Retreat Prep: Reuben’s Section — “Bringing It All Together” (Handbook 2.0)

When: Thu Jun 11, 2026 — Team Retreat 2 “The Revenge of the Studio,” closing section of the day Who: Reuben (facilitating), Josh, Joe Time: 45–90 min (90-min run of show below; 45-min compression at the end) Framing: The Advantage (Patrick Lencioni) — the handbook is our playbook: it answers Lencioni’s six critical questions (Why do we exist / How do we behave / What do we do / How will we succeed / What’s most important now / Who must do what). Today we re-answer them with a year of evidence instead of opinions. Best Idea Wins: show your work.

Energy note: This is the last section of an all-day offsite. Everything is cards, walls, and timeboxes — no open-ended discussion blocks.

---

Run of Show (90 min)

0. Frame (5 min)

One line: “A year ago we wrote down who we’d be. Today we check it against who we were. Every claim needs a receipt.”

1. The Cache — “I can help with…” (25 min)

Build a capability deck from the year.

• Round 1 — Self (5 min, silent): Each person writes index cards: “I can help with ___” on the front, one real moment from the past year on the back. No receipt, no card.
• Round 2 — Each other (5 min, silent): Write cards FOR your two teammates: “Josh, you can help with ___” + the moment you saw it. (This is the good stuff — people undervalue their own capabilities. Perpetual Trust Fall in action.)
• Share & swap (7 min): Read the cards you wrote for others out loud, hand them over. Quick reactions only.
• Sort (8 min): Tape three columns on the wall — Design / Develop / Disseminate (the handbook’s WORK cycle). Sort the whole deck. Then stand back:
  • Where are the pile-ups? Where are the gaps?
  • Does the handbook’s “What do we do” section describe this wall, or a different team?

Artifact: The deck survives the retreat — it becomes a “menu of services” doc that feeds messaging (The Words theme), the role-play in step 5, and the decision-framework question “do we have unique value to add?”

DONE (Jun 11): Deck captured and interpreted → reference/menu-of-services.md (cleaned deck + proposed ROLES rewrite) and reference/capability-map.html (visualization). Raw CSV in Downloads.

2. Values Audit — Receipts, Taxes, Drift (20 min)

Lencioni’s test: a real core value occasionally costs you something. If it never has, it’s decoration (or “permission-to-play”), not core.

For each of the four values (Serious Play, Perpetual Trust Fall, Best Idea Wins, Cards on the Table), capture on stickies:

• Receipt — a concrete moment someone lived it. Rule: tell a story about a teammate, not yourself.
• Tax — a moment the value cost us something (time, a deal, comfort) and we paid anyway.
• Drift — a moment we violated it or quietly skipped it.

~4 min per value, then 4 min to decide per value: keep / sharpen the wording / merge. A value with receipts but no tax gets flagged for honest discussion. This directly produces the Values section amendments Reuben wanted for handbook 2.0.

3. Codebook Tag Triage — the “Confident by July” checkpoint (10 min)

Framing: In March we said we wanted to be “dead confident in the data by July” (see prep/2026-04-08-tag-analysis-plan.md, presented at quarterly planning). Phases 1–2 of that plan actually happened — the May–June retag cleaned the unofficial tags (ADVISE 33→2, BOOST 14→0) and woke up dead ones (EQUITY-WIN 0→6, SPARK 0→2). This retreat IS the Phase 3 checkpoint. Print the fresh Airtable table (below) and decide:

• Two bases, two opposite stories — lead with this. AC:DE Macro layer: STUDIO-BUILD on 70% of records, CHECK-IN nearly absent (3). Project Support Tracking (seed grants): CHECK-IN is the #1 tag (105), SOUNDING-BOARD healthy (27), STUDIO-BUILD only 11. April’s worry — “Engagement is our weakest category… we do this work, we’re just not tagging it” — was a one-base artifact. We WERE tagging it, in the other base. Combined, the portfolio looks like what we actually are: a build shop at the project layer and a relationship shop at the touchpoint layer. Celebrate, then ask: should the handbook’s OKR categories be read across both bases?
• Workhorses (combined): STUDIO-BUILD (133), CHECK-IN (108), BRIDGE-BUILD (99), CO-CREATE (88), TOOLBOX (72). The April question still stands at the macro layer: is STUDIO-BUILD at 70% of AC:DE records accurate mix or reflexive default?
• True ghosts (combined, both bases): SYSTEM-SHIFT (0, ever, anywhere) and BOOST (0). BREAKTHROUGH has only 4 uses (all Learning Through Creation) — and the codebook predicted “[BREAKTHROUGH] often follows [VULNERABLE-SHARE],” a sequence never once recorded. EQUITY-WIN (6, all macro layer): tagging problem or portfolio finding?
• Squatters still alive in the seed base: TRUST-MARKER (3) and CONFIDENCE-SHIFT (1) — the two April unofficial tags that never got resolved. Promote or fold.
• Hygiene list for handbook 2.0 (cheap wins, decide in 2 min): THOUGHT-PARTNER is spelled two ways across tables ([THOUGHT PARTNER] vs [THOUGHT-PARTNER]) which silently breaks counting; some records carry 7–11 stacked tags (one SofIA check-in has 11) vs. the one-best-tag discipline; 17 untagged records in the seed base; tag field is named differently per table.
• Goal 1 is now measurable: “75%+ seed grantees with 3+ tags” — the Project Support base is where that lives. 225 records, only 17 untagged. Worth a live look if there’s energy.

