The Facilitator's Stack: A Checklist From Invitation to Follow-Up

Series: How to Run a Workshop People Remember (10/10)
Nine posts of opinions, mechanics and tool links later, here's the whole series compressed into the artifact people actually asked me for: the runbook. Print it, paste it into Notion, ignore half of it — the checklist is the point, not the compliance.
T-minus 2 weeks: design
- [ ] Write the outcome sentence: "By the end, participants can ___" (part 1)
- [ ] Write the evidence sentence: "I'll know because ___"
- [ ] Draft the four-beat arc with minutes: Open → Explore → Converge → Close
- [ ] Pick activities from folders with run sheets, props and debrief scripts — WorkshopSkill, or file your own materials in the folder shape (part 2)
- [ ] Pressure-test the plan with ChatGPT or Claude; keep the one-page plan in Notion
T-minus 1 week: prep
- [ ] Print props (poker-spec cards, A3 posters, both manual sizes)
- [ ] Create the sign-up card and check-in card — QRCard AI, printable, live status after printing (part 3)
- [ ] Send the invite with the card in it; if tickets are paid, the money lives in Eventbrite, the card handles the day
- [ ] Build the pre/post measure — FormLM, from the pre/post-training quiz template, one description and the scoring is generated (part 6)
- [ ] Optional, if you do this monthly: let an agent do the above from the folder's SKILL.md via skills.json, and build the forms over FormLM's MCP server — npm package @formlm/cli,
npx -y @formlm/cli mcp(part 7)
Day before: the room
- [ ] Pre-build the remote/hybrid board if needed — Miro or FigJam, frames per group, never a blank canvas (part 5)
- [ ] Wi-Fi card on each table — free in-browser generator, nothing uploaded
- [ ] Charge everything; pack the clicker; check the projector with the presenter deck
- [ ] Deal tomorrow's groups in advance — random group generator, cross-matching if the goal is introductions
Day of: run it
- [ ] Pre-measure open as people arrive (under 10 minutes, scored)
- [ ] Open with pairs, not the big screen — Two Truths and a Lie, two minutes (part 4)
- [ ] One live-screen moment, genuinely used: trainer live screen — counts, not names (part 4)
- [ ] Every group task = artifact + visible countdown + midpoint callout (part 5)
- [ ] Debrief scripted, not improvised — What? So what? Now what? Exit ticket at the door; anonymous wall for the unsayable (live message wall) (part 8)
- [ ] Close with one written commitment per person
- [ ] Post-measure before people leave; feedback card on the table
Within 48 hours: follow-up
- [ ] One email: commitments, materials, what changes next time (part 8)
- [ ] Structured feedback survey out — workshop feedback form template
- [ ] Ten minutes of margin notes on the run sheet: what ran over, what got cut, which debrief question produced silence
- [ ] If a client sponsored this: the per-respondent knowledge-lift report forwards itself (part 6)
Monthly: keep the program alive
- [ ] Same date every month: the client report — attendance and deltas at the top, what the group worked on, one chart, one note (ReportSkills; the trainers' section and the monthly client report guide define the shape) (part 9)
- [ ] The templates hub has the starting points; paste the tables you already have, send per-recipient on a schedule:
The whole stack, every URL in one place
The tools from my own shop (I'm one of the makers — full disclosure below):
| Tool | The job | URL |
|---|---|---|
| WorkshopSkill | Ready-to-run activity folders: run sheet, deck, props, debrief script, pre/post measure; SKILL.md for agents | workshopskill.com |
| QRCard AI | Printed QR cards: sign-ups, check-ins, feedback, live screen, message wall, group generator | qrcard-ai.com |
| FormLM | Scored forms and assessments; per-respondent PDF reports; open-source MCP server | formlm.me |
| ReportSkills | Recurring branded client reports from pasted tables; scheduled, per-recipient delivery | reportskills.com |
The other people's tools this series leans on:
| Tool | The job | URL |
|---|---|---|
| ChatGPT / Claude | Outcome sharpening, drafting, arc pressure-tests | chatgpt.com / claude.ai |
| Notion | The one-page plan | notion.so |
| Google Slides / Canva | Teaching-block slides, one-pagers | slides.google.com / canva.com |
| Eventbrite | Paid ticketing, refunds, receipts | eventbrite.com |
| Calendly / Cal.com | Scheduling around the session | calendly.com / cal.com |
| Mentimeter / Slido / AhaSlides | Heavyweight polling when you need the features | mentimeter.com / slido.com / ahaslides.com |
| Miro / FigJam | The remote room's shared artifact | miro.com / figjam.com |
| Zoom | Breakout rooms for remote programs | zoom.us |
| Google Forms | Free-and-manual measurement when that's enough | forms.google.com |
| Otter.ai | Debrief transcripts, with consent | otter.ai |
| Looker Studio | The dashboards world, when your data lives there | lookerstudio.google.com |
The sentence to keep
If the whole series had to compress to one line, it's this: a workshop is not a deck; it's a folder, a measure and a follow-through. Every tool in the tables above just makes one of those three cheaper. A folder of markers, a show of hands, and a follow-up email you actually send beats a premium stack with a missing debrief every single time.
That's the series. If you run workshops, I'd genuinely like to hear which part of the pipeline eats your evenings — the fastest way to reach me is through any of the contact pages above.
Every part of this series: 1 — Design before the slides · 2 — The activity folder · 3 — Registration and check-in · 4 — Opening the room · 5 — Group work · 6 — Pre/post measures · 7 — Agent-assisted prep · 8 — The debrief · 9 — Client reporting.
*Disclosure: I'm one of the makers — the tools from my own shop named across this series. *