You ask your team what they think. Someone offers a confident answer. The next person agrees, with a small addition. A few minutes later, the room appears ready to move on.
That might be genuine agreement. It might also leave you wondering what people thought before they heard everyone else.
You do not need to turn every meeting into a debate. For decisions that depend on information spread across a team, it is worth separating two activities: forming an initial view and discussing that view with others.
The practical question is simple: how do you make sure useful information reaches the conversation before the team settles on a position?
Agreement is not the same as a complete picture
In a classic experiment, Garold Stasser and William Titus asked small groups of students to choose between fictional candidates. Some groups received the relevant information in common; in others, important information was distributed across members. Groups were much less likely to choose the candidate favoured by the complete information when that information was distributed. Having something known somewhere in the group did not ensure it shaped the decision. This was a controlled student task, not a measure of workplace performance. Stasser and Titus, 1985
That distinction is useful for leaders. Asking for opinions is different from asking what information those opinions contain.
Three people favouring a proposal may be relying on the same report. One person questioning it may have seen a customer problem that the report does not capture. Counting preferences cannot tell you whether you have heard the relevant evidence.
This builds on the broader habit of asking better leadership questions. Here, the focus is the order in which you ask them.
Protect the first view, then improve it
Research on numerical estimates offers another caution. In a study by Jan Lorenz and colleagues, exposure to other people's estimates made answers more similar and increased confidence, without a commensurate improvement in collective accuracy. The experiment involved factual estimation, not a workplace discussion. Its useful warning is narrow: convergence alone is not proof of better judgement. Lorenz and colleagues, 2011
It would be a mistake to conclude that people should stop influencing one another. A later experiment found that social influence in decentralised networks could improve collective estimates even as answers became less diverse. The structure of that influence mattered. Again, this was an estimation experiment, not a test of a management meeting. Becker, Brackbill and Centola, 2017
The routine below is a practical application of these concerns, not a scientifically validated meeting formula. Its purpose is to preserve initial input and make room for information exchange. People should still be able to change their minds.
1. Write the decision before inviting an answer
Start with a specific choice.
“Thoughts on customer service?” is too broad. “Should we trial a different service roster at one location next month?” gives people something they can assess.
Write down the options, important constraints and who will make the final decision. Include the possibility that more information is needed. Make it clear whether you are gathering advice, seeking agreement or delegating the choice.
Everyone should work from the same basic brief. Independent thinking does not mean withholding essential information.
For high-stakes or safety-critical decisions, use the appropriate specialist review and established procedures. This lightweight routine cannot replace them.
2. Ask for a private first response
Before anyone advocates an option, invite each person to note:
- My provisional recommendation is…
- The evidence I am using is…
- Something others may not know is…
- One uncertainty, or fact that could change my mind, is…
Allow enough quiet time for the complexity of the decision. For a substantial proposal, gather these responses before the meeting instead.
The leader completes the same prompts. Holding back your preferred answer until the initial responses are recorded is a facilitation choice, not a claim that the studies above prove leaders must always speak last.
Explain the purpose: these are starting points, not positions people must defend. Do not use the notes to embarrass someone who later changes their view.
3. Share information before counting preferences
Reveal the responses together where practical. Begin with evidence, uncertainties and information that is new to the group.
Ask: “Which detail have we not considered yet?” Then: “How do we know that?”
Separate observations from interpretations. “We received six complaints about the current hours last week” is different from “customers hate the roster”. If the number itself has not been checked, record that too.
Do not treat every contribution as equally reliable simply because everyone had equal space to contribute. Relevant expertise, source quality and the limits of the evidence still matter.
An information gap is worth noticing even when it comes from someone who does not favour your preferred option.
4. Discuss the differences that could change the choice
Consider a hypothetical team deciding whether to extend service hours.
One manager favours the change because customers have requested it. Another is concerned about staff coverage. A third questions whether those requests came from regular customers or a short-lived promotion.
A quick vote gives you a preference count. A useful discussion asks which facts would make each option workable.
You might discover that demand is concentrated on one evening, or that the coverage problem has not yet been assessed. Those are possible findings in this fictional example, not evidence about any real business.
Choose the unresolved question that matters most to the decision. Assign someone to check it if necessary, rather than letting the discussion produce confident guesses.
5. Invite a revised view and record the reason
After discussion, ask each person what, if anything, changed their recommendation.
A changed view can be a sign that someone learned something. An unchanged view can also be reasonable. Ask for the reasoning rather than rewarding agreement with the leader.
The decision owner then records:
- What we decided and why.
- The important evidence and trade-offs.
- What remains uncertain.
- Who owns the next step.
- What would cause us to reconsider.
You may still decide without unanimous support. Make the remaining disagreement visible and explain how it was considered. Do not describe an advisory discussion as a team decision if one person retains authority.
Catch, Challenge, Change in the meeting
Catch–Challenge–Change provides a simple way to remember the leadership move.
Catch the assumption that a quiet room or quick agreement means the whole picture is understood.
Challenge it by asking for independent input and evidence that has not yet been heard.
Change the sequence of the conversation: initial thinking first, information exchange next, decision and reasons last.
You do not need a new process for every routine choice. Try this with one upcoming decision where different people hold different parts of the picture. Afterwards, ask whether you surfaced anything important that the usual conversation might have missed. That is a more useful starting point than declaring the meeting a success because everybody agreed.
Bring a real decision into the room
If your leaders need practice questioning assumptions and making more considered decisions, explore Andy Nunn's Leadership Mindset Workshop.
For a conference audience, The Human Advantage leadership keynote introduces broader themes of clearer thinking, assumptions and leadership through change. Share your audience and the decisions they face to discuss which format may fit.
