New Product Development Blog

Is That New Product a Cannibal?

Companies that expect to survive must introduce new or improved products regularly.

The reasons for this are numerous. Consumer attention is fragmented and new products create or increase awareness. Private labels and other followers can launch similar products that eat into profits. Retailers demand the latest and greatest offerings to create “new news.” With these pressures from purchasers, competition, and distribution channels, companies are faced with the task of rapidly introducing new products, sometimes at the expense of current ones.

Certainly products that are today’s good performers may not be tomorrow’s stars. Product wearout happens. Over time, purchasers grow tired of eating, drinking, and using the same things. If a company fails to refresh its offerings, purchasers may look elsewhere. Even so, companies can be reluctant to launch new products that may steal share from current ones. On the flip side, companies may be overly eager to put something new on the physical and metaphorical shelf, eschewing thorough assessment of potential consequences. Both situations can cause companies to experience uncomfortable, and largely unnecessary, uncertainty.

The best way to counter new-product launch anxiety is having solid information in the form of source-of-volume estimates and incremental sales potential. Smart and sophisticated companies use multiple sources of information to triangulate on probable market impact after a new product has launched.

Dig Deep Into Product Launch History

How well did previous new product launches go? What was the cannibalization rate of current products? How quickly did it happen? Were there differences by distribution channel? If you are fortunate enough to have historical competitive sales data, dive into that to see the extent to which new products stole share from the competition. Notice the commonalities, if any, across all of your company’s past new product launches. Begin to develop hypotheses to test in subsequent primary research.

It’s true that past performance is not necessarily an indicator of the future. Yet, sales patterns can serve as a springboard for more future-oriented intelligence-gathering.

Ask Product Purchasers

Grounded in past company performance, primary research can be the crystal ball for how much of your current sales (and the competition’s!) will be sourced by the new product. The hope is that the new product adds to company sales even though some cannibalization might occur.

Direct Questioning. Ask a single question of respondents who are likely to purchase the new item. If this new product were available where you normally shop, would you buy it (1) instead of your usual product or (2) in addition to your usual product? The answer to this question can be interpreted in light of current product-purchase rates to yield a basic understanding of the likelihood that the new product will cannibalize particular brands. This is the simplest and also the least reliable method to assess source of volume. Use this method with caution.

Pre-Post Allocation. Respondents are asked to allocate 100 points, for example, to indicate preference for products currently available, then they evaluate the new product concept on likelihood to purchase, among other measures. The last step is to perform a second 100-point allocation that includes the new product. Adjustments to the new product share can be made using the concept-evaluation data (to adjust for respondent overstatement). Whether analyzed in aggregate or on the individual-level, this technique allows for a share calculation pre- and post-new product launch. Source of volume can be calculated and interpreted.

The key to this exercise is to include a good sampling of current brands in the category, including your own. This way both source of volume (share stolen from competitors) and cannibalized volume (share taken from your existing products) can be assessed.

Choice Modeling. Choice modeling (product and pricing optimization) provides a window into purchaser decision-making. Using a realistic respondent exercise and advanced mathematics, choice models quantify the impact of introducing a new product on market outcomes (share, units, revenue, margin). With proper calibration to the current market reality, the model can simulate the number of units that new products will steal from competitive products as well as units cannibalized from existing products. If pricing and new product features are varied in the choice exercise, these marketing levers can be examined in the simulator to determine if some combination of new features and pricing leads to more incremental sales (and less cannibalization of existing products). Based on the comprehensive intelligence yielded, choice modeling is often regarded as the best approach within the primary research toolkit to quantify source of volume.

Place the Product in a Test Market

This is the gold standard for assessing new product performance. Roll out the new product to a select number of stores and/or channels and observe sales for a few months. How quickly did the new product cannibalize existing product sales? Does this match with primary research findings? Why or why not? Test markets are the most realistic way to measure source of volume and sales lift. Unfortunately, this method takes time and money. For these reasons, few companies go this route unless the financial risk is quite high.

New product introductions are necessary to fuel company growth. Companies can experience new product launch anxiety, born of either the hesitation to hurt current sales or the “fail-fast” approach that advocates launching new products quickly and hoping for a good outcome. The best treatment for new product launch anxiety is to assess potential performance using historical data analysis, primary research, and/or test markets. Once any cannibalistic tendencies are revealed, companies can make better decisions regarding a new product’s fate.

Elizabeth Horn

Senior VP, Advanced Analytics

Beth has provided expertise and high-end analytics for Decision Analyst for over 25 years. She is responsible for design, analyses, and insights derived from discrete choice models; MaxDiff analysis; volumetric forecasting; predictive modeling; GIS analysis; and market segmentation. She regularly consults with clients regarding best practices in research methodology. Beth earned a Ph.D. and a Master of Science in Experimental Psychology with emphasis on psychological principles, research methods, and statistics from Texas Christian University in Fort Worth, TX.

• 
•