Driver-based planning brings data modeling to EPM software

Use driver-based planning tools to identify and model the variables that have the biggest impact on forecasts.

Sales example

Using Units and Rate is a simple way to model sales. Here's a more sophisticated example.

Pipeline. Each salesperson maintains a pipeline of what they hope to sell in the near term. Picture a matrix with salespeople down the side, months across the top and the pipeline amount for each month.

Forecast. Sales managers know which salespeople are realistic and which ones are overoptimistic, and can use drivers to reflect these qualifiers in their sales forecasts. For example, if Fred predicts $10,000 in sales but always estimates high, then his sales manager's driver -- almost a fudge factor -- might be 80%.

Attrition. A certain percentage of the sales force will quit or be fired during the year. One driver is the percentage of the existing sales force that will be lost in a given month. Another driver is the number of months, on average, it takes to replace a salesperson .A third driver is the number of months it takes a new salesperson to be productive.

Competition. The competition will make adjustments to the size of its sales force and pricing policy. This, too, will impact the driver-based plan.

Since all vendors of modern enterprise performance management (EPM) software seem to offer driver-based planning, now is a good time to get a firm grasp on definitions and consider some key questions.

What is driver-based planning?

The "planning" in driver-based planning is about building models that forecast where the company is headed based on information from the departments that must work together to be successful. Here are some questions that a plan might address, by department:

  • Sales. How much revenue can we produce?
  • Operations. How do we provide enough products to satisfy customer demand?
  • Human resources. How many employees do we need to hire? How long will it take for them to be productive?
  • Finance.  How do we fund all these activities and stay profitable?

"Driver-based" refers to the variables that drive key results. Rather than make an estimate for total sales, for example, you can use drivers to model the variables that "drive" sales, such as the total number of customers. Here are some examples of drivers:

  • Sales. Number of salespeople by region, number of customers and average demand by customer.
  • Operations. Number of factories, availability of raw materials and capacity.
  • Human resources. Number of employees, active recruiting programs and months to make a new employee productive.
  • Finance. Cost of goods sold (COGS), revenue by region, capital investments and margin.
(Credit: Alight Planning)

This illustration from Alight Planning, the most visible vendor in driver-based planning, shows drivers in action. In the illustration, Units and Rate are drivers. By modeling at the Units and Rate levels, you can get a more reasoned answer about software license revenue.

Managing dimensionality

Referring back to the diagram, in the leftmost column, three regions (North, South, West) are represented, with two or three products in each region. For each product, by region and by month, you will want to estimate units and rates.

Let's examine the "so what" of all this. Units and Rates are drivers, which help us model revenue. Product, Region and Month are dimensions, which are keys to the definition of Units and Rates, and hence how revenue is computed. With three products, three regions and 12 months, there are 108 potential rows of data.

Could all this be handled in Excel? Sure. But once you understand that the world of planning is highly dimensional, you should expect a software vendor to provide techniques to both represent this data and report on it. Most EPM software vendors do.

Production example

At the most basic level, COGS equals the number of products available times the cost per product. But there is more to consider. The success of any plan to bring a new product to market often depends on whether manufacturing can assure that the added capacity is available. Driver-based planning lets you use multidimensional analysis to factor in two key drivers of capacity:

Production. Due to any number of drivers, manufacturing may not be able to deliver the number of products requested by sales.

Raw materials. Raw materials may be late in arriving at the factory, thereby delaying production. The price of raw materials may go up, increasing COGS.

Exploring the model further

Once you have estimated all the Units and Rates, you can compute total revenue for the year. Having done the hard work to build the model, there are two main kinds of analysis you can do.

"What if" analysis lets you change a unit somewhere and recompute the revenue, while goal seeking allows you to decide what you want revenue to be, for example, and have the system adjust a Unit or Rate to meet that goal.

Driver-based planning can do all of that and has become an important component of EPM software.

My advice? Take the time to model a three-dimensional scenario, such as sales by region, by product and by month. Go further and model more complex sales and production scenarios like the ones described in the sidebars. EPM software vendors who can help develop these complex models are quite likely to be on your short list.

About the author:
Barry Wilderman has more than 30 years of experience as an industry analyst, researcher and consultant at such companies as META Group, Lawson Software, SalesOps Analytics and McKinsey and Company. He is currently president of Wilderman Associates. Contact him at [email protected] and on Twitter at @BarryWilderman.

Dig Deeper on Financial analytics and reporting