Committees & Decision Making: Lies, Damned Lies & Heuristics
TL;DR: How business say they use committees to make decisions, and what happens in practice rarely align. Decision processes can be corrupted in ways that are invisible to decision-makers. If risk teams aren't empowered to challenge the design, operation and outcome of decision systems, major organisational risks remain unmanaged.
I have noticed a consistent gap between how businesses say committees make decisions and what actually happens. An organisation may hold contradictory versions of the story, none of which reflect reality. Whilst this gap is rarely deliberate, it is always a barrier to effective risk management.
The views and opinions expressed on this account are my own and do not reflect the official policy or position of my employer. Any content provided is for informational purposes only and should not be considered or relied upon as professional advice.
Committees and decision-making
A committee is usually imagined as a diverse group that assesses evidence objectively and makes balanced, reasoned decisions. Sometimes this is accurate. Often it isn't. Committee decision-making often falss short in three ways:
The committee's role is theatre, with the decision made elsewhere. It may make sense to make decisions elsewhere, but decision theatre makes effective risk challenge harder.
The committee’s review was neither fair nor effective. Members defer to the decision sponsor rather than testing the proposal objectively.
Sponsor limits oversight. The committee lacks the information, time or scope needed to do their job.
Risk teams often aren't empowered to challenge the design or operation of the management systems used to make decisions. Decisions made outside committees, may come to risk teams too late for meaningful challenge. Even when a committee appears to own a decision, flaws in the process can still limit risk effectiveness.
The belief that committee ownership produces better outcomes rests on assumptions that are rarely stated, tested or reviewed. To see where those assumptions fail, we need to examine committee design, committee operation and the handling of individual decisions.
Committee design
The wisdom of crowds
We assume that diverse, independent groups make better decisions.
Aristotle wrote about the intelligence of crowds, but the modern idea is often linked to Francis Galton. He observed that the median estimate of people guessing an ox’s weight was within 1% of the true figure. Averaging independent judgements cancels out some individual error and produces a better estimate. But the effect has important caveats:
Independence matters. If members influence each other, diversity of thought declines.
The effect works best for numbers. Committees rarely break decisions into measurable comonents to take full advantage of this effect.
Simple models can outperform groups. By reducing noise from human judgement, models often produce more consistent results.[i]
Context matters. Group resistance to new ideas increases:
The further away from the status quo a new idea is.
The closer a new idea gets to an expert’s speciality, the more likely they are to think it is bad[ii]. This is because an expert can generally see more ways in which an idea can fail.
Mediocrity bias arises because group opinion tends to be more polarised around more innovative ideas. Mediocre ideas tend to attract less negative attention, so they can more easily achieve consensus.
A committee needs to preserve its status to retain its relevance. It sometimes has to balance objectivity with meeting organisational expectations. That can make it reluctant to delay or reject ideas backed by powerful sponsors, or to reach conclusions that more powerful groups may reverse.
Thought diversity through role diversity
Committees design tends to assumed that cross-functional representatives improves decisions. This is because they will bring more relevant “local knowledge”, and be exposed to different ideas. [iii]
Role diversity passively increases diversity of thought: organisations rarely formally assess whether committee members are updating their knowledge to align with decision-making responsibilities. A system that depends on individual diligence is vulnerable to knowledge gaps, blind spots, “busyness”, and underperformance.
Discipline diversity
Academic and professional backgrounds can act as a proxy for how they think. A committee made up of different disciplines is assumed to consider a wider range of evidence, reasoning and success criteria. For instance, when considering traffic on a road, a safety expert sees an accident risk, an ecologist environmental harm, and a doctor sees a cause of respiratory illness.[iv]
The assumptions underlying this theory can break down:
Where a discipline’s lens is not particularly relevant, the individual may recuse themselves from the discussion, or seek to inappropriately utilise their lens.
Some lenses can be explicitly or implicitly favoured by the organisation, effectively concentrating decision-making in the hands of the few representatives of the “correct” specialism.
Skillsets can be applied inconsistently. For instance, a data scientist may be more likely to utilise their training where it supports the management view, whilst playing up limitations in their analysis when it contradicts it.
Group identity can trump discipline identity. The balance between the two can change over time, and vary with decision type.
Lens diversity
The limits of discipline diversity can be offset by introducing structured lenses. Surgical checklists, for example, improve decision-making by providing a procedural lens while still allowing doctors to use their expertise when circumstances fall outside the standard process[v].
