Creative decision mechanism workshop / 6 MIN READ
Check a creative vote for agenda-dependent outcomes
The order of comparisons can change the selected route without changing anyone’s preferences.
AI-assisted research and analysis for the SNS Creative Fieldbook
Sources checked 2026-11-07
A creative review needs an honest account of how the recommendation was produced, especially when a quick show of hands is presented as a single shared preference.
What you will leave with: A complete fictional comparison record and a revised creative-decision brief.
Use this when
- A team chooses among more than two creative routes through successive votes.
- A review summary describes the final survivor as the option everyone preferred.
Not an election forecast, a real vote or a claim about any participant’s motives.
Keep the routes and the decision question fixed
A fictional museum is choosing a poster direction for an exhibition. Route A leads with an object photograph, B uses a typographic statement and C uses an illustrated scene. Each is a complete concept rather than a mixture of the strongest features from several presentations. The review concerns which direction to develop, not whether factual exhibition details may be omitted.
Nine fictional reviewers rank three creative routes. Four rank A>B>C, three B>C>A, and two C>A>B. Rankings are complete, strict and unchanged throughout this exercise; every reviewer votes for the preferred route in each comparison.
Prepare comparable concept sheets before discussing preferences. Use the same exhibition information and intended placement for each route. If one concept arrives as a finished animation while another is a rough sketch, record that presentation difference rather than treating every reaction as a clean assessment of the underlying idea. Keep factual corrections separate from aesthetic preference.
Ask the facilitator to state what the meeting can decide. A recommendation for development does not certify an image licence, approve a factual claim or establish production feasibility. Those questions should not disappear merely because participants prefer one visual direction. The fictional vote gives the team a specific review problem to inspect.
Read all comparisons, not only the last one
A beats B by 6–3; B beats C by 7–2; C beats A by 5–4. This majority cycle has no route that defeats both others. Individually consistent rankings can therefore produce a cyclic group comparison.
Put the full comparison record next to the recommendation. If the slide shows only the final contest, a reader cannot see what was discarded earlier. Preserve the route labels and the original concept versions so the summary cannot accidentally combine a response to an old draft with a later redesigned option.
Invite the group to explain the design reasons behind its reactions without demanding that everyone adopt the same taste. A participant may value immediate recognition, another may value readability, and another may value an unexpected visual story. Those explanations can inform the next design iteration. They are not measurements of actual visitors’ attention or evidence of future ticket sales.
Change the agenda as a diagnostic, not a tactic
Consider sequential elimination: compare two routes, then compare their winner with the third. Starting A/B selects C finally; starting B/C selects A; starting C/A selects B. The preferences never changed, only the agenda.
This demonstrates agenda dependence, not that voting is always defective or every review has a cycle. It does not identify a universally superior replacement rule.
The exercise asks the facilitator to expose a process choice, not to arrange a desired result. Before a real review, describe the decision procedure and how unresolved disagreements will be handled. If the announced procedure proves unsuitable, stop and explain the proposed revision openly rather than silently changing the comparison order after seeing which concept is ahead.
Do not accuse a facilitator of manipulation merely because a chosen sequence mattered. The person may have used an ordinary presentation order without recognizing its effect. Describe the observed process and the limits of the resulting claim. A useful correction can be specific and non-accusatory: the recommendation followed the announced sequence, while the complete responses remain available for interpretation.
Turn the discussion into a defensible design brief
Write the next brief around the exhibition’s actual communication need. Identify the intended viewer, the essential information, the display context and the design question worth resolving next. A small, focused revision may answer a shared concern better than repeatedly asking the room to choose a favorite. Keep the reason for that revision separate from the numerical comparison.
When the responsible owner selects a direction, record the rationale and remaining work plainly. The owner might choose a concept for a documented practical reason even though the discussion remains divided. Do not describe that accountable choice as unanimous approval or hide the disagreement to make the meeting appear more successful.
Close the exercise by rewriting the review summary. It should say which concept is moving forward, who made that decision, what evidence informed it and what still requires validation. Preserve the fictional response record as a teaching aid. No real ballots were collected, no campaign was released and no conclusion about visitor behavior follows from this example.
Make the reasoning concrete.
A museum’s review summary claims that the final surviving poster route is the group’s unambiguous favorite. The facilitator has shown only the last contest and omitted the earlier comparisons.
Publish the reassuring summary
Opening: Describe the surviving concept as a complete consensus.
Hypothesis: The design team receives a simple instruction.
Risk: The record overstates what the review established.
Explain the decision accurately
Opening: Retain the full comparison and distinguish the owner’s choice from the group’s discussion.
Hypothesis: The design team understands both the direction and unresolved concerns.
Risk: The next brief still needs specific communication requirements.
The fictional team revises the summary and prepares a concrete design question rather than pretending the discussion produced unanimity.
Take the questions into your next brief.
Use these prompts in your own notes. Nothing is submitted from this page.
Question
What can this creative review actually decide?
Comparability
Which differences between concept presentations could affect the discussion?
Record
What is missing when only the last contest is shown?
Procedure
What process will be explained before participants respond?
Brief
What specific design question should the next iteration address?
Watch for the wrong conclusion.
- Converting a preferred concept into evidence of visitor demand.
- Replacing an incomplete decision record with a claim of consensus.
- Changing a review procedure without explaining the change.
Your next action
Reconstruct the fictional comparison and rewrite the review summary so it states the actual decision and remaining design work.
Sources & scope
All business examples and numerical fixtures are original and fictional. Sources support the identified mathematical or implementation distinctions. Qualified analysts own model specification and inference. No causal result, model fit, actual customer data or operational change is claimed.
- Stanford Encyclopedia of Philosophy: Voting Methods →Scholarly reference · Supports: Supports majority cycles.
- Easley and Kleinberg: Voting, chapter 23 →University-hosted textbook chapter · Supports: Supports sequential agenda dependence.