creativity · small groups · offline + online
Worst Idea First.folder
Don't ask how it succeeds — ask how it fails, badly. Then flip every answer. The worst ideas are the easiest to say and the sharpest to act on.
// kit v2: a stage-ready HTML deck (presenter drawer, run-of-show timer, built-in QR codes) · poker-spec print-and-cut cards with backs and blanks · an A3 activity poster · manuals in A4 and US Letter with bookmarks · three license tiers (CC BY free / single event / organization).
Reverse brainstorming, rebuilt as a complete workshop folder: a five-minute disaster round on a real target the group actually cares about, a flip into preventive actions, and a pre/post measure that shows how the room's safety actually moved. The structure gives permission — the game asks for awkward truths, so people bring them.
The exercise runs on a real target and a wall — no projector required. The run sheet carries the pre-mortem, anonymous-shuffle and paired variations, plus the rescues for the three ways this exercise usually goes wrong; an optional poker deck (target, sabotage, the worst, flip, timing) hands each round its own card.
the run sheet
// this is the guide file, word for word — read it here, run from it here. the .md at the bottom of the page is the same text, offline.
Folder:
worst-idea-first· Category: creativity · 3–4 per group · 20 min · offline + online · optional print · assessment-typeLicense: CC BY 4.0
Reverse brainstorming: instead of "how do we make this succeed", the group spends five minutes engineering failure — "how do we make this absolutely flop" — then flips every item. The worst ideas are easier to say, funnier to hear, and flip into the sharpest actions.
Objectives
By the end of the activity, participants will have:
- Said the awkward things — <mark>inside a game that rewards saying them</mark>.
- Watched criticism transform into actionable items with one flip of the frame.
- Practiced the psychological-safety move: naming risks before they happen, not after.
Setup
Space: tables or breakout rooms, one group of 3–4 each. A shared wall or whiteboard for the flip.
Materials: none — sticky notes if the room loves them; the flip works as well spoken. An optional poker deck (target, sabotage, the worst, flip, timing) is in this folder for rooms that want paper in hand.
Grouping rule: groups of 3–4, mixed departments. The worst ideas come from people who see different failure modes.
Pick a target — say it out loud: one real project, product or event the group actually cares about. "This year's customer conference." "The new onboarding flow." Real targets make the flip land.
Minute-by-minute script
| Time | Phase | Facilitator does | Participants do |
|---|---|---|---|
| 0–3 | Frame + target | Name the target. "Question one: how do we make it fail? Badly. Five minutes." | Form groups |
| 3–8 | Worst ideas (5 min) | Float. Push for quantity: "dumber, cheaper to say, faster" | Generate failure plans |
| 8–10 | Pick the worst | "Each group: your single most catastrophic idea. Say it proudly." | Vote, laugh, present |
| 10–16 | The flip (6 min) | "Now flip each item: how do we prevent exactly this?" | Flip worst → actions |
| 16–20 | Bridge | "Each group reads its two sharpest flipped actions — they go to the owner list." | Read out, transition |
Then run the debrief — see worst-idea-first-debrief.md.
Known failure modes & rescues
- The worst ideas are boring. "Nobody shows up." Rescue: demand specifics — "nobody shows up because of what we did? Design that." Sabotage is a creative brief.
- A group flips too early. They skip the fun and start writing preventions at minute four. Rescue: hold the flip — "stay in the disaster a moment longer."
- Someone's worst idea is a real complaint. The laughter dies; it was a genuine grievance in disguise. Rescue: honor it — "that's the most valuable card in the room. Flip it first, and let's talk after."
Variations
- Pre-mortem version: the target is a real project plan; the worst ideas become the project's risk register, verbatim.
- Silent version: worst ideas written on cards, shuffled, read aloud by another group — anonymity for rooms still building safety.
- Online: breakout rooms for the worst round, shared doc for the flip. The shuffle version works well with anonymous forms.
the debrief sheet
// the debrief questions, in full — the .md version prints on one page for the room.
8 questions, 3 failure modes, one rule: the worst ideas are the data.
Use with
worst-idea-first-guide.md. Debrief runs 10 min after the flip.
The 8 questions
1. The easiest bad idea.
"Which worst idea was easiest to say out loud? Why that one?"
→ Easiest usually means safest to say — which maps exactly what the room feels free to criticize.
2. The one nobody wrote.
"What worst idea did you think of but not say? One word is enough."
→ Voluntary, anonymous, no follow-up. The gap between thought and said is the safety meter.
3. The comedy curve.
"Notice how the room got louder and funnier? What changed?"
→ Laughter is the audible sound of safety rising. Naming it teaches the group what it feels like.
4. The flip quality.
"Which worst idea flipped into the sharpest action? What made the flip sharp?"
→ Sharp flips are specific: 'make the invite boring' becomes 'rewrite the invite subject line' — not 'improve communication'.
5. The disguised complaint.
"If a worst idea was a real grievance in disguise — what made it finally sayable?"
→ Only ask if one surfaced. The answer is usually: 'because the game asked for it.'
6. The daily version.
"Where in your week do risks only get named after they happen? What would a five-minute worst-idea round have caught?"
→ Transfer question. Expect: launch post-mortems, incident reviews, 'we knew this would break'.
7. The one habit.
"Which flipped action goes on a real list with a real owner before we leave this room?"
→ Commitment question. One action, one name, one date — the minimum honest output.
8. The next target.
"What is the next project that deserves a worst-idea round — and when?"
→ Closing question. It schedules the exercise into the future instead of ending it.
3 known failure modes of the debrief itself
- The debrief reverts to serious. The fun ends and the room puts its guard back up.
Rescue: keep question 1 playful; the analysis can ride on the laughter.
- The quiet group's flip gets skipped. Louder groups read more actions; quiet ones get cut by time.
Rescue: two sharpest actions per group — cap the loud groups, protect the quiet ones.
- Everything becomes a risk register. The room converts the whole exercise into process and kills the safety lesson.
Rescue: question 2 and 3 re-anchor the debrief on what got said, not what got filed.
If you are running the pre/post measure (worst-idea-first-assessment.json), note these observables:
| Observable | Points to |
|---|---|
| Room volume and laughter rise during the worst round | Safety rising |
| Specific sabotage ideas beat generic ones | Creative engagement |
| Flipped actions carry owners and dates | Commitment quality |
The report your participants receive is a self-comparison, never a group ranking — keep it that way.
who it's for · what it leaves behind
Who is the Worst Idea First activity for?
Teams and their leads ahead of a launch, a reorg or any bet that deserves a pre-mortem — and any room where brainstorming has gone polite and useless. It runs on groups of 3–4 from 8 to 20 people, works online in breakout rooms, and lands hardest when the target is a real project with real stakes.
What results does it produce?
Two or three preventive actions per group that pass the specificity test — auditable by anyone who missed the session — plus a visible rise in what the room feels free to say. A pre/post group-level delta from the assessment preset measures voice, risk-naming and self-censorship. It measures the room's safety, not wit.
Can my AI agent use this folder?
Yes. Every folder ships a SKILL.md skill file: frontmatter for matching, plus runnable prep commands — pull the guide and the debrief with two curl commands. The three-step workflow (discover, install, run) lives on the skills page, every folder ships a one-command zip bundle, and the machine index is skills.json.
Is it really free?
Yes — CC BY 4.0, no account, no email wall inside a folder. Download it, run it, copy it, rebrand it for your own workshop. Attribution keeps everything free.
make it live
The folder's paper pre/post measure is built for rooms with nothing but markers. If you want the digital version — the same 6 items as a live form, each participant's before/after delta computed automatically — two ways:
for AI agents
$ formlm-cli smart generate --input "Pre/post measure for the Worst Idea First reverse-brainstorm: 6 Likert items on voice, risk-naming, specificity and safety to say awkward things — the same 6 items before and after the activity, so each participant's delta is computed automatically"
Step 4 of the agent workflow: npm install -g @formlm/cli, log in once, and the prep above runs itself. The form and its data live in your own workspace.
— or, with a mouse —
Open the free pre/post training quiz template on FormLM
and paste the 6 items from worst-idea-first-assessment.json — the template already runs the before/after comparison and scores it.
// optional, not required — the paper version in this folder is complete on its own. free account on their side.
put it on the wall
The folder ships a single-file HTML deck that runs full-screen from one laptop: press N to open the presenter notes drawer, T to start the round timer, B for big-screen viewer mode, and the closing slide renders its own QR code so the room can scan and grab the deck on their phones. Open worst-idea-first-slide-deck.html in any browser; worst-idea-first-slide-deck.pdf is the same deck for print.
// no companion service, no account — the deck in this folder is the whole projection.
take this folder with you
// every file downloads as a plain file. no account, no email wall inside a folder.