Each organization has its own approach to risk. Many have centralized the process to better  track risk probability and coordinate mitigation strategies. Others have invested large amounts of money in data analytics and business intelligence software. When it comes to actually involving company personnel, strategies rarely  extend much further than training programs or employee handbooks.

These are all serviceable strategies. Yet, arguably the most underutilized resource to help identify and mitigate risk is the one that will help achieve the best results: the collective wisdom of the organization. The invaluable experience and knowledge of company employees, vendors and other affiliates often remains largely untapped. Finding the pulse of the company and putting this resource to work is a critical component of managing risk.

Prediction markets are still relatively unknown as business tools. They first became popular after being used by the University of Iowa to more accurately predict presidential elections. The Pentagon has used them to predict terrorist attacks and casualty counts. In the entertainment world prediction markets are used to forecast movie ticket sales at the Hollywood Stock Exchange.

More recently, the publication of James Suroweicki's book The Wisdom of Crowds and the buzz around Web 2.0 popularized prediction markets as viable and effective tools for the business community. Suroweicki cited several examples of prediction market use in business, most notably forecasting printer sales at Hewlett-Packard, predicting FDA approval of new drugs at Eli-Lilly and anticipating project completion dates at Google. Using prediction markets in these ways has value, but just begins to scratch the surface of their usefulness.

For instance, imagine that we want to predict when a project milestone will be completed. Through a prediction market, a company could ask the question: "When will milestone X be completed?" Managment would then set up stocks representing possible answers: "Week 1," "Week 2," "Week 3" and "Week 4." Participants in the market, which often include employees, suppliers and other partners, then "buy and sell" shares in the possible answers. Some consumer prediction markets use real money allotments, while others give their traders-the participants-an allotted amount of fantasy money that they use to make stock transactions through a network interface reflecting what date they expect be correct.

The term "market" is used because people buy and sell shares in stocks representing the possible answers to questions. Whichever answer has the highest stock price has the highest probability of occurring according to the participants in the market. And more often than not, the market speaks the truth.

The ability of a diverse group of people (a "crowd") to generate a more accurate prediction than an individual is still not widely understood, but nevertheless, a small group of "like-minded" or "single-discipline" employees regularly do just that.

Because of this phenomenon, prediction markets are increasingly helping companies improve forecasts of key performance metrics, increase collaboration with customers, better allocate resources and foster a culture of innovation. Besides sales forecasts and project management, prediction markets are being used to better allocate financial resources by forecasting performance metrics, filtering new innovations the company may pursue, and performing  competitive analysis. Perhaps most interestingly, by bringing additional transparency to employees' feelings, prediction markets can also be a valuable part of any risk mitigation strategy.

Using Prediction Markets to Mitigate Risk
When companies are faced with missed revenue projections, tardy product launches or interruptions to a supply chain, upper management often thinks, "How did this happen?" or "Why didn't we know about this?" Risk managers can better forecast these mishaps by implementing an internal prediction market platform. The benefits of doing so are several:

1) Constant risk tracking. Traditionally, companies do not track specific risk factors or, when they do, it is on a quarterly basis. Alternatively, they deploy expensive IT infrastructure to collect large amounts of data about specific processes. With prediction markets, a company can have a constant gauge on risk probabilities. Quantification is available 24/7 and reacts to internal and external events in near real-time.

2) Transparency. Prediction markets provide companies with a level of transparency never before so readily accessible. Perhaps for the first time, management can begin to understand what employees really think across a broad range of issues because stakeholders are participating anonymously. Suddenly the company is able to circumvent the office politics that invariably hinder the free flow of information.

3) Collaboration. Working on a project together with all the relevant employees within an organization is always difficult-let alone trying to do so with external business partners, vendors, investors and customers. Prediction markets can be an effective way of easily and cost-effectively bringing together all of an organization's stakeholders in risk analysis.

4) Awareness. Typically, when management makes employees aware of a metric, they will do a better job meeting it. Risk is often an abstract concept for most employees, but when organizations ask specific questions about risk factors, people will start to think seriously about operations, strategy and even their own jobs in the context of risk aversion.

5) Identification. One of the reasons consumer sites like YouTube and Flickr are so popular is because the content is generated by users; there are no editors or gatekeepers. Prediction markets can be used in the same way. Companies can use them to augment their formal risk factor identification process where employees can identify possible risks and run their own markets to gauge the outcomes.

Getting Started
Prediction markets can be surprisingly straightforward to implement. The following are some suggestions to get started: 

Find a vendor offering a pilot program that allows your company to quickly get a prediction marketplace up and running. Make sure the prediction market application is easy to use and understand but also offers the level of security necessary to protect sensitive data. Difficult software often gets in the way of people's willingness to participate.

• Identify specific questions for a pilot marketplace (e.g. "Will we have an operational interruption this quarter at X location?") and determine how answers to those questions will be measured in order to cash them out and pay out your participants. Only questions that have quantitative outcomes are appropriate for a prediction market.
  
• Make sure there is an internal administrator of the marketplace responsible for introducing the concept, defining the rules, managing the prediction market software, providing "customer support" for participants, and communicating the results to as broad an audience as possible. Depending on the vendor you choose to work with, this does not need to be someone with an IT or financial background.
  
• Recruit internal "champions" who will give their support to the pilot. These leaders must be supporters of the concept and regularly encourage their people to participate. One of the best ways to drive participation is to show that those at the senior levels are paying attention.
  
• Identify a broad range of traders to participate according to the markets being run. Remember, widespread participation generally yields more accurate predictions. Typically, a market needs at least 12-15 participants to yield statistically "relevant" results.
  
• Define what incentives participants will receive. While the obvious approach may be to award prizes, the best incentive programs can also be "soft" incentives. Many organizations include both business and entertainment markets around everyday cultural topics (movies, music and sports) to keep the marketplace "sticky," allowing participants to ask their own questions, or even setting up a competitive environment between groups of people. Surprisingly, this method often proves more effective in maintaining trader interest than awarding prizes for participation levels.

Beyond the Prediction
Prediction markets are best known for their uncanny ability to produce more accurate forecasts, but it is the insight gained from analyzing the data generated during trading that may be the most important.

Mapping trading data to other demographic information about your participants may yield some surprises about who the most prophetic employees are. Contrarian viewpoints can also be easily detected and investigated further to see if there is merit to the opinion. The data may also reveal characteristics of the company previously unknown. For example, a recent study on internal prediction markets at Google showed newer employees were much more optimistic than veterans of the company. The study also revealed that employees who sat near each other in the office tended to have similar opinions about the issues. Both of these factors were previously unknown and such identification-and quantification-can be very valuable when thinking about organizational behavior and its impact on risk mitigation.

Through such means, prediction markets can be an effective way of augmenting any risk mitigation program to capture the collective intelligence of employees, business partners and customers.

Because of their consistently high degree of accuracy, prediction markets are sometimes seen as the proverbial crystal ball. But this is a mischaracterization of their capabilities. Instead, prediction markets help a business understand the probabilities of an event occurring and give insight into what employees' thoughts are on important business topics. Generally, the process of acquiring this untapped wisdom is virtually impossible given the cultural dynamic of today's corporation. With prediction markets, however, these insights can be transformed into quantifiable and actionable information.

Adam Siegel is the CEO and co-founder of Inkling Markets, a Chicago-based prediction market provider whose clients include Cisco, CNN and Wells Fargo.