There is an unlikely field that studies this problem with unusual clarity: avalanche safety.

In avalanche safety training, there is one rule that overrides all others: if a single person in the group says “no,” everyone turns around. Corporate boards could learn something from that.

Corporate boards make some of the most consequential decisions in business — acquisitions, strategic pivots, leadership transitions, major capital allocations. Yet when those decisions appear in board minutes, they are almost always recorded as unanimous. Research suggests dissent occurs in only about 1% of board decisions. That unanimity often reveals as much about group dynamics as it does about genuine agreement.

The rule exists because of a pattern instructors see again and again. Someone senses something is wrong — unstable snow, deteriorating conditions, a risky route — but speaking up means challenging the plan and slowing everyone down. In larger groups especially, that voice often stays quiet.

The most dangerous variable, instructors often say, is not the snowpack. It is the group.

Corporate boardrooms operate under strikingly similar conditions. Directors must make consequential decisions with incomplete information: often within the compressed time frame of a board meeting. The question isn’t whether boards face pressure to align. It’s whether that pressure is silencing the most important voices in the room.

Consensus has obvious virtues. Boards function best when directors ultimately align behind a course of action. A unified board gives management clarity and confidence in execution.

But consensus can be a signal. It can also be a warning.

Anyone who has spent time in boardrooms recognizes how quickly the momentum of a conversation can tilt toward agreement. Management presents a proposal. A director offers a supportive observation. Another suggests refinement. Gradually, the discussion shifts from whether the proposal is sound to how it should be implemented.

Eventually the chair looks around the table and asks a familiar question: “Is everyone comfortable moving forward?”

Directors sometimes recognize the dynamic only after the meeting ends. Following a unanimous decision on a major initiative, someone may quietly remark in the hallway, “I had some reservations about that.” Another director admits they did as well. In the room itself, however, those doubts never surfaced.

Seasoned investors understand the value of dissent. Warren Buffett has long argued that the best boards are those where directors are willing to challenge assumptions rather than simply ratify them. But even strong boards can find that once a discussion begins to converge, raising a late objection becomes psychologically difficult.

Psychologists call this dynamic groupthink: the tendency of cohesive groups to suppress disagreement in pursuit of harmony. Boardrooms are particularly susceptible: — directors meet periodically, relationships are collegial, and open disagreement can feel unnecessarily disruptive.

Avalanche educators warn about the same pattern. As groups become larger, responsibility diffuses and individuals become less likely to challenge the emerging consensus. The very structure of the discussion can begin to suppress caution.

If that dynamic shows up in boardrooms — and the evidence suggests it does — improving board decisions isn’t only about who sits at the table. It’s about how decisions are made once everyone is there. Boards have spent decades focused on composition: independence, diversity, expertise. The next frontier is deliberation.

Some boards already experiment with structured disagreement. In evaluating major transactions, directors may organize “red team/blue team” exercises, assigning one group to argue for a deal while another is tasked with challenging it. The objective is to stress-test assumptions before committing capital.

Yet most board deliberation still takes place in a single conversation around a table. That format encourages the emergence of a dominant narrative before competing analyses have had time to develop.

Boards might consider what could be called parallel deliberation: briefly breaking into smaller groups before reconvening to compare conclusions.

After management presents a proposal, the chair divides directors into small groups and asks each to answer the same three questions: What assumptions must be true for this plan to succeed? What could cause it to fail? Under what circumstances would we say no? Fifteen minutes later, the board reconvenes and compares conclusions before continuing the discussion.

Such a structure introduces several useful dynamics. Smaller groups lower the social cost of dissent. Independent discussions generate multiple lines of analysis rather than a single conversational path. And by interrupting the momentum of a room-wide consensus, the structure helps surface concerns that might otherwise remain unspoken.

The goal is not to manufacture disagreement. Boards ultimately need alignment. But alignment reached through rigorous debate is far stronger than consensus that emerges quietly around the table.

In avalanche training, the group turns around when one person says no.

In boardrooms, that same voice is the one most likely to stay quiet — and the one most worth hearing.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.