This Initiative Isn’t Working Like We Thought: The Problem of Cannibalization

This Initiative Isn’t Working Like We Thought:  The Problem of Cannibalization

One of the mistakes businesses make is thinking about new initiative launches in a silo.  As a result, they sub-optimize the incremental growth the new product can deliver.  Even if the consumer insights manager is an expert on the “Three A’s” of trial, sometimes new products don’t perform as expected.  Further, there are cases where even big-volume initiatives don’t yield much growth to the business.  If the consumer insights manager or volume forecaster predicted trial and repeat correctly, business partners may be left scratching their heads as to why the business is not growing.  The answer is cannibalization.

What is Cannibalization?

Cannibalization is volume a new product sources from other items in a company’s portfolio.  Unless the new product creates a totally new-to-the-world category, there will be some cannibalization impact.  Typically, most of this volume comes from other products or variants within the new product’s own brand. Marketers and consumer insights managers are typically fine with the volume cannibalization comes from other brands’ products.  They may even be OK with cannibalization if it means that the new product trades shoppers up from their own lower-priced/margin items to higher-priced/margin items.

However, retailers are concerned with cannibalization within their overall category is well.  Therefore, shopper insights managers need to understand what drives cannibalization too.

What Drives Cannibalization?

If you understand the “Three A’s” of trial and the Fourt-Woodlock model of volume forecasting, then predicting cannibalization is easier.  New products typically cannibalize their brand in one of several ways:

  • Appeal:  There is an overlap in consumer appeal with other products in a company’s portfolio.  People who already bought the company’s brands now find this new product more appealing.  As a result, they buy the new product.  Unfortunately, these aren’t new buyers.  They would have bought the company’s existing products even if the new one hadn’t launched.  If this is the case, the company is just “shifting sand.”
  • Awareness:  Money is a finite resource.  If the new product’s launch takes media spending away from another product in the portfolio, that other product’s volume will decrease making the launch somewhat of a wash.
  • Accessibility:  This is an oft-overlooked cause of cannibalization.  Brick & mortar retailers do not (yet) have “endless aisles.”  If your company launches a new product, its space on the shelf must come from somewhere.  Ideally, it would come from weaker competitive items.  However, most retailers will expect companies to remove their own poor-performing items to make room for new product launches.
  • Volume per Purchase:  In some cases, new products launches down-size either the price or amount of product sold v. the rest of the portfolio.
  • Purchase Rate:  This can actually be a negative effect of products that promise “longer-lasting” benefits.  If these products truly delay repeat purchases, they can potentially elongate the purchase cycle.  This increase in price or volume is actually a net negative v. the faster-consumption items that required more purchases.

If this sounds familiar, it is.  These are the same variables as the Fourt-Woodlock formula.

How Can I Predict Cannibalization?

Four of the five factors above are self-inflicted cannibalization.  They are the result of business decisions that don’t pay-out.  Shifting marketing spend, deleting items at a retailer, launching a product that changes purchase size and frequency are all known risks.  These can all be modeled across multiple scenarios pre-launch.  The hard one to predict is cannibalization of appeal.

A common way to predict cannibalization of appeal is through SOVA.  SOVA stands for “Source of Volume Analysis.”  Typically, when a concept is tested, there is a question in the survey that asks something like, “If you went to the store and this product was not available, what would you buy instead?”  Respondents then claim what product they would buy instead (usually what they normally buy).  This claimed data is used to estimate where the new product’s volume will come from. Unfortunately, claimed data isn’t exactly perfect.  As a result, in-market behavior may differ drastically.

If you want to go the extra mile and look at sourcing within a competitive set, choice modeling, share of preference modeling, and RichMix analysis are all predictive methods that do this.

Conclusion:

In conclusion, the good news is that you can identify most cannibalization issues pre-launch. Most causes of cannibalization are actually the result of the company’s own decisions.  Now that you know what you are looking for, hopefully you can help ensure big initiatives deliver big incremental sales for your company.

3 thoughts on “This Initiative Isn’t Working Like We Thought: The Problem of Cannibalization”

Leave a Reply

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