Infosurv Concept Exchange

The Infosurv Concept Exchange (iCE) is a prediction market created by Infosurv, a global market research firm founded by Jared Heyman in 1998. As of June 2010, iCE is the only prediction market specifically designed for market research purposes, namely predicting the commercial success of new products or advertisements. A provisional patent on the technology behind iCE was filed in November 2008 and a utility patent filed in November 2009. iCE clients include mostly consumer packaged goods companies such as the Wm. Wrigley Jr. Company and Nestlé. In 2010, iCE was named a finalist for the IIR EXPLOR Award recognizing the most innovative new research methodologies of the year.
iCE allows regular consumers to bet virtual dollars on which new product concept or advertising concept they think will be most successful in the real-world marketplace. iCE traders can earn up to $10 USD in actual currency for making good predictions. iCE is similar to other prediction markets such as the Iowa Electronic Markets and the Hollywood Stock Exchange in that attempts to make predictions about future outcomes that are more accurate than traditional research methods such as surveys or polls by aggregating the wisdom of crowds.
Sampling
A key differentiator between iCE and traditional market research techniques like surveys or focus groups is that the respondent technique used for iCE is non-targeted. A survey or focus group is always targeted at a particular type of consumer, often defined by some combination of demographics, psychographics and behavioral criteria. Market researchers must take a targeted approach when using these techniques because they are essentially asking a particular consumer, “What would you buy?” However, in an iCE market traders aren't asked “What would you buy?” but instead “What will sell the most?” The first question must always be directed at a targeted consumer, whereas the second question can be directed towards anyone in a good position to predict which product would sell the most.
Mechanism
iCE traders are granted $1000 virtual iCE dollars which they may use to buy "virtual shares" in the new products, packages, logos, or advertisements concepts that they think will be most successful.
Initially, shares in every concept being tested in a given market have the same starting price, which is equal to the number of concepts tested divided by $100. In other words, if 4 concepts are being tested, they each have a starting price of $25 iCE dollars. Each concept’s share price reflects the market’s current estimated probability of that concept being the winner. Their total must always equal $100 because there’s always a 100% chance that one of them will be the winner.
Once trading activity begins, the share prices begin to fluctuate based on the market’s estimated probability of that outcome, just as shares of a given NASDAQ stock will start to rise if everyone starts buying it. As the price of one iCE stock begins to rise, the system automatically adjusts the other concepts’ share prices downward so that they always total $100.
By observing how share prices move over time, iCE can make predictions about the real world success of each concept being tested.
Resolvability
iCE is unique amongst commercial prediction markets in that it's a second generation (G2) market. This means that iCE markets cannot be resolved or paid out based upon an observable real-world outcome. Since most of the concepts tested in iCE markets are never actually launched, iCE markets must be resolved differently from first generation (G1) markets. There is an ongoing debate in the prediction market community about whether G2 markets are accurate, although recent academic research has found G2 markets to be equally predictive as G1 markets. iCE management will not divulge the details of how their markets are resolved.

Main Menu

See Also

Main Menu