Lessons Learned: How to Effectively Design an Online Product Name Test

Choosing a name for a new product or line extension is a critical decision influencing a product's success. Online research can help clients make an educated decision about what to name their product in order to increase its chances for success in the marketplace.

Getting Started

Clients need to provide the following:

  • Concept image (preferably without name on the concept)
  • Names to be tested – Testing five to ten names will yield the best results. A longer list of names may be possible with a very short questionnaire

Online Data Collection

The product name test is conducted via the World Wide Web, with online panelists being sent an e-mail directing them to the survey site (where the questionnaire resides). Panel members input a security code to access the survey.

Survey Design

An online product name test begins much like an online concept test. Respondents are screened for category usage, and then shown a concept image and asked a purchase intent question. Respondents who choose one of the top three boxes (Definitely will buy, Probably would buy, or Might or might not buy) proceed to the next question; those choosing one of the bottom two boxes (Probably would not buy or Definitely would not buy) are terminated. Only respondents who show an interest in the concept are allowed to continue with the survey.

Once positive purchase interest is established, respondents are told that the manufacturer is considering several possible names and we would like their opinion of the names under consideration. Respondents are then asked to rate each name on overall appeal and appropriateness. The order in which test names are shown is randomized to avoid order bias. These questions can be asked in a grid or sequential monadic format. Although either design can be effective, the grid format is preferable as it gives better differentiation between the names (see below).

Example of Grid Format

Listed below are different possible names for this product. Thinking about the product description you read, how appealingis each name to you? (Select One Box Across for Each Name)

The same kind of grid can be used to rate appropriateness of the alternative names.

Next, respondents are asked to rank the potential names. Attribute ratings and reasons for favorite and least favorite choice may be asked if survey length allows.

Analysis

While many product name tests are conducted as "field and tab" studies, several analytic techniques can be used to add valuable insights into the data. The following analytic techniques can prove to be especially useful:

  • Bradley-Terry Analysis: This analysis can provide additional insight into the ranked name preferences. Instead of making decisions based on a straight ranking, where the "distance" between rankings is unknown, the Bradley-Terry analysis allows us to determine how much one name is preferred over another (e.g. name A is twice as preferred as name I).

    Click for larger version of image
  • Perceptual Map: The names and dimensions (appeal, appropriateness, and overall rating) can be shown in a map for a more visual approach to the analysis.