Prediction market research is unique in a few ways. First, it’s game based and participants get to select the topics they want to partake in. Second, unlike survey research we don’t collect opinions. Instead, we ask for a judgement on a specific outcome. For example, “Thinking about consumers who eat yogurt, which 1 of these 10 product names would motivate the most consumers to try this new Yummy Brand yogurt product?”. Once participants provide their answer, they invest tokens to indicate their confidence in the outcome. Next, participants are asked to justify investments with a qualitative open-ended rationale for their decision. Finally, our participants have “skin in the game” with the top 20 percent of participants earning additional incentive based on their predictions.
The latter. Unlike other traditional approaches, a prediction market asks participants to predict the behavior of others based on their confidence in the outcome, rather than on what they would personally do. This questioning technique takes the individual out of the equation and allows participants to consider how their friends, family, and social network would react to each question.
Generally speaking, a participant does not need to be a user of the category discussed in the prediction market; however, they do indicate familiarity with the topic by choosing to participate in the question. In fact, one advantage to the self-selection design of prediction markets is we yield a more diverse or “gen pop” audience which typically delivers the best result in most categories. Also, when compared to traditional methodologies, prediction market research requires smaller sample sizes in order to make the same business decisions which saves time and money.
Yes, participants see current results as an intentional component of the methodology. When participants evaluate a question, and the possible outcomes, they will also consider the current likelihood displayed in the market. This is equivalent to a sports bettor understanding the current odds and betting accordingly.
Our Huunu prediction market research platform provides the following metrics:
Prediction Likelihood Index summarizes overall prediction score on key success metrics
Preference Prediction reveals the proportion of participants betting for and against each outcome or answer choice
Strength Meter calculates the “intensity” of positive and negative betting activity on each outcome or answer choice
Qualitative Guidance summarizes of open-ended consumer feedback