3 ways internal prediction markets can surface strategic signals amid data noise

For a long time, corporate forecasting has seemingly resembled old-school weather prediction: heavy on models fed with historical data in the hopes that patterns in what happened before could yield insight into what will happen next. Incorporate some sensing data on the current state, and many organizations hoped they’d have enough to go on to forecast future outcomes. 

There’s a better way—or several better ways—in which modern-day organizations continue to predict in this data-driven, AI-enabled era. One promising approach to consider is prediction markets. Deloitte is researching how prediction markets could evolve as investment and hedging tools in the financial services industry, and organizations should reconsider developing their own internal prediction markets as insights tools—a way to capture and leverage the wisdom of the crowd. 

Prediction markets are trading platforms where participants trade contracts based on how they believe future events will unfold. It’s a bet on the probable outcomes based on a dynamic market estimate that moves when new evidence arrives, courtesy of the anonymous prediction market participants.  

In addition to using prediction markets for market sensing and hedging, organizations can also set up their own internal prediction markets. It’s an idea that was introduced in the 1990s but didn’t take off for a variety of reasons. Should organizations breathe new life into this tool—shaping it with smart incentives and strong ethical boundaries to tackle the challenges faced today? 

In the not-too-distant future, organizations might set up their own internal prediction markets, inviting employees to consider probable future business outcomes based on their unique knowledge, perspectives, and experiences—the kinds of insider knowledge that generative AI can’t help with—and then using that crowdsourced insight to help drive strategic decision-making. 

By aggregating information from across teams and crowdsourcing perspectives on future events, organizations could surface signals that they otherwise might have missed about direct or indirect events affecting things like pipeline conversion, launch timing, churn risk, capacity constraints, or strategic milestones. These internal prediction markets could act like numerous fingers in the wind predicting future weather patterns heading the organization’s way. 

Here are three ways organizations could benefit. 

1. Sensing: Many companies already model energy consumption, raw material supply, geopolitics, and more in scenario planning. Internal prediction markets could provide new signals on the same uncertainties and, crucially, on events that indirectly affect their business—information that parts of the organization might be fully up to speed on and others could benefit from knowing. Tapping into the wisdom of the crowd can help organizations determine the likelihood of a regulation update (which the legal department is likely tracking) or the projected arrival date of quantum computing (which probably is baked into the CTO’s scenario planning), for example. Internal prediction markets can enhance traditional sensing efforts because they are designed to capture a broader set of perspectives, helping to counteract “group think” or a single data stream for decision-making. 
2. Hedging: Many companies already hedge interest rates, FX, commodities, and input prices. Internal prediction markets that cover more events could help companies manage exposures that are currently hard to price—for instance, weather-driven demand shocks, supply chain disruptions, or risks embedded in technological innovations. 
3. Evergreen insights: Internal prediction markets are always-on data aggregators, constantly updated with new inputs, so they can yield signals that reflect current-state insights, rather than providing limited perspectives from a point-in-time data upload.  

Like the investment tools they’re modeled after, internal prediction markets are probabilistic and, therefore, not infallible. They won’t eliminate uncertainty—but they can help expose it and make planning and forecasting a more robust discipline, when triangulated with other data sources. In an era of increasing complexity, volatility, and data proliferation, they could offer a strategic signal worth tuning into.