Q&A: The Case for Crowdsourcing War Games

One Washington-based consulting firm, whose clients include the Pentagon, is betting on the wisdom of the crowd.

For the past few years, Wikistrat has drawn upon a network of about 2,000 analysts, posing questions about potential international trends and crises. For instance, what are the geopolitical implications of outcomes in the Turkish elections? The result of an interoceanic canal in Nicaragua? In addition to the Pentagon, its customers -- the ones posing the questions -- include the Royal Air Force, the French Air Force and NATO. 

Last week, the company announced it was starting a few new services -- letting customers poll teams of analysts about outcomes in simulated war games and using the simulations to test-drive their own crisis management skills, among other new features. About a year and a half ago, Wikistrat started a subscription service that gives customers access to an information dashboard, constantly re-populated with new insights from analysts.

Shay Herkshkovitz, who oversees what Wikistrat calls its "analytic community," talked with Nextgov about why the company is putting so much stock in the crowd. This conversation has been edited for length and clarity.

NG: How do you choose the "crowd"?

SH: Each one of them is a subject matter expert in various fields -- that varies from regional expertise to topical ones. People who are experts on the Middle East, people who are experts in naval forces, grand strategy and so on. We do leverage the wisdom of the crowd, but this is a crowd of experts. 

Either we initiate contact with people who we know, who we found through research. Alternatively, we have people asking to join our community from which we carefully select those who meet our professional criteria. The third way we bring people on board is word of mouth. People who are members in our community refer their peers to Wikistrat.

NG: What incentives do you give these analysts, who are weighing in on issues of global importance?

SH: The first is professional in nature. People want to exchange ideas, they want their voices to be heard. The second is their desire to actually influence. Those who perform well in our internal activities are invited to a client project. Since our community knows that most of our clients are government clients, they want to make a difference. 

When people are participating in client simulations or engagements, they get paid. We pre-determine the amount of money we pay them. 

We are working under the assumption that neither the identity nor the research questions of our clients will be disclosed.[Participants] will know what the research question is, [but] will not know who they are working for.  

NG: What kinds of questions do your clients ask?

SH: What we are normally asked is to provide is trend analysis, and some kind of forecasting service and sometimes early warning mechanisms. 

[An example could be], "What will be the key drivers that will influence security issues?" If someone wants know how will the conditions in Libya -- some kind of a black hole that attracts all kinds of extremists -- "How it will affect the entire Mediterranean Basin?"

In the case of a university and academia, they can actually experience what it means to manage a crisis. We also offer these services to government entities who'd like to train their people in real-time decision-making. 

NG: Where do you host your simulations?

SH: Our platforms enable us to conduct roleplaying games, simultaneously involving numerous potential participants. 

Let's say we wish to play [out], for example, threats to the U.S. Navy in South East Asia -- not a topic we deal with. We can assign a group to play the Chinese, another group of experts to play the U.S. Navy. We're not asking random people to play these actors; we actually use experts on these fields. 

If you would like to play the Indian fleet, then we have a former senior executive with the Indian military, for example, on the strategic level.

We did an internal simulation -- there isn't any end user for this, the simulation was basically for product development. The basic scenario was a drone strike of the U.S. government in Chad that went bad, and, according to the story, they hit a civilian facility. Obviously, a tragedy, but also a crisis. 

(Image via venimo/ Shutterstock.com)