Visual Planning – Predictions

Power ON is presenting Azure Machine Learning Predictions in Power BI for Budgeting, Planning and Forecasting.

Azure Machine Learning and ‘R’ are powerful tools that can be applied to corporate data to create projections and estimates to support Budgeting and Forecast processes.

Visual Planning from Power ON is now launching the ability to upload Accounting data from mainstream ERP’s including SAP, Oracle, Dynamics, Sage, and others and run powerful ‘R’ scripts and Machine Learning(ML) analysis and automatically create a forecast model for revenue data with dimensional data based on historical information from the ERP. How?


Screen Shot 2017-04-25 at 3.42.23 PM

Machine Learning Forecast on ERP Data

In this example, Visual Planning – Predictions has loaded data from SAP, ran the predictive script and estimated sales for the next 12 months per customer, per product!

The first step is to load your data to Power BI with Power Query and schedule it with Power Update.

Then by using Visual Planning solution enabled for Power ON,  users can deploy the Predictive Custom Visual that connects to our Azure ML Web Service.

What is different with Visual Planning Predictions from other predictive analytics modeling is that it returns the data point from the R script to Power BI so you can save it in Power BI and consume it later with any Power BI Visual.

Machine Learning Script

Visual Planning – Predictions has sample scripts users can take advantage of to get started. Customers using Visual Planning may also insert their own scripts, using Power ON’s templates.

The ML Engine is connected to Visual Planning’s web service and can receive the result set from the ML data and save it in Power BI for consumption and visualization by other users and different scenarios.

All features of Azure ML Studio is supported, and it can consume data from any ERP or corporate solution including SalesForce, Oracle Data, Hyperion Planning, HFM.

This is the world first solution with ML Studio in Power BI. The only other comparable solution Is IBM Planning Analytics for Corporate Performance Management and integrated BI.

Screen Shot 2017-04-25 at 12.05.37 PM

Advanced Settings Panel

Power ON’s Visual Planning – Predictions custom visual has advanced settings for driving powerful predictions for sales and expenses in the budgeting model. Models can be Multiplicative, or Additive.

Users may also control the Forecast Length, the Trend Component Type, Seasonal Component Type and Confidence levels. After changing the values, users can re-run the ML script and observe different outcome models, and compare results and setting for optimal model selection.

The custom prediction visual is connected to Power ON’s web service that is enabling powerful analytics in a simplified yet controlled environment.

Paradigm shift

Legacy budgeting and forecasting processes started with copying usually 9 months of actuals and 3 months of budget into a preliminary budget version for next year. This was necessary because the budget process started in October, and companies needed a projection for the year. This was a big band-aid, and dysfunctional process.

Welcome to 2017! With Power ON Visual Planning – Predictions you load all historical information into Azure Machine Learning from Power BI. Step two is to import the custom visual. Step three is to set the parameters, and you are ready to review the machine generated,  Budget and Forecasts.

Our largest client states: “Visual Planning – Predictions has completely changed the way we produce and review our budgets. Power ON’s solution has made dysfunctional budgets into living metrics that our consumers think is fun to compare themselves against”.

You get the data to Power BI, Power ON does the rest. Reach out today for a demo.


Screen Shot 2017-04-25 at 3.44.45 PM