Why it Pays to Heed the Herd
By Brad Wieners

(Business 2.0) – The title of New Yorker Columnist James Surowiecki's terrific first book, The Wisdom of Crowds (Doubleday, May 2004), pays homage to Charles Mackay's Extraordinary Popular Delusions and the Madness of Crowds, an 1841 classic that's enjoyed a revival since the dotcom bust. But while Mackay insisted that men "go mad in herds," Surowiecki makes the opposite argument--that "under the right circumstances, groups are remarkably intelligent, and are often smarter than the smartest people in them."

Surowiecki's book is bursting with vivid examples. The Iowa Electronic Market is a mock futures exchange that predicts election results with uncanny 98 percent accuracy. Google's page-ranking methodology is another instance of enlightened groupthink: By ordering search results according to the number of websites that link to matching pages, it harnessed the collective wisdom of Web surfers and made previous systems obsolete.

Surowiecki notes that groups are best at solving "cognition problems" (complex scenarios involving facts that are finite in nature), such as estimating the number of jelly beans in a jar. To be wise, groups must be large, diverse, and genuinely independent. "The best collective decisions are the product of disagreement and contest, not consensus or compromise," he writes.

What accounts for egregious groupthink failures like the Internet bubble? Surowiecki faults investors' willingness to suspend their judgment in favor of cues from high-profile Wall Street "experts." Maybe. Or maybe even a smart mob can be dulled by greed or panic. At which point it becomes just a mob. -- BRAD WIENERS