Survival on the Moon (NASA exercise)

Survival on the Moon is a consensus exercise designed by Jay Hall in the early 1970s. Participants embody a crew stranded on the sunlit side of the Moon, approximately 300 km from the mother ship, with 15 intact items to prioritise for survival and to reach their goal. The game unfolds in three phases: individual ranking, team ranking by consensus, and then comparison with a reference grid assigned by NASA. It serves to work on the quality of collective decision-making, listening to minority opinions, and the difference between consensus and voting.

Walkthrough

1. 1

   Set the framework and distribute the scenario

   5 min

   The facilitator presents the objective: "We will observe how a group decides, not just seek the right answer." They distribute the scenario, the list of 15 intact items, and an individual ranking sheet. They remind participants of the key data: the crew is on the sunlit side of the Moon, about 300 km from the mother ship, and must prioritise the items from 1 to 15. Participants read silently and may only ask questions for understanding the scenario, not for strategy.

   Tip — Emphasise from the start that the score is just a tool: this prevents the more competitive individuals from overshadowing the listening dynamic.

2. 2

   Conduct the individual ranking

   8 to 10 min

   The facilitator gives the instruction: "Individually, rank the 15 items from most useful to least useful for reaching the mother ship: 1 means the most important, 15 the least important." They ask participants to work without discussion, to achieve a true individual decision. Participants fill in their grid and may briefly note their reasons. The facilitator ensures silence and announces the remaining time halfway through and one minute before the end.

   Tip — Ask participants to keep track of their reasons: this will enrich the discussion later and prevent random rankings.

3. 3

   Form teams and establish the consensus rule

   5 min

   The facilitator forms groups suitable for the number of participants, keeping teams small enough to allow everyone to speak. They specify: "Your team must produce a single common ranking, from 1 to 15. You are not voting: you are seeking an agreement acceptable to all." They define consensus as a decision that everyone can support, even if it is not their first choice. Participants settle in with a blank team grid and their individual rankings.

   Tip — Clearly write 'no voting' on the board: as the pressure of time increases, groups will spontaneously revert to majority voting.

4. 4

   Build the team ranking

   20 to 25 min

   The facilitator initiates the work: "Compare your priorities, listen to disagreements, and produce a unique ranking of the 15 items." Teams discuss item by item or by families of utility, then record their final ranks. The facilitator circulates without giving hints about the solution; instead, they prompt with "Who hasn't spoken yet?" or "Which minority argument deserves to be tested?" They remind participants that the goal is to decide together, not to rush at all costs.

   Tip — Watch out for self-proclaimed experts: invite them to formulate verifiable hypotheses rather than imposing an authoritative answer.

5. 5

   Compare with the reference grid

   8 to 10 min

   The facilitator distributes or projects the reference grid assigned by NASA, without detailing it before the end of the work. They indicate the known benchmarks: in this reference, oxygen tanks and water are at the top; matches are last, as there is no oxygen to burn; the magnetic compass is nearly useless due to the lack of a global magnetic field. Each participant calculates their individual score by adding up, for the 15 items, the absolute difference between their rank and the reference rank. Teams then calculate their collective score using the same method.

   Tip — Provide an example of calculation on a single item with fictitious ranks: this avoids errors without revealing or discussing the entire correction.

6. 6

   Read the results and observe collective performance

   5 min

   The facilitator collects individual scores and team scores, then displays them simply. They remind the group that the lower the sum of the differences, the closer the ranking is to the reference. The group observes whether the collective decision has improved, matched, or degraded the individual decisions. The facilitator asks: "What in your decision-making explains this result?" to transition to process analysis.

   Tip — Do not only celebrate the best score: also look for the team that learned the most about its decision-making process.

7. 7

   Debrief on consensus and behaviours

   10 to 15 min

   The facilitator leads a debrief focused on decision-making mechanisms. They explore how teams handled minority opinions, technical arguments, disagreements, and time pressure. They explicitly distinguish between consensus and voting: consensus aims for an agreement that is sustainable for all, while voting resolves a majority preference. Participants finally relate the experience to their real-life situations in meetings, projects, or crises.

   Tip — Return to observable facts: "Who spoke? Who changed their mind? On what argument?" This avoids an overly abstract debrief.

Variants

• Observer variant: in each team, one person does not participate in the ranking and observes decision-making behaviours, particularly the flow of conversation, handling of disagreements, and moments of transition to consensus.
• Time constraint variant: reduce the team discussion time to reveal voting, authority, or giving up reflexes. Debrief afterwards on the effect of time pressure on listening quality.
• Remote variant: use a shared spreadsheet or collaborative board with a column for each participant, a team column, and a reference column revealed only after the decision. Ask teams to work in breakout rooms and designate a consensus keeper.
• In-depth learning variant: after the first calculation, ask teams to identify the three items where the gap with the reference is greatest, then reconstruct the hypotheses that led to those choices.