You May Get More Than You Pay For
Want Better Product Ideas? Try Smart Incentives
By Rajan Sambandam, PhD, TRC's Chief Research Officer. Published in Quirk's, November, 2005.
How often have you done surveys where open-ended questions soliciting new product/feature ideas have produced poor response? If your answer is “innumerable times,” then it is time to consider smart incentives. Smart incentives are a way of motivating survey respondents to think about the questions asked and to generate the best ideas they can.
Consider what happens in a regular consumer survey. Respondents are qualified and asked to participate in a study. Often an incentive is used and while they vary in type (e.g., cash incentive, entry in a lottery, points redeemable for rewards), they all have one thing in common: The incentives are for participation in the survey, not for the quality of the responses given. The primary aim of the respondent is to complete the survey rather than provide the most thoughtful response. While this may be reasonable in some types of studies (say, measurement of customer satisfaction), it is not very fruitful when the objective is idea generation.
Innovation is the engine of growth. Yet idea generation is a difficult task for companies in all industries. Significant numbers of ideas are generated internally, but there is no denying the fact that consumers can and should participate in the process. This is where smart incentives come in. Smart incentives are a way of rewarding respondents for coming up with the best ideas and they work as follows:
Inform respondents to the survey that their ideas will be evaluated and that the best idea will receive a significant reward. Then simply allow the magic of free market competition to work as respondents compete with each other to produce the best idea.
That is all it takes. Since each respondent is providing ideas in isolation (as opposed to collaborating as in a focus group), the potential for a large and varied set of ideas to emerge is quite high. Research has shown that when a large number of reasonably well-informed people generate ideas independently, the outcome is much more positive. In other words, this method overcomes the problem of "group think." Apart from producing good ideas, the task itself is more interesting for the respondents because they are now competing to be the best.
What kind of results can we expect from using smart incentives? We ran an experiment using respondents to a Web panel in which two random groups of people were asked two simple questions. The first question was about checking accounts and asked respondents to suggest one new feature or modify an existing one. The second question was similar and related to Internet service providers (as this was a Web panel all respondents were familiar with ISPs). One group was offered a standard participation incentive, which in this case was entry in a lottery for a chance to win one of several $20 gift certificates. The second group was offered the same participation incentive and was also offered an additional incentive of $200 if their idea “won.” The winning idea, these respondents were told, was to be picked by another group from the same Web panel.
When the results came back it was quite clear that the smart-incentive group was responding quite differently as seen by several measures.
The number of non-responses was higher for the participation-incentive group for both questions. That is, the smart-incentive group was more motivated to answer the questions.
The smart incentive group was following instructions more accurately in suggesting only one idea per question (as asked).
When it came to the question of actual number of ideas, the smart-incentives group (across the two questions) produced 100 percent more responses than the participation-incentive group.
Looking at the ideas it was also clear that the smart-incentive group was trying harder and providing more “meat” around each idea. Frequently, they provided more elaborate descriptions of their ideas, sometimes running into paragraphs. On the other hand, the participation-incentive group’s answers tended to be single phrases or single sentences.
Checking account ideas from the participation-incentive group tended to be fee- or access-related. Examples include free checking and bill payment, ability to access account and check status online, and e-mail notification. Ideas from the smart-incentive group covered these topics, but also went beyond them and were more creative. A points/rewards system for maintaining minimum balances, linkage of bank card to grocery store card, online account balancing and various biometric recognition ideas were some that came from this group.
ISP ideas from the participation-incentive group were generally related to control (blocking pop-ups) and security (spam). Again the ideas for the smart-incentive group covered these topics and went beyond. Bundled services, phone service over Web, automatic link to best dial-up connection in area, and portable e-mail addresses were some ideas that came from this group.
Of course, not all ideas are created equal. The process described so far is primarily useful for identifying a variety of ideas. To truly understand if one or more of these ideas will prove to be popular, a second phase of research is necessary. In this phase, ideas that are deemed practical and innovative are tested to gauge market response.
In our experiment another wave of data collection was conducted with the Web panel. In this case, respondents were asked to rate the ideas generated in the first phase. The idea that got the highest rating was declared the winner. In each product category, the winning idea was suggested by a single respondent. This is interesting because in each case a solitary respondent was able to come up with an idea that ultimately proved to be the most popular. If smart incentives were not provided or if the second phase of the research were not conducted, it would not have been possible for these ideas to emerge as winners. Of course, the client is under no obligation to just go with the winner. It is entirely possible that the client may like other popular ideas that did not win because of feasibility or other reasons. But the point is that there are many more ideas to choose from and they have all been put to a popularity test that provides a good indication of what the market thinks of them.
Needless to say, the two respondents who won were very glad to receive the $200 cash prize!
What we see here is the power of a very nominal smart incentive to provide substantially better returns. There are different directions in which this research can be taken including additional stages for fine-tuning ideas, and addressing other objectives such as concept testing and enhancement. But the basic idea of motivating the respondent to be thoughtful remains the same. At a time when new products are failing at a high rate and companies are looking for any sort of differentiating edge, smart incentives can provide a very effective and economical boost to the idea generation and product modification processes.