Because negotiation is a key strategic element of both transactions and litigation, the question is whether we can draw insights from large data sets that help lawyers add value for their clients in real time.

Human Behavior Varies—Often Irrationally—But It Is Predictable Even When Irrational

It turns out that the negotiation of litigated cases is more nuanced than these one-sized general rules. The negotiation of litigated cases usually involves a dance that divides into roughly three phases. Some are tangos while others are waltzes, but effective negotiators engage in a pattern of reciprocating behavior that tests the strike price for a deal over multiple rounds. Short-circuiting the negotiation dance often leaves money on the table. Figure 1 shows actual negotiations plotted with dollar moves coming together along the horizontal axis and time running from the start of the mediation down the vertical axis to a deal.

Patterns Emerge from Large Data Sets

Humans are predictable—really predictable. The National Security Agency wants our cell phone data because the phone companies can predict where we’ll be tomorrow with 93 percent accuracy. Make a credit card charge outside of your established pattern, and you’ll get a text or call from the bank within seconds.

Lawyers in legal negotiations are also very predictable. Not only do their early moves telegraph where they are headed when matched to historical patterns, the pace of play is also predictable. Picture It Settled has spent years building a system of neural networks and learning algorithms that compares each move in a legal negotiation to more than 15,000 other cases—a much larger data set than would be available in a clinical trial. (Full disclosure: The author is president of Picture It Settled.) After a few moves, the system can predict your opponent’s next move within minutes and dollars. Armed with that information, you will know with high certainty where the other side is headed before they get there—much less guesswork. You can fine-tune your strategy to subtly affect the pace of concessions and the eventual outcome.

Of course, there is no cookie-cutter way to negotiate a case. But the larger the data set, the smaller the chances become that someone has an untried pattern that works.

Extreme Positions Sometimes Pay Off but Don’t Work Most of the Time

The data indicate that taking an extreme position early in a negotiation sometimes pays off but much more often results in impasse or sudden drops to avoid impasse that end up conceding more than a strategic concession plan would have produced. Holding an extreme position too long and then conceding at the last minute can leave 15 percent or more on the table. That’s $150,000 in a $1 million claim. This insight flies in the face of the conventional wisdom and mythology of legal negotiation. The definition of an extreme negotiating position, however, varies by venue, claim type, and other variables.

John Travolta played a lawyer in the movie A Civil Action whose opening offer was so outside of the social convention for such negotiations in Boston in the early 1980s (over 35 times the eventual settlement) that it failed to even draw a response. The plaintiffs’ lawyers and their financier had valued the case at $25 million. Had Travolta’s character had the benefit of modern analytics combing data in similar cases from the Boston area, he would have known that a 2.5 multiple was more in line with convention for the venue and case type. Had he started around $62 million, there was a much better chance he could have landed a settlement in the $25 million range. Instead, his 35 multiple failed to draw a response and he and his partners lost their homes and went bankrupt pursuing the case for years to an $8 million settlement.

Insight Becomes Actionable

Accurate forecasts are insightful, but only helpful if you act on the information.
Once you know where the other side is headed, you can adjust the target settlement (represented by the dot at the bottom on Figure 3) to improve the round without increasing the risk of impasse. The system recalculates suggested offers that will get you to the adjusted target settlement incrementally, rather than with sudden moves. Because these moves are based on successful rounds, your odds improve.

If you get too aggressive, the model will show an increased risk of impasse. By continually adjusting expectations and strategy to the current forecast, you can test whether your trial alternatives are better than the projected deal. Even small percentage improvements usually yield much better settlements. Because the strategy is informed by successful and unsuccessful historical rounds, the improvement comes without out unnecessarily increasing the risk of impasse.

Conclusion

Big data and smart analytics will rapidly extend what experimental psychologists, behavioral economists, and other disciplines have learned about predictable if seemingly irrational human behavior.

Current technology allows us to play Battleship with sonar in negotiations. Knowing with some certainty where the other side is headed in time to improve your position through a research-based, fine-tuned concession plan will improve your results. It’s not a substitute for well-honed intuition developed through experience. It’s an aid to test and calculate optimum positions. It’s really nothing more than adding a scope to a gun so the human takes a better shot. A 5 percent improvement in a $10 million case amounts to $500,000. That’s worth some planning.

Read original article


Tags: