communication · pairs · offline + online
Draw What I Mean.folder
Describe it, draw it, then compare — three rounds that make instruction drift visible. Twelve minutes, two people back to back, and the gap between what was meant and what was drawn.
// tool pack: poker-spec print-and-cut cards with backs and blanks (picture cards, a shared word card, a blank to author your own) · an A3 activity poster · the run sheet and debrief bound in A4 and US Letter with bookmarks · paper and pen, no projector, no account.
A back-to-back describe-and-draw tool, built as a complete workshop folder: a timed run sheet, printable picture and shape-word cards, and three rounds — free words, a shared shape-word card, open questions — that show a shared vocabulary and the chance to ask beat a confident description.
The props are poker-spec print cards (63.5 × 88.9 mm, bleed and cut guides included) — picture cards, a shared shape-word card, blanks to author your own. Nothing needs a projector, and nothing needs an account to download.
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:
draw-what-i-mean· Category: communication · pairs (2 people) · 12 min · offline + online · print props · tool packLicense: CC BY 4.0
Back to back, one partner describes a small composed picture — three to five simple shapes in a fixed arrangement — and the other draws it from the words alone. Then you compare. Three rounds, one variable changed each time: ordinary words, then a shared word list, then the right to ask questions. The gap between what was meant and what got drawn is the lesson, and the round-by-round comparison tells you which repair actually closed it.
How this differs from blind-portrait: Blind Portrait puts one description through a wall of silence and shows you the damage. This folder isolates the causes — vocabulary first, then feedback — so the team leaves knowing which repair worked, not just that communication failed. Run Blind Portrait to open a session on instructions; run this one when the team already agrees something is leaking and needs to know what to fix first.
Objectives
By the end of the activity, participants will have:
- Seen on paper how far ordinary words drift — size, position, count, missing element.
- Felt a shared word list close most of that gap with no extra effort from either partner.
- Used a clarifying question from the receiving end, and measured what it recovered.
When to use this activity
- A team routinely builds from spoken or half-written instructions and keeps shipping the wrong thing.
- You want a fast, hands-on demo of why a shared vocabulary and clarifying questions beat "just explain it better".
- You have 12 minutes and can seat pairs back to back; no drawing skill needed — they are line shapes, not portraits.
Setup
Clock: twelve minutes for the three rounds and the bridge. Give it twenty if the room debriefs properly — the comparisons are the material.
Space: two chairs per pair, back to back, no table shared between them. The describer holds the picture card facing themselves; the drawer holds a pen and the sketch sheets.
Materials per pair:
| Item | Source |
|---|---|
| 3 × picture card (Round 1 / 2 / 3) | draw-what-i-mean-prop-cards.pdf — one per round, same difficulty |
| 1 × shared shape-word card | draw-what-i-mean-prop-cards.pdf — both partners may read it in round 2 |
| 1 × rule card — back to back, words only | draw-what-i-mean-prop-cards.pdf, or read it aloud |
| 3 × blank sketch sheet | any plain A4; the sheet template is on the print page of the same PDF |
| 1 × pen per drawer | marker or ballpoint, not pencil — a confident wrong line is the evidence |
| 1 × blank card per pair | draw-what-i-mean-prop-cards.pdf — for authoring a picture in the team's own terms |
Grouping rule: pair people who actually receive briefs from each other. If everyone already shares a tight domain vocabulary, pair across teams instead — the round-2 effect is bigger when the shorthand is not already there.
Roles: describer (may not show, point or draw), drawer (may not peek; in round 3 may ask anything), and you on the clock.
