deFacto Global Blog

Effective Business Analytics In or Out of the Cloud

Business analytics and the cloud typically are seen as two naturally compatible offerings—and for good reason. Business analytics runs on massive and inexpensive computing that is optimized to process and make sense of huge volumes of unstructured data, and is accessible to a large number of users.

However, not all users of business analytics are enamored with the cloud. Security remains a major concern, particularly among larger companies that are highly regulated or publicly traded. While these companies may be willing to entrust their less sensitive marketing or business operations data to the cloud, they are reluctant to put their sensitive data in the hands of external cloud services, particularly their core financial information.

To reduce risk, these businesses restrict their most critically sensitive data to private clouds and other on-premises solutions.  Fortunately, new on-premise options are available that give these companies many of the benefits of cloud-based machine-learning predictive-forecasting solutions.

The Last Mile of Business Analytics

Before a company can develop a meaningful business analytics strategy, it is important to understand what business analytics systems can and cannot do well. Business analytics systems are designed to manage and add structure to unstructured data, with social media content being a good example. The analytics systems are not designed to automate or manage process flows around specific business processes.

While marketing or operations managers might use business analytics for predictive analytics, they still depend on CRM, ERP, and budgeting and planning systems to address the “last mile” problem, which is the ability to request, process, and deliver forecasts to business users in the timeframe and business context they need to make timely decisions. Business managers require dashboards with user-friendly interfaces with visual displays and controls that make it easy to grasp and manage the data.

ISVs Add Value

The “last mile” problem is best left in the hands of vendors that are able to seamlessly integrate predictive analytics capabilities into their core offerings. Microsoft, on its side, has begun to integrate machine learning (ML) capabilities into its CRM and Dynamics AX platforms.

Microsoft traditionally has relied on third-party ISV partners to extend its solutions, and ISVs can add value to Azure Machine Learning while enriching their own offerings with more intelligent analytics. deFacto Global has solved the last-mile problem for customers by introducing the first Predictive Financial Forecasting solution based on Microsoft Azure Machine Learning.

For those not familiar with Microsoft Azure Machine Learning, it is cloud-based solution that provides the resources needed to design, build, and operate predictive analytics models in a fraction of the time and cost previously required.

In partnership with Microsoft, deFacto is making available a solution that enables business managers to more easily deploy predictive analytics via Azure Machine Learning. Through integration with the deFacto platform, customers can create business models that enable them to perform predictive forecasting in concert with a broad range of business planning analytics, all based on a centrally shared model of the business.

deFacto Predictive Forecasting eliminates the need for data scientists and expensive data processing capacity and adds the critically needed user-friendly interface that makes it possible for managers in every area of the business to use predictive analytics. This solves the “last mile” problem and enables a broader range of companies to gain the benefits of predictive analytics.

Easier and Faster Deployment

In the deFacto Predictive Forecasting solution, the combination of the deFacto Business Planning Platform with Microsoft’s Azure Machine Learning significantly reduces the complexity and shortens the time required to get predictive analytics up and running. The ease and quickness of deployment are what make Azure Machine Learning and deFacto Predictive Forecasting an important game changer.

Before deFacto Predictive Forecasting, it took months or even years for data scientists and IT teams to create business models and build predictive applications. With deFacto Predictive Forecasting, the entire process is simplified, enabling businesses to build predictive applications in a fraction of the time.

Businesses of all sizes and stripes can now enjoy the benefits of predictive analytics that were previously accessible only to the largest enterprises. Businesses can use Predictive Forecasting to completely automate their forecasting processes to benchmark manual forecasts or completely replace manual forecasts with superior results. Improvements can be made in every strategic and operational area, including sales, marketing, financial planning, logistics, manufacturing, and human resources.

In the Cloud or On Premises

As noted above, many organizations are reluctant to put their sensitive data in the cloud because of security concerns. With the introduction of R-scripting support in SQL 2016 and Azure Machine Learning, Microsoft customers now have two great alternative ways for business users to tap into business analytics capabilities, either on-premise or in the cloud.

With R-scripting, more risk-averse companies can get started with SQL Server 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.

To demonstrate Microsoft R Server’s capability, a team of Microsoft data scientists recently built a model based on a huge dataset to predict whether a New York taxi passenger would pay a tip. Using R Server on a single Windows machine, the team collected and analyzed 600 million taxi records spanning four years. The team was able to perform the end-to-end process of downloading and cleaning four years of data, as well building and evaluating a machine learning model, in just less than 12 hours.

Have It Your Way

With deFacto Predictive Forecasting, customers can take advantage of the capabilities of Azure Machine Learning in the cloud or perform predictive analytics using SQL Server 2016 on internal servers on-premises. Both approaches enable users to deploy predictive financial forecasting to achieve similar results quickly and easily.

Having an on-premises option gives CFOs, finance analysts, and business unit managers the ability to perform budgeting, analysis, and forecasting using predictive analytics just as easily as they have been doing with their traditional methods, enabling all of these business planners to realize increased accuracy and significant operational advances.

Developing a meaningful business analytics strategy is a challenge for many organizations, especially those that are concerned about security. Thankfully there is now a range of solutions that can help, either on-premise or in the cloud.

Show More

Leave a Reply