Wiser: Getting Beyond Groupthink to Make Groups Smarter: Summary & Key Insights

There’s a seductive idea that groups, by their very nature, are smarter than individuals. It’s captured in the phrase “the wisdom of crowds.” When conditions are right—when independence is preserved, diversity of opinion is celebrated, and participants draw upon distinct pieces of information—groups do indeed outperform their members. Studies of market forecasts, political predictions, and even jelly bean counting contests show how aggregated judgments can yield remarkably accurate results.

But the romantic story of the “wise crowd” fails in most real-world settings, because those ideal conditions are quickly violated. Independence collapses when members hear what others think before offering their views. Diversity erodes as groups self-select or fall under the sway of dominant personalities. And instead of aggregating distinct insights, groups tend to amplify shared beliefs.

One of our key examples involves corporate forecasting and governmental decision-making. When analysts or policy advisors begin conforming their views to perceived group norms, errors multiply. It isn’t because they lack intelligence, but because they read cues of approval and disapproval in subtle ways: a nod from a superior, a hesitant pause, the early framing of a problem. In short, groups become echo chambers, not engines of discovery.

The myth of the wise crowd invites us to relax, to trust that collective processes will take care of themselves. We urge the opposite. Group wisdom is not automatic—it must be cultivated by preserving independence, encouraging informational diversity, and managing the influence of authority. Otherwise, the crowd becomes, quite literally, more foolish than the individuals within it.

At the heart of most group failures lies a deceptively simple dynamic: people focus on what everyone already knows. It is far easier to discuss familiar facts than to bring forth unique, potentially controversial information. Psychologists call this the *common information effect*. Imagine a hiring committee evaluating candidates. Each member has reviewed certain files and formed private impressions. During discussion, they naturally gravitate to shared experiences: the same impressive credential, the same recommendation letter. Rarely does someone insist on attention to a detail known only to them—perhaps a subtle warning in a reference—or persist long enough for it to influence collective judgment.

The result is surprisingly consistent: groups spend over 70 percent of their time discussing shared information, and only a small fraction on data known to a subset of members. The tragedy is that unique information is often decisive. By neglecting it, groups discard the very diversity that could make them smarter.

We saw this pattern vividly in research with juries deliberating complex evidence. Jurors arrived with distinctive pieces of memory and interpretation. Yet through discussion, their varied observations compressed into an artificial uniformity. Once a dominant narrative took root, disconfirming data was quickly dismissed. The same dynamic pervades boardrooms, think tanks, and crisis management teams.

The lesson is straightforward yet deeply challenging: groups must be designed to surface unshared information early and fully. Structured round-robin formats, deliberate calls for minority opinions, anonymized input systems—each of these can help. But at a cultural level, the group must send a clear message: the highest loyalty is not to consensus, but to truth.