Bonus calibration beat (2 min, feeds Joe’s “how we track data” section): Reuben’s notes system, the AC:DE base, and the seed-grant base each tell a different story about the same year (e.g., RIPPLE: 29 in notes, 11 in AC:DE, 2 in seed base). Which source do we report from, and how do we reconcile them?

4. Amendment Sprint (15 min, heads-down)

Each person writes actual replacement text, not notes-about-edits. Suggested split:

• Reuben: Values section rewrite (from step 2 output) — wording + a “receipts” example under each value
• Josh: Priorities / decision framework — does it match how decisions actually got made this year?
• Joe: The Work (Design/Develop/Disseminate) — recalibrated against the cache wall, + codebook tag changes from step 3

5. Stakeholder Role-Play (10 min)

Stakeholder cards face-down — each person draws one: Isabelle / brand-new seed grantee / skeptical faculty member / external funder / GSE dean. 90 seconds: “what the Studio does and what’s changing this year,” using cache-deck language, no handbook jargon. Listeners write down every word or phrase that wouldn’t land with that stakeholder. That list feeds The Words theme directly.

6. Ratify (5 min)

Fist-to-five per amendment. Anything under a 3, the holder speaks; then disagree-and-commit (Cards on the Table). Assign one owner to merge everything into handbook 2.0 and a date. Lencioni’s point: clarity dies without overcommunication — also decide where the handbook shows up in the weekly ritual so 2.0 doesn’t become shelfware.

---

45-Minute Compression

• Frame (3)
• Cache rounds 1+2 + quick sort (15)
• Values: receipts + tax only, skip drift (12)
• Tag triage with the printed table (8)
• One spoken amendment each + fist-to-five (7)
• Amendment writing becomes 48-hour async homework; ratify over Slack or at next weekly.

Materials

• Index cards in two colors (self vs. for-others), sharpies
• Painter’s tape for the Design/Develop/Disseminate wall
• Printed tag-frequency table (1 per person)
• Printed current handbook (1 per person, for marking up)
• 5 stakeholder cards, a visible timer

Tag Frequency Data (pulled live from Airtable, Jun 10–11, 2026)

Combined view — both bases (AC:DE Macro Manager Layer 173 records + Project Support Tracking 225 records)

Tag	AC:DE	Seed	Total		Tag	AC:DE	Seed	Total
STUDIO-BUILD	122	11	133		ANCHOR	5	14	19
CHECK-IN	3	105	108		REFRAME	4	11	15
BRIDGE-BUILD	45	54	99		SPARK	2	11	13
CO-CREATE	76	12	88		RIPPLE	11	2	13
TOOLBOX	29	43	72		VULNERABLE-SHARE	4	8	12
THOUGHT-PARTNER	31	21	52		SCALE-MOMENT	11	1	12
VALUE-VISIBLE	41	11	52		EQUITY-WIN	6	0	6
NUGGET	12	26	38		BREAKTHROUGH	0	4	4
MULTIPLIER	34	2	36		ADVISE	2	1	3
LEVEL-UP	16	19	35		TRUST-MARKER*	0	3	3
SOUNDING-BOARD	7	27	34		CONFIDENCE-SHIFT*	0	1	1
ALIGN	12	20	32		SYSTEM-SHIFT / BOOST	0	0	0

* = not in the handbook codebook. Seed base = sum across all 5 tables (Learning Through Creation 113, Joyful Learning 67, Tequity 23, PWHOPL 15, Other Projects 7; 17 untagged). Seed-base recent window (Apr 1–Jun 11, 45 records): CHECK-IN 16, BRIDGE-BUILD 15, NUGGET 10, TOOLBOX 10 lead; PWHOPL nearly dormant (1 record since Apr 1); recent activity skews Tequity (17/45).

AC:DE Macro Manager Layer detail (173 records)

Tag	All-time	Apr–Jun ‘26		Tag	All-time	Apr–Jun ‘26
STUDIO-BUILD	122	29		NUGGET	12	5
CO-CREATE	76	18		RIPPLE	11	0
BRIDGE-BUILD	45	13		SCALE-MOMENT	11	4
VALUE-VISIBLE	41	11		SOUNDING-BOARD	7	0
MULTIPLIER	34	10		EQUITY-WIN	6	6
THOUGHT-PARTNER	31	12		ANCHOR	5	1
TOOLBOX	29	7		VULNERABLE-SHARE	4	0
LEVEL-UP	16	4		REFRAME	4	0
ALIGN	12	5		CHECK-IN	3	2
				ADVISE	2	2
				SPARK	2	1
				BREAKTHROUGH	0	0
				SYSTEM-SHIFT	0	0

Untagged records: 5 all-time, 0 in Apr–Jun (tagging discipline improved). Per-lead tag volume: Joe 339, Reuben 135, Josh 61 (co-led records double-count). Note: Month field uses “Sept 2025” spelling, and 10 records sit in an “Ongoing” bucket outside the monthly timeline.

April 7 baseline for comparison (prep/2026-04-08-tag-analysis-plan.md, 140 records): STUDIO-BUILD 98, CO-CREATE 64, ADVISE 33, BOOST 14, SPARK/EQUITY-WIN/SYSTEM-SHIFT 0, 7 untagged. The May–June retag is visible in the deltas.