How to run it
| Time | Phase | Facilitator does | Participants do |
|---|---|---|---|
| 0–1 | Set up | Seat pairs back to back. Hand the drawer three sketch sheets, the describer three picture cards face down. | Take cards; nobody sees the other's material |
| 1–4 | Round 1 — free | "Describer: say it however you like. Drawer: words only, no questions." | Describe, draw, resist the urge to look |
| 4–5 | Compare 1 | "Card up, beside the drawing. Note what drifted." | Mark the drift: size, position, count, missing element |
| 5–8 | Round 2 — word list | Hand each pair the shared shape-word card. New picture. "These words are now legal — and they are the only new thing." | Both use the named shapes and position words; draw; compare |
| 8–11 | Round 3 — questions | "Drawer: you may ask anything. Describer: you may still not show it." | Drawer interrogates; describer answers; compare |
| 11–12 | Bridge | One line each: which round closed the most gap, and what your work calls that. | Say it out loud; keep the three sheets |
Then run the debrief — see draw-what-i-mean-debrief.md.
The card
Each card shows three to five simple shapes in a fixed arrangement — a circle above a tall rectangle, two triangles flanking a square, a line under everything. Nothing needs drawing skill: the difficulty lives entirely in the words. The three cards are deliberately similar in difficulty so the only thing changing between rounds is the rule. The shape-word card names shapes, positions, sizes and counts, which is what turns "a thing over there, kind of big" into "tall box, round one on top-left, two pointy ones either side."
Failure modes
- The describer starts drawing for the partner — in the air, on the wall, on their neighbour's paper. Rescue: back to back is not staging, it is the instrument. Hold it.
- Round 3 becomes a joint rebuild. Rescue: questions are allowed, showing is not. The drawer still holds the only pen.
- It reads as a kids' game. Rescue: land the bridge on a real artifact — "put your round-1 drawing next to the ticket you misread last month."
- A pair finishes in ninety seconds. Rescue: hand them the blank card to author a fourth picture in their own domain terms; that is the hardest version of this exercise.
Variations
- Online: one camera on the describer's card only, the drawer on a shared whiteboard. Keep back-to-back in spirit — no screen share of the picture, no questions until round 3.
- Large groups (20+): every pair runs at once; at the end, tape the three rounds of every pair to the wall as a gallery and walk it. The pattern across the wall is the debrief.
- Domain reskin: author the blank cards with shapes from your own work — a screen, a service, a data store, a dependency arrow. Round 2's word card then becomes your team's real glossary, and the exercise turns into a test of whether that glossary exists.
- Long version (25 min): add a fourth round where the pair may only use the shared word list and no other nouns, then debrief on which words went missing from the list.
- No-print fallback: describe a room the drawer has never seen — same three rounds, no cards at all. Weaker evidence, same argument.
Facilitation notes
- The strongest moment is comparing round 1 to round 2 — same describer, same effort, only the shared word list changed.
- Say up front that nobody is graded on drawing; these are line shapes, so the shy drawer still plays.
Put it on the wall
Print draw-what-i-mean-activity-poster.pdf (A3) to put the run of show, the three rounds and the bridge question on one wall while you facilitate.
Props in this folder
draw-what-i-mean-prop-cards.pdf— poker-spec print sheet (63.5 × 88.9 mm, bleed and cut guides): three picture cards, shared shape-word card, rule card, bridge card, author-your-own blank, branded backsdraw-what-i-mean-guide.md— this run sheetdraw-what-i-mean-debrief.md— six debrief questions and the failure modes of the debriefdraw-what-i-mean-run-sheet.pdf— the print edition: this run sheet + the debrief, bound as one A4 document with bookmarksdraw-what-i-mean-activity-poster.pdf— the whole tool on one A3 wall sheet: run of show, the three rounds, the bridge, watch-outsdraw-what-i-mean-start-here.md— one page, thirty seconds, then print and godraw-what-i-mean-license.md— what you may do with all of thisSKILL.md— machine-readable manifest for AI agents*-Letter.pdf— US Letter editions of the print documents
Agent version
Every file in this folder has a stable public URL — an AI agent can fetch the cards, the debrief and this run sheet with three curl commands (see SKILL.md). The paper version needs nothing but two chairs, three cards and a pen.
the debrief sheet
// the debrief questions, in full — the .md version prints on one page for the room.
Six questions, three failure modes of the debrief itself, one rule: all three drawings stay on the table, in order, until the last question is done.
Use with
draw-what-i-mean-guide.md. The bridge line at minute 12 is the compressed version; this sheet is the full one.
How to run this debrief
- Lay the three pairs out in order — card beside drawing, one row per round. The argument of this whole tool is a picture, and it only works if the row is visible.
- Ask to the drawer first. The receiving end has the information about what was missing; the describer has already forgiven their own words.
- Resist the compliment. "You described it well" closes the finding. Point at the paper: "where did the circle end up?"
Read the three comparisons
- Round 1 — ordinary words. The drift is usually size, then position, then a missing element. Nobody did anything wrong; the vocabulary was simply private.
- Round 2 — the shared word card. Same describer, same effort, only the words changed. This jump is the single most useful piece of evidence in the folder.
- Round 3 — questions allowed. Watch who asked and what they asked about. The questions that were not asked are as informative as the ones that were.
The six debrief questions
1. The biggest gap.
"Which round closed the biggest gap — better words, the shared card, or the chance to ask?"
→ Targets: which repair is worth investing in. Push for one answer; "all three" means the pair enjoyed the game and missed the experiment.
2. The word card in your real work.
"What is your word card at work — the list of names your team already shares? What is missing from it?"
→ Targets: glossary awareness. Expect either nothing written down, or a document nobody opens. Both are the finding.
3. The swallowed question.
"Drawer: which question did you want to ask in round 1 and not ask? What stopped you?"
→ Targets: the cost of silence at the receiving end — not wanting to look slow, not wanting to interrupt, running out of time.
4. Same person, different result.
"Describer: you tried exactly as hard in round 1 as in round 2. What does that do to the advice 'just explain it better'?"
→ Targets: the strongest single indicator here. Effort was constant; only the shared vocabulary changed. If the pair cannot see that, run the two cards side by side again.
5. The drift signature.
"Where did your drawings drift most — size, position, count, or a missing element? Where does your work drift the same way?"
→ Targets: the team's specific failure signature. Counts and missing elements are the classic silent losses: nobody notices the fourth shape was never drawn.
6. The one-word fix.
"Name one word your team should define before the next build — one noun currently doing two jobs."
→ Targets: commitment in singular form. One agreed definition ships; a glossary project does not.
Failure modes of the debrief
| Failure mode | Why it happens | The rescue |
|---|---|---|
| It becomes an art critique | The drawings are on the table and people are funny about them | Every question must name a word, not a line: "which noun was missing here?" |
| The pair agrees the shared card did it | Round 2 felt easier, and easy feels like the answer | Ask round 3's question next: which single question recovered the most? |
| The lesson stops at the paper | The room had fun; Monday looks identical | Close on question 6 and write the word on the wall poster's blank line. |
who it's for · what it leaves behind
Who is Draw What I Mean for?
Facilitators and team leads who keep getting work built from vague or half-specified briefs. It runs in pairs back to back — any even number of people from 4 up, several pairs at once — and needs no art skill, only line shapes and words.
What does it produce?
Three side-by-side comparisons — the picture card against the drawing, for free words, a shared word card, and open questions. The visible jump between round 1 and round 2 is the value of a shared vocabulary; the round-3 cleanup is the value of asking.
Can my AI agent use this folder?
Yes. Every folder ships a SKILL.md skill file: frontmatter for matching plus runnable prep commands — fetch the cards, pull the debrief. 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 yours
This tool is paper and a pen — no account, no projector. The one file you will actually change is the shared shape-word card: replace the generic shape and position words with your team's real glossary, reprint at poker spec, and the whole round-2 effect carries your vocabulary. The white-label guide in the folder walks through it.
Files in this folder
All files are free under CC BY 4.0. Attribution: “Draw What I Mean · WorkshopSkill · CC BY 4.0.”