Lens diversity fails when people are not incentivised to contribute, or are discouraged from contributing, their perspective. Incentivisation need not be financial: limited time, heavy workloads, culture and business norms can all push members towards a mechanical use of prescribed lenses.
Groups can also become complacent. If the prescribed lenses work for a run of decisions, members can become over-reliant on the standard process, and incurious about identifying its limitations.
Committee operation
The everyday operation of committees often creates shortcuts that limit members’ ability to contribute. This is rarely intentional.
The first speaker in a discussion often shapes the final outcome, even when their argument is no stronger than anyone else’s. That person is often senior, controls the information presented and already has a preferred outcome. This makes it harder for the group to identify and reject a weak proposal.
Group polarisation
Group polarisation is a feedback effect in which the group reaches a more extreme position than its members would have reached independently. If most members favour investing in a new product, discussion can lead to a larger investment and greater confidence in success. This caused by two dynamics:
Persuasive arguments: Most arguments will support the majority view. As people hear new reasons for a position they already hold, they become more confident and move towards a more extreme version of it.
Social comparison: Members may adopt a more extreme position to signal loyalty and strengthen their group status.
Information pooling
People are more likely to repeat information that others already know than to share information unique to them. Whilst we assume that a group has access to more knowledge than any individual, in practice, valuable information may never enter the discussion.
Information leakage
Institutional memory imperfectly captures the nuances, limitations and compromises of a decision. People tend to discount facts that contradict their worldview. Urgency, denial, conventional thinking and sheer busyness can all narrow the information considered and remembered. Short-term operational pressure and rigid decision structures leave little room for research, development or deeper investigation. A desire to maintain consensus provides a further incentive to downplay any disagreement.
This loss of information weakens later decisions that depend on the original one and makes it less likely that the organisation will reopen a decision when new evidence emerges.
Identity fusion
Identity fusion arises when people merge their identity with a group or leader[vi]. They may then adopt a leader's, or a majority's, ideas uncritically. Diversity and independence of thought disappear, while the committee becomes political theatre: a performance of robust governance that can shield the group from genuine challenge.
We are not (necessarily) talking about cult-like behaviour. Keynes’s “reputation calculation” [vii] shows that dissent carries personal and professional costs, while displaying loyalty to the in-group may bring influence and advancement. A successful contrarian looks reckless; an unsuccessful one looks foolish.
Decision theory also suggests that when knowledge is uncertain, doing nothing can look safest[viii]. Change creates execution risk for the organisation and reputational risk for its sponsor; the status quo usually has no obvious owner. Most organisations do not punish people for moving away from the status quo too slowly. The more established a process is, the less likely any one person is to face consequences when it produces a poor decision.
How committee decision-making becomes corrupted
Decision sponsors can weaken a committee’s ability to reach an outcome that conflicts with their preference. Their influence usually increases with their organisational status and power. Their actions can be intentional or accidental; pursued in good faith or bad.
Limiting committee effectiveness
Selective or non-engagement with a committee limits its ability to reject or amend bad proposals. Mechanisms include:
Restricting the level, robustness, or quality of analysis provided to support decision-making. Typically this is done by removing or downplaying uncertainty analysis, or taking advantage of methodological limitations.
Limiting the range of options for which analysis is performed.
Using the limited number of formal meetings available to pressure the committee to make a decision with limited or flawed analysis.
Bringing the committee in late on in the decision-making process, removing its ability to comment on decision purpose, scope, and methodology.
Politics
Reviews of committee effectiveness often assume that papers and minutes capture the full decision process. In reality, the decision may have been settled before the papers were even drafted. Mechanisms include:
Pre-meetings: the decision sponsor attempts to persuade a majority of members before the meeting on the merits of the argument, carrying out extra work to address concerns. This can bias the analysis performed in favour of the decision, and puncture the group’s diversity of thought.
Horse trading: the decision sponsor arranges support from a majority of members before the meeting takes place, not based on decision justification, but on personal power deals, intimidation, and favour-trading.
Agreeing the decision in advance with a senior stakeholder to whom the decision-making group is likely to defer, increasing the personal cost of dissent.
Utilising the opinion of other, less powerful groups, over which the decision sponsor has more sway, to gain legitimacy for their preferred outcome.
Rhetoric
How an idea is presented shapes how a committee understands its risks and uncertainties. Rhetorical arguments appeal to emotion, logic and values[ix]; shaping the assumptions on which the decision rests.
