Evaluating Source of Volume and Cannibalization for New Packaged Goods Line Extension Ideas


Many introduced new packaged goods products are line extensions of existing brands. Line extensions use an existing brand name to introduce a new variant--often a new size, flavor, form, package, ingredient, or feature. Typically line extensions are introduced with a positioning, which explains how the line extension addresses a consumer need, touts functional or emotional benefits, provides `proof' that the benefits will be delivered, shows how the product is unique, and, finally, asks consumers to try. Strong ideas can be very successful at generating sales but these sales may be at the expense of the parent brand. If the objective is to grow the parent brand franchise or if the line extension has lower margins, then a line extension that sources heavily from the parent brand may be considered a failure even if its sales exceed objectives. Therefore, a proven method for understanding how much volume a line extension sources from a parent brand is critical.

For purposes of clarity, let's define source of volume as the percent of a line extension's volume that is sourced from a parent brand; and let's define cannibalization as the percent of a parent brand's volume that is lost to the line extension. Although source of volume is a useful diagnostic, it is cannibalization that is critical to understanding whether introducing the line extension will be profitable.

Source of Volume Two approaches are commonly used to estimate source of volume from a concept test. One approach asks consumers a direct question such as "if this new product was not available, what brand (or brands) would you buy instead?" The second approach asks consumers for their last ten purchases across a competitive set of brands, exposes the concept, and then asks consumers for their next ten purchases across the same competitive set, which now includes the line extension. The purchase allocation approach typically is somewhat more accurate than a direct question, but the direct question works better when the competitive set is more difficult to define. Plus, the direct question can provide a rough estimate of how much of the line extension's volume will be obtained outside of the category by examining the percent of consumers who say `none' instead of selecting a brand. As a best practice, both the direct question and purchase allocation approaches should be employed; then either approach or both may be used depending on the results.

Analysis Results are analyzed among consumers favorable to the line extension (i.e., those who express positive purchase intentions or allocation of at least one purchase to the line extension). A sufficient sample size (typically 300 or more) is required to provide enough consumers who both have positive intentions toward the line extension and are purchasers of the parent brand.

Survey results should be calibrated to the market (via retail scanner data), because often in surveys the advertised brands are overstated while the price and store brands are understated. Also, in categories with many brands, surveys responses may be inaccurate because consumers are confused or the task is more difficult. Calibration corrects these biases. Validations against models based on in-market household panel or retail scanner data show that calibration improves accuracy considerably.

The chart below provides an example of a source of volume analysis for a line extension. Three columns are shown: (1) the market share for the parent brand and each competitor, (2) the percent of the line extension's volume sourced from the parent brand and each competitor, and (3) the index of source of volume to market share (called the `fair share' index).


In this example, the line extension is expected to take 40% of its volume from the parent brand, 21% from Competitor 1, 23% from Competitor 2, and so on. The line extension will take more than `fair share' from the parent brand and Competitor 3, and less than `fair share' from Competitors 1, 4 and 5. And although the index indicates the highest sourcing from Competitor 3, the absolute volume sources from Competitor 3 are fairly small. Line extensions are expected to source more than `fair share' from the parent brand. The cannibalization analysis answers the question, "how much volume sourced from the parent brand is too much?"

Cannibalization Cannibalization is a more difficult measure to obtain than source of volume. Forecasts are required for the line extension and the parent brand in addition to the source of volume analysis. To obtain the forecasts several marketing questions need to be answered:

  • What was the marketing plan for the parent brand in the last 12 months? What is the expected marketing plan for the parent brand during the introductory year of the line extension? What is the marketing plan for the line extension?
  • Will the parent brand's spending be changing? Will any of the parent brand's spending be allocated to the line extension?
  • Is the parent brand growing or declining prior to the introduction of the line extension?
  • Will the line extension be advertised and, if so, will the advertising be dedicated?
  • Will the parent brand be introducing any new SKUs prior to the introduction of the line extension?

The second chart provides an example of a cannibalization analysis. For a complete analysis, costs would be evaluated to calculate profitability for the line extension, parent brand, and total franchise.


Using Source of Volume and Cannibalization Analyses When are source of volume and cannibalization analyses appropriate for concept testing? For some line extensions, the relevant competitive set may be unclear. For example, the parent brand may be extending outside of its current category or into a totally new category. If the relevant competitive set is unclear, or if the consumer must use the line extension to have a better understanding of when the line extension would be used or what products would be substituted, then it is best to defer source of volume and cannibalization analyses to research which is conducted at a later stage in the development process (e.g., a concept/product test).

Source of volume and cannibalization analysis are important tools for evaluating the success of line extensions. These tools must be applied appropriately to obtain business information that aids in decisions regarding whether or not line extensions should be introduced.

More insights about Public Sector