The entrepreneurial instinct is to bring important decisions to the team. This instinct is correct under specific conditions and actively harmful under others. Understanding which conditions produce which outcome changes not just how meetings are run but whether to run them at all.

The four conditions that make groups smarter than individuals

Surowiecki’s (2004) synthesis of collective intelligence research established four conditions necessary for group judgment to outperform individual experts. The first is diversity of perspective: different people must bring genuinely different information and analytical frameworks, not variations on the same worldview. The second is independence: individuals must form their views without social influence from each other before sharing them. The third is decentralisation: people draw on local or specialised knowledge rather than deferring to a central authority. The fourth is aggregation: a mechanism must exist for combining private judgments into a collective decision without those judgments contaminating each other in the process.

When these four conditions are met, groups reliably outperform the average individual and frequently outperform the best individual. When they are violated, the group produces worse decisions than many of its members would have produced alone.

Most entrepreneurial strategy meetings violate all four simultaneously. They bring people with overlapping backgrounds and shared investment in the company’s narrative — low diversity. They discuss openly before private views have formed — no independence. They are led by the CEO whose framing anchors the discussion — no decentralisation. And they aggregate through open conversation that produces conformity and anchoring rather than genuine preference aggregation. The group is meeting without the conditions that would make the meeting productive.

Why the most cohesive teams make the worst decisions

Janis’s (1972) groupthink analysis of major foreign policy disasters established the conditions under which groups produce systematically worse decisions than the best individuals within them: high cohesion, insulation from outside opinion, directive leadership that signals preferred solutions, high stakes, and the absence of systematic decision procedures.

Each of these conditions describes a high-performing entrepreneurial team. They are cohesive — often built around shared founder identity and collective belief in the mission. They are directive — the CEO’s view dominates, explicitly or implicitly. They are isolated — the team has built a shared consensus reality about the market, the product, and the competition that outsiders challenge and insiders validate. And they are under constant high-stakes pressure that would make dissent feel like disloyalty.

The failure modes Janis documented are each observable in entrepreneurial decisions that fail: the team that collectively decides the competitor is inferior without examining the evidence. The team that never voices serious doubt about the core product assumption because silence has been mistaken for agreement. The team that mistakes the absence of expressed disagreement for genuine consensus. Each of these is groupthink operating in a startup context — and each produces the confidence-without-calibration that the individual team members, questioned privately and independently, would not have endorsed.

What happens to good ideas in group discussion

Diehl and Stroebe’s (1987) brainstorming research documented that face-to-face groups consistently produce fewer and less creative ideas than the equivalent number of individuals working alone — through production blocking (people cannot speak simultaneously and lose their ideas while others talk), evaluation apprehension (people self-censor in front of others), and social loafing (individual effort reduces in group contexts). The same mechanisms that suppress idea generation in brainstorming suppress the contribution of dissenting assessments, minority views, and devil’s advocate positions in decision discussions.

The person who has a dissenting view about the strategic direction is navigating production blocking, evaluation apprehension, and social pressure simultaneously in the meeting. Lorenz, Rauhut, Schweitzer and Helbing’s (2011) PNAS study confirmed the specific social influence problem: when people can see others’ estimates before forming their own, the statistical diversity of judgment that produces collective intelligence is eliminated. The meeting where participants hear the CEO’s view first, then contribute their own, is a social influence study in progress — and it produces the same result.

What structural changes actually recover group decision quality

The independence condition is the most tractable to address. Requiring meeting participants to write individual assessments before convening discussion — the brainwriting principle established in Paulus et al.’s research — preserves the diversity of judgment that open discussion destroys. The Amazon six-pager is the institutional version: participants read the narrative memo in silence before discussion begins, forming individual views before social influence operates. This single structural change recovers a substantial portion of the collective intelligence that the discussion format would otherwise eliminate.

The red team and the devil’s advocate address the evaluation apprehension mechanism. Schwenk’s (1990) meta-analysis confirmed that explicitly assigning a person or sub-group to generate the strongest possible case against the group’s preferred option significantly improves decision quality in cohesive groups. The structural assignment removes the self-censorship that evaluation apprehension produces: the devil’s advocate is not expressing personal doubt — they are performing an assigned function. The institutionalisation of the dissenting role makes dissent available at exactly the moment that groupthink dynamics would otherwise suppress it.

Pre-commitment of individual views before the group convenes addresses the anchoring mechanism. Investment committees that require written private recommendations before the committee discussion consistently show fewer anchoring effects and less status-driven deference than those that discuss without prior commitment. The private pre-commitment preserves the independence that the discussion would contaminate.

When group decisions are worth making and when solo decisions are better

The conditions framework provides a practical decision rule for when to involve groups. Decisions where group members have genuinely different information, formed independently, that can be aggregated without social contamination — these benefit from groups. Decisions where the group shares the same information, the same framework, and the same stakes — these produce groupthink, not collective intelligence.

For the entrepreneurial team, this identifies a specific set of decisions where group involvement is likely to degrade rather than improve quality: strategic direction decisions where the team has shared investment in the existing direction, product assumptions that the whole team believes, and competitive assessments that the team has collectively constructed. These are precisely the decisions most likely to be brought to the group — and most likely to be degraded by the groupthink dynamics that cohesive, high-stakes teams reliably produce.

Books worth reading on this

Victims of Groupthink by Irving Janis is the foundational account of collective decision failure — the specific psychological mechanisms, the observable symptoms, and the case studies that make the failure pattern legible in real decision contexts. For the entrepreneurial team that has experienced the post-mortem feeling that everyone privately had doubts nobody voiced, Janis’s analysis of how that silence was produced is the most precise available diagnosis. His account of the structural corrections Kennedy implemented after the Bay of Pigs provides the most historically documented case study of how groupthink is specifically addressed.

If the dynamics described here are significantly affecting your wellbeing, speaking with a psychologist is the right next step. UK: Samaritans (116 123, free, 24/7). Mind (0300 123 3393). BACP: bacp.co.uk/search/Therapists. Crisis Text Line — text HOME to 741741 (US, UK, Canada, Ireland). International: internationaltherapistdirectory.com.

This article is for educational and informational purposes only. Sources: Surowiecki, J. (2004), The Wisdom of Crowds, Doubleday. Page, S.E. (2007), The Difference: How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies, Princeton University Press. Janis, I.L. (1972), Victims of Groupthink: A Psychological Study of Foreign-Policy Decisions and Fiascos, Houghton Mifflin. Diehl, M. & Stroebe, W. (1987), Productivity Loss in Brainstorming Groups: Toward the Solution of a Riddle, Journal of Personality and Social Psychology, 53(3), 497–509. Lorenz, J., Rauhut, H., Schweitzer, F. & Helbing, D. (2011), How Social Influence Can Undermine the Wisdom of Crowds, PNAS, 108(22), 9020–9025. Schwenk, C.R. (1990), Effects of Devil’s Advocacy and Dialectical Inquiry on Decision Making: A Meta-Analysis, Organizational Behavior and Human Decision Processes, 47(1), 161–176. Paulus, P.B. et al. (2006), Asynchronous vs. Synchronous Idea Generation in Face-to-Face and Computer-Mediated Environments, Applied Ergonomics, 37(6), 803–812. Janis, I.L. (1972), Victims of Groupthink, Houghton Mifflin. Syed, M. (2019), Rebel Ideas, John Murray. Gawande, A. (2009), The Checklist Manifesto, Metropolitan Books.