Oracle EPM Cloud for Trend-Based Modeling

Published October 9 2020 by Vatsal Gaonkar
Back to insights


Today’s modern economy - whether related to individuals, communities, organizations, or nations - has a complex network of competing agents for supply and demand. The belief in this economic model, in most set-ups, is that actions carried out by agents advocating economic outputs have a measurable pattern or trend in inputs that predict outputs. One of the economic models is the empirical model which, when defined in simple terms, says that the economic outputs or financial results can be modeled based on a set of variables or variable relationships that have been input. One relationship modeling technique in this empirical modeling paradigm is Trend-Based Modeling.

For immemorial planners from ancient sea traders to the corner mom-and-pop store in our neighborhoods, this technique has been used. Simply put, the model calculates future financial revenue and costs based on empirical data or past experiences with some assumptions made for extraneous factors such as economic slowdown, inventory depletion, etc.

Oracle EPM Planning Cloud has one such model built in the Financials Module in the form of a prescriptive out-of-the-box content. The feature is aptly called the Trend-Based Model. In our travels working with many clients, one of the early adopters of this prescriptive content was Vitamix Corporation. Founded in 1921, Vitamix manufactures high-performance blending solutions for home and commercial use. As a recognized leader in marketing and product innovation, Vitamix founder William. G. “Papa” Barnard created one of the first infomercials in 1949 to demonstrate how the Vitamix blender could help people improve their health with whole foods. The company developed the first true commercial blender in the early 1990s, which ignited the smoothie movement in the United States. With a rich culture of innovation and arduous research running through the company’s veins, the FP&A team at Vitamix was welcoming of newer technology options that could make their planning cycle more robust.

The solution needed to compile assumptions at a SKU (Stock Keeping Unit) level and summarize it to the planning product category level for users to make broad assumptions for product margin planning. What was more intrinsically needed was an ability to use trends in these “scaled-up” product category assumptions to extrapolate product category margins. Oracle EPM Cloud’s out of the box Trend based Model provided methodologies to use patterns in historical data to forecast numbers into future forecasting months. Some of the salient trend models in the Financials Module that were implemented at Vitamix were –

  1. Prior Year Actual Average – assumes historical actuals is loaded
  2. Prior Year Actual Average with Seasonality – assumes historical actuals is loaded
  3. Hold Forecast to Current Year Plan – assumes the system is used for monthly forecasting in addition to Budget
  4. Monthly Growth & Prior Month Run Rate – builds model based on provided factors (rates)

    As a first, the underlying data for these salient forecasting methodologies needed to be aggregated from a SKU level input to Product Category, where Vitamix FP&A plans.

For example, drivers such as Unit Volume, Unit Cost (Standard Price), and Unit Returns across approximately 20000 SKUs needed to be loaded and then summarized to approximately 240 product categories to plan against. Trend based models in the prescriptive content was applied at these product category levels to compute product category margins across entities and locations. The design needed a collection place for these drivers at a highly sparse dimension, which to begin with would house 20000 SKUs with room for growth in the future. The solution would load all these SKU level data to an aggregate storage option (ASO) plan type, aggregate it using Cloud Data Management capabilities and move over to the Financials Module plan type (OEP_FS) at the product category level.


Configuration for the trend-based data modeling is set-up to work expansively for monthly/quarterly forecasting, annual operating plan, and rolling forecast owing to the scalability of the framework. Most planning applications will need a combination of different models such as –

  • Trend Based models – based on actionable data that feeds the application, regardless of the granularity of data. In an upcoming blog, we will discuss Intelligent Performance Management (IPM or EPM Auto Predict) that is in EPM Cloud Planning since 20.08 (August 2020) update.
  • Driver Based models – differ by industry. As an example, Volume of visits/ procedures are an essential driver that drives variable revenues and costs in the forecasting model.
  • Zero Based models – These models could be mainly for non-variable lines of a forecast. Typically, year over year numbers are given a bit of an adjustment and re-entered. Today’s ever-changing world, COVID-19 pandemic notwithstanding, requires many industries such as manufacturing, health care, and financial services to get down to the actionable data level to derive their planning drivers and rates. The automation of getting these rates to flex and evolve based on a change in underlying actionable data is where the magic sauce is for FP&A planning.

Oracle EPM Cloud Planning provides a state-of-the-art prescriptive content for Trend-based modeling. With the availability of Machine learning algorithms in Oracle EPM Cloud Planning, the trend-based models can be further flexed to model key drivers, say Volumes and variable costs in healthcare, product margins based on SKU level detail change in Manufacturing & Interest Rates/ Borrowing Cost Rates in Banking. In our next blog, we will discuss the flexing of drivers using Oracle EPM Cloud Auto Predict. You can find more information on EPM Auto Predict here.

For comments, questions, or suggestions for future topics, please reach out to us at  Visit our blog regularly for new posts about Cloud updates and other Oracle Cloud Services such as Planning and Budgeting, Financial Consolidation, Account Reconciliation, and Enterprise Data Management.  Follow Alithya on social media for the latest information about EPM, ERP, and Analytics solutions to meet your business needs.

Contact us