Why Driving Trial Matters So Much in Volume Forecasting

Why Driving Trial Matters So Much in Volume Forecasting

If you work in the CPG world long enough, you will be exposed to volume forecasting.  Volume forecasting is a kind of quantitative analysis that predicts new product sales in years 1-2 of launch.  Unless your company does volume forecasting in-house, you are likely familiar with BASES.  BASES is a volume forecasting supplier owned by Nielsen and considered Best-in-Class in the field.  Even if you are not responsible for owning the volume forecast yourself, it is still important to know what determines its outcome.

The Fourt-Woodlock Equation of Volume Forecasting:

Since 1960, the Fourt-Woodlock equation has led our understanding of how product volume builds.  Depending on where you learn volume forecasting, you might call the individual variables different things.  Whatever you call them, the variables of the Fourt-Woodlock equation breakdown product volume into two categories:  Trial Volume and Repeat Volume.

Trial Volume:  (Households) x (% Households Trying) x (Volume at Purchase)

  • Households:  The total number of households in the country.
  • % Household Trying:  The estimated number of households that will buy the product once during the year.  You may also know this metric as % household penetration.
  • Volume at Purchase:  How much of the product a consumer is expected to buy the first time they purchase.

The trial volume is added to the repeat volume for the brand, item, or product.  Repeat volume is calculated as follows:

Repeat Volume:  (Trial Households) x (% Repeat) x (Volume per Repeat Purchase) x (Repeats per Repeater)

  • Trial Households:  From the trial equation, the number of households that purchased the product once in the year.
  • % Repeat:  The percentage of households that buy the product that will buy the product at least one more time during the remainder of the year.
  • Volume per Repeat Purchase:  How much of the product a consumer is expected to buy on average each repeat purchase.
  • Repeats per Repeater:  The average number of times a trial household will purchase the product again over the course of the year.

Therefore,  the full equation looks as follows:

[(Households) x (% Households Trying) x (Volume at Purchase)] + [(Trial Households) x (% Repeat) x (Volume per Repeat Purchase) x (Repeats per Repeater)] = Year 1 Volume.

Why Trial is the Center of Volume Forecasting:

Accurately estimating trial is the most important part of the equation.  If no one tries the product, there is no volume at purchase, and there is no repeat.  Yet, it’s amazing how quickly our business leaders forget the importance of trial.  Loyalty programs and trading-up consumers are popular strategies, but they totally ignore the fundamentals of volume forecasting.

From just the math alone, we see how important trial is to the Fourt-Woodlock equation of volume forecasting.  Trial is the ONLY variable that appears on both sides of the equation.  Further, it is the most limiting variable on both sides of the equation.  Trial estimates immediately cut the universe of potential buyers down from everyone in the country to something smaller than that.  Then trial sets the limit for the number of households that can possibly repeat.

This, is a simplistic way to think about why Dr. Byron Sharp and the Ehrenberg-Bass Institute of Marketing Science is so adamant that brands continue to recruit new users.  Penetration is the first step to loyalty, and consumers are fickle.  If a brand does not continuously try to expand its trial footprint, natural buyer churn will erode away the brand’s volume and share.


In conclusion, volume forecasting is often seen as an esoteric black box in market research.  Don’t fall into this trap.  As a market researcher you need to at least be conversant on the variables that make-up the Fourt-Woodlock equation, if not have the equation itself memorized.  It will make you a better strategist and steward of your business.

5 thoughts on “Why Driving Trial Matters So Much in Volume Forecasting”

Leave a Reply

Your email address will not be published. Required fields are marked *