Concern about rhetoric obscuring truth is not new. Plato’s Gorgias, written around 2,500 years ago, explored its dangers in the pursuit of knowledge. In politics and law, rhetoric meets competing rhetoric. In committees, however, the proposer often holds the only rhetorical weapon. For a committee to do its job, members must first recognise this, then decide when it is obstructing analysis, and finally work out how to counter it.
Common uses of rhetoric in organisations include:
Over-emphasising the likelihood or severity of a future unknown risk, or missed opportunity.
Presenting a new idea as familiar to overcome status quo bias, for example by overstating its similarity to existing processes, decisions or business lines. Brexit supporters’ greatest rhetorical trick was presenting the biggest political change in a generation as the lower risk option.
Presenting a decision as if it were not really a decision. For example, approving a small amount of revenue in a new business line, knowing that this will remove obstacles to later expansion.
Simplifying complex issues, or overemphasising uncertainty to make the preferred decision more attractive.
Appealing to group identity – emphasising group loyalty, or the emotional impact on a competitor, over the merits of the decision.
Presenting the decision as a foregone conclusion: “if we don’t do this, someone else will”.
Using jargon: complex, technical language is used to disguise weaknesses in argument, or bury relevant details.
Presenting a false choice between action or inaction, when there are a range of potential actions that could be considered.
Being selective over the information provided to make the preferred outcome seem more appealing.
Lessons for risk teams
Risk teams should question untested assumptions. Organisations often lack clear guidance on good decision-making and rarely document the assumptions underpinning their design. Decision process design and operation tend to sit outside meaningful risk challenge.
Risk teams need strong decision-analysis capabilities. Many risk frameworks address only the symptoms of poor decisions, while any review of decision-making processes is often unstructured and superficial.
Redesigning risk frameworks and risk activity around organisational decision-making is a useful first step. But risk challenge will remain incomplete unless it tests the full suite of assumptions behind how decisions are made.
Risk teams should focus on:
Challenging the design, assumptions and operation of decision-making systems. Where group ownership is justified, risk should ask whether risk oversight belongs at framework level, individual-decision level, or both.
Developing a sociological view of how groups and individuals interact, including where behaviours can corrupt the decision process.
Pushing for clearer quantification of expected decision-outcomes and uncertainty.
Building feedback loops that assess both the quality of decisions and the value added by the management system.
Ensuring that risk is involved in material decisions from initiation through to outcome. Risk teams should be empowered to act as “decision referees” and call out where the process has been distorted or corrupted.
I hope this blog sparks ideas and discussion. If you found it interesting, please share or connect with me on LinkedIn to contribute or provide feedback!
[i] Noise: A Flaw in Human Judgement. 2022, Daniel Kahneman, Olivier Sibony and Cass R Sunstein, 978-0008309039
[ii] Range: How Generalists Triumph in a Specialized World, 2020, David Epstein, 978-1509843527
[iii] An individual’s peer group is most likely to be derived from those people they sit closest to As most office-based organisations arrange their desks by function, it is therefore highly likely that, once an organisation has passed a certain size, role diversity will also correspond to peer group diversity.
Knight, Caroline & Olaru, Doina & Lee, Julie & Parker, Sharon. (2022). The Loneliness of the Hybrid Worker. MIT Sloan Management Review.
[iv] Risk is a Construct: Perceptions of Risk Perception, 1993, Ulrike Becker et all, 3-926901-65-9
[v] Treadwell JR, Lucas S, Tsou AYSurgical checklists: a systematic review of impacts and implementationBMJ Quality & Safety 2014;23:299-318.
[vi] Identity Fusion, February 18 2015, William B. Swann, Jr. and Michael D. Buhrmester, Volume 24 Issue 1, Current Directions in Psychological Science
[vii] The General Theory of Employment, Interest and Money: with The Economic Consequences of the Peace, Chapter 12, John Maynard Keynes, 978-1840227475
[viii] Understanding Scientific Reasoning 5th Edition, 2006, Giere, Bickle, and Mauldin, 2006, 978-0155063266
[ix] HOEFER, ROLF L., and SANDY E. GREEN. “A RHETORICAL MODEL OF INSTITUTIONAL DECISION MAKING: THE ROLE OF RHETORIC IN THE FORMATION AND CHANGE OF LEGITIMACY JUDGMENTS.” The Academy of Management Review, vol. 41, no. 1, 2016, pp. 130–50. JSTOR, http://www.jstor.org/stable/43699322. Accessed 21 Aug. 2026.
Comments