the next generation of applications and business processes. “Algorithms,” says Gartner, “are moving front and center in the race for competitive differentiation as leading organizations uncover their true value.”[10] To remain competitive, Gartner advises business and IT leaders to step up to the growing business opportunities enabled by algorithms.

Machine Learning Explosion

Studies show that interest in predictive analytics is rising at a rapid pace and that an increasing number of businesses are reaping the benefits of machine learning technology. As relates, an Accenture study found that:

  • At least 40% of companies surveyed said they were using machine learning to improve sales and marketing performance.
  • 38% of those surveyed credited machine learning for improvements in sales performance metrics.
  • 76% said they were targeting higher sales growth with machine learning.
  • Several European banks were able to increase new product sales by 10% while reducing churn by 20%.

A recent McKinsey study found that a dozen European banks are replacing statistical modeling techniques with machine learning. The banks are seeing an increase in customer satisfaction scores and customer lifetime value as well.

Machine Learning Parts and Marts

Along with machine learning infrastructure and templates, vendors are making available libraries of algorithms that can be downloaded and run.  Businesses can access machine learning applets specific to the types of analytics they want to perform, such as financial planning, sales forecasting, marketing, human resources, credit scoring, manufacturing, supply chain, and more. Custom algorithms and models can be created and shared.  Microsoft is making available large datasets for specific subjects as part of its Azure Machine Learning service.

As KDnuggets reports, in the algorithm economy state-of-the-art research is turned into functional running code and made available for others to use.  As companies adopt the micro-service development paradigm, the ability to plug and play different machine learning models and services to deliver specific functionality will become progressively more mature.

As Gartner research director Alexander Linden explains, machine learning algorithm marketplaces are similar to the mobile app stores that created the app economy.  “The essence of the app economy,” says Linden, “is to allow all kinds of individuals to distribute and sell software globally without the need to pitch their idea to investors or set up their own sales, marketing and distribution channels.”[11]

Better Forecasting

Machine learning makes automatic adjustments to refine and improve analyses.  As Information explains, every time a miscalculation is made, machine learning algorithms correct the error and begin another iteration of the data analysis. The calculations are performed in milliseconds, which makes machine learning such a powerful tool for predicting outcomes and optimizing decisions.

As Gartner reports, the explosion of data sources and the complexity of information make manual classification and analysis unfeasible and uneconomic for many types of data sets. Machine learning, Gartner explains, automates these tasks and makes it possible to address key challenges related to “the information-of-everything trend.”[12]

Machine learning can eliminate the need to manually conduct predictive analyses for business planning purposes. Businesses can compare manual forecasts vs. machine learning results to gauge the accuracy of the analyses, using machine learning as a reliability check to benchmark against manual results. In instances in which predictive analytics outperforms manual processes, companies can eliminate the manual processes and automate their forecasting to achieve consistently superior results.

The ability to continually learn and improve enables machine learning to be applied across a broad spectrum of business areas, including contract management, customer service, finance, legal, sales, quote-to-cash, quality, pricing and production applications.

New On-Premise Options

While the new crop of machine learning services are cloud-based, businesses that are averse to putting their sensitive information in the cloud have new options that enable them to reap the benefits of machine learning via on-premise systems based on Microsoft SQL Server 2016 and R scripting.

With R-scripting, cloud-averse companies can get started with SQL 2016 to perform predictive forecasting, then seamlessly graduate to the Azure Machine Learning platform based on their need for higher performance or added capabilities to support more sophisticated applications, if and when they feel comfortable.

deFacto Leads the Way in Predictive Forecasting

deFacto has been at the forefront of simplifying business planning and forecasting and making  advanced business planning systems accessible to businesses of all sizes. In partnership with Microsoft, deFacto has introduced the Predictive Forecasting platform that fully integrates Microsoft Azure Machine Learning with the deFacto Business Planning Platform.  As deFacto CEO Bob Bedard notes, this is a groundbreaking advance that solves the “last mile” problem and enables businesses to have a machine learning solution up and running in a matter of days.

Like the algorithm marts described above, deFacto is making available loadable model applications that customers can access and share from an online Model Exchange. The models run on the deFacto Business Modeler Platform to accelerate the deployment of planning solutions in virtually any area of finance or business operations.

With the Smart Machine Age upon us, it’s time for all businesses to explore opportunities around machine learning and predictive forecasting. Organizations that fail to adopt advanced business analytics risk being left behind. To stay competitive, says Gartner analysts, organizations must create five to 10 viable business scenarios over the next six to 12 months using advanced machine learning.[13]

[1], “Machine Learning Trends and the Future of Artificial Intelligence” by Matt Kiser, June 2016,

[2], “Machine Learning Trends and the Future of Artificial Intelligence” by Matt Kiser, June 2016,


[4] Gartner, “Smart Machines See Major Breakthroughs After Decades of Failure,” Tom Austin, September 8, 2015.

[5] Gartner, “Entering the Smart-Machine Age, Tom Austin Bettina Tratz-Ryan, Frances Karamouzis, Whit Andrews, Alexander Linden, October 21, 2015.

[6] Gartner, “Predicts 2016: Advanced Analytics Are at the Beating Heart of Algorithmic Business,” Alan D. DuncanAlexander Linden, Lisa Kart, Nick Heudecker, Jim Hare,November 6, 2015.

[7] Gartner, “Use Analytic Business Processes to Drive Business Performance,” Neil Chandler, Thomas W. Oestreich, February 27, 2015.

[8] Gartner, “Algorithm Marketplaces Are Bringing the App Economy to Analytics,” Alexander Linden, October 15, 2015.

[9], “Machine Learning Trends and the Future of Artificial Intelligence” by Matt Kiser, June 2016,

[10] Gartner, “Explore Algorithmic Business to Drive Differentiation,” Stephen Prentice,March 9, 2016.

[11] Gartner, “Algorithm Marketplaces Are Bringing the App Economy to Analytics,” Alexander Linden, October 15, 2015.

[12] Gartner, “Top 10 Strategic Technology Trends for 2016: Advanced Machine Learning,” Mike J Walker, David W. Cearley, Brian Burke, February 26, 2016.

[13] Ibid.

Show More

Leave a Reply