An accurate forecast is an essential part of how CPG organizations plan and make confident business decisions. 

A volume forecast is a prediction of expected future sales volume based on the information available across the organization. In the CPG industry, volume is frequently referred to in cases (shippable units) but can also be measured in other common units relevant to the organization (such as eaches or pounds).  

Forecasts have wide-ranging use cases within an organization, first and foremost serving as the primary driver within the supply chain. Raw materials, production plans, inventory targets, and distribution plans are all driven by volume forecasts. Without accurate forecasts, organizations can struggle to align supply with demand while maintaining high levels of service to customers. Beyond that, volume forecasts also play an important role in financial planning, budgeting, and sales planning as key inputs to developing accurate plans and supporting strategic decision-making. Below, we unpack considerations for building accurate forecasts in CPG.

Components of a Forecast  

In leading organizations, forecasts aren’t created solely as a total volume number. Forecasts are commonly developed with base and incremental components building up to a total forecast. A true base volume represents expected sales in each period (e.g., week, month) without any promotional activities or other factors that would increase sales. In contrast, incremental volume represents sales driven by promotional, marketing-driven activities, or other non-recurring volume drivers. 

Within base and incremental volumes, there are subcomponents that drive each: 

Base Volume Drivers 

• Everyday Volumes: Amount sold given current distribution, shelf placements, and pricing with existing retailers 
• Custom Items Sold Daily: Similar to everyday volumes, but involves unique products only created and offered to a specific retailer (e.g., special case configuration, special unit size) 
• Seasonality: Regular, repeated fluctuations in volume differing periods (weeks, months, quarters) driven by changes in consumer behavior across different periods in a year (e.g., sunscreen sales increasing in the summer) 
• Velocity: Overall upward or downward trend in a forecast driven by changing consumer preferences over extended periods of time 
• Distribution Changes*: New ongoing distribution at existing retailers within current product portfolio, new retailers carrying the existing product portfolio, or changes in store counts that could increase everyday sales as well as losses in distribution or changes in store counts that would decrease sales 
• Innovation*: New product launches that will expand everyday ongoing volume at existing or new retailers  
• Corporate Price Changes: Increases or decreases to list prices that influence retailers to adjust shelf price and affect the volume consumers purchase (price increases drive volume reductions, while price decreases can increase volumes) 
• Shelf Position Changes: Shifts in the number of product facings or shelf location that influence the volume purchased by end consumers (e.g., adding facings or improving location on shelf can increase volume) 

*Note: Distribution changes and innovation can result in cannibalization of existing SKUs, eroding existing baselines 

Incremental Volume Drivers 

• Promotions: Investments with retailers to offer limited-time price discounts or deals to consumers that increase sales within the promotional period (e.g., “Buy 1 ,Get 1 Free” or “2 for $4.00” vs. regular retail price of $2.99) 
• Displays: Limited-time special product locations in a store (aisle endcap, shippers) that create temporary additional locations within store, creating additional opportunities for consumers to purchase products 
• Special Packs: Limited-time product offerings (e.g., product size increase from 12oz. to 16oz.) that create new opportunities for consumers to purchase, but will not remain as an everyday item 
• Seasonal Programs: Limited-time offerings of seasonal products (e.g., holiday themed product flavor or seasonal product display aisle in store) for specific periods of time that will not remain as everyday items or displays 
• Marketing Programs: Advertising (commercials, billboards), digital marketing (social media, eCommerce advertising), loyalty programs, and other marketing initiatives are additional levers organizations utilize that drive volume and can add incremental volume sales within periods of time that initiatives are running 

Total Forecast 

Base volume drivers and incremental volumes drivers are combined to become a total forecast representing the total expected sales for each period 

Information Sources Used in Developing a Forecast 

Forecasts are developed using many different information sources, including shipment history, point-of-sale (POS) data, promotion plans, customer forecasts, customer inventory levels, and macroeconomic data. Across CPG companies, these inputs are frequently analyzed in statistical models to generate forecasts and develop demand plans. 

While statistical modeling is important, it doesn’t replace the judgment from demand planners and sales teams. Without their qualitative insights, forecasts would be missing critical information that can play a significant role in forecasts.  

Looking at base volume drivers, there’s a mix of quantitative and qualitative inputs to the forecast. Everyday volumes, custom items sold daily, seasonality, and velocity are all quantitative forecast inputs. Quality data and consistent purchasing patterns enable statistical models to do an excellent job at forecasting these inputs to the forecast. However, distribution changes, innovation, corporate price changes, and shelf position changes are all qualitative inputs to the forecasting process that a statistical model wouldn’t know without a demand planner or sales team member providing inputs to the forecast. 

Incremental volume drivers also rely heavily on qualitative input. Promotions, displays, special packs, seasonal programs, and marketing programs all need to be added into the forecast, as statistical models don’t generally have forward-looking visibility into these planned activities.   

Looking Ahead

This is the first article in a series about forecasting, with a focus on both demand planning and sales forecasting. Throughout the series, we’ll dive deeper into each of those critical forecast types, how they’re created, and what demand planning and sales teams should consider as they build more accurate plans. We’ll also explore best practices for forecast development, planning cycles, key metrics, and the connection to the overall S&OP process.  

Our next article will focus on the differences between demand and sales forecasts, including how each team’s perspective shapes the planning process. Visit our consumer products supply chain planning for more on the topic.