12. Sample readout

Client: Ferrovane Components (fictional). Industrial manufacturer. 4,200 employees. Five plants (A–E), headquarters, and a shared-services centre created in 2025. Engagement: Full Census, Organizational Health & Effectiveness pack v3.2 (firm fork), 12 executive interviews, document review. Fieldwork: 3 weeks. 8 languages. Email, QR/kiosk, magic link. Response: 2,562 of 4,200 = 61%. Salaried 1,300 invited → 996 (77%). Frontline 2,900 → 1,566 (54%). Plant C night shift 31%. Anonymity: k = 5, text-k = 10. 11 of 186 units rolled into parents. No suppressed cell recoverable by differencing. Weighting: post-stratified by site × level. Weighted index 56; unweighted 57.4. Difference 1.4 pts (below 2-pt banner threshold). Benchmarks: Firm book — manufacturing, 9 clients, 21,400 respondents: median index 61. External reference shown as indicative only. Legacy 2024 survey imported; 6 items mapped to current practices.

All figures below are fictional and illustrate the format.

12.1 The answer in one page (pre-read, 8-minute partner view)

Health index 56. Second quartile of our manufacturing benchmark (median 61). The company is not short of effort or loyalty. Motivation and work environment score in the top half. It is short of three things: a single set of priorities below the executive layer, supervisors with the time and skill to lead, and ownership of the work the shared-services redesign moved.

Three burning findings

1. Priorities are clear in the boardroom and gone by the third layer.
2. Supervisors are the constraint, not the shop floor.
3. The shared-services redesign moved cost, not work.

One bright spot: Plant A runs a local operating system worth copying.

If we only do three things: one enterprise priority set with kill criteria; a 90-day supervisor operating cadence; named owners for the eight interfaces that now run on tickets.

Ask of the steering committee today: approve the three moves, name owners, agree to re-measure the linked items in a pulse at month 9.

12.2 Burning finding 1 — Priorities are clear in the boardroom and gone by the third layer

Claim. Direction scores 74 for the top two layers, 49 for supervisors, 41 for the frontline. "I understand our strategy" is 71% favourable. "When two priorities conflict, I know which one wins" is 28% favourable.

So what. People can recite the strategy and cannot use it. Every plant is resolving priority conflicts locally, differently, every week. That is the source of the rework described in finding 3.

Evidence

• Metric: Direction by layer, n = 2,562, differences significant after FDR correction, d = 0.71 (top two layers vs frontline).
• Contradiction radar (strength: high): strategy-understanding item vs priority-conflict item; text theme "competing priorities" present in 34% of respondents who wrote comments.
• Documents: three 2025 leadership decks list 11, 14, and 9 "must-win" priorities, with 4 in common.
• Quote: "Every VP has a top priority. We have seven top priorities." — Supervisor, Plant D
• Quote: "The plan changes when a different person visits." — Operator, Plant B
• Clip: 0:48, maintenance planner, Plant E, shareable consent.
• Driver: Priority clarity is the #2 practice by relative importance (17% of explained variance, 95% CI 12–22%).

Confidence: High (quant, text, documents, video converge).

Recommended move (pack DIR-R1): One enterprise priority set of no more than five, each with an owner, a measure, and explicit kill criteria for work that does not serve it. Cascade reviewed in the month-9 pulse.

12.3 Burning finding 2 — Supervisors are the constraint, not the shop floor

Claim. Supervisor-level leadership and accountability practices explain 38% of the explained variance in the health index (95% CI 31–44%) — more than site, function, or tenure. Coaching & development scores 38 (Band 1, Constraint). Supervisors report 44% of their time on administration and firefighting and 12% on coaching.

So what. The frontline is willing (Meaningful work 72, top quartile of the book). The layer that converts strategy into shift-level work has neither the time nor the training to do it. Adding programmes at the frontline will not move the index while this layer is saturated.

Evidence

• Driver: relative weights analysis, n = 2,388 complete cases, bootstrap 1,000×.
• Structure: median supervisor span 1:31 in Plants B–E vs 1:17 in Plant A.
• Time-allocation grid: supervisors (n = 214).
• Theme "promoted without preparation": prevalence 22% among supervisors, 2.1× skew vs overall.
• Quote: "We promote the best welder and never train them to lead." — Supervisor, Plant B
• Quote: "My supervisor spends the shift on spreadsheets." — Operator, Plant C
• Clip: 0:42, team lead, Plant D, shareable consent.
• Interviews: 8 of 12 executives named supervisors as "the gap"; 2 of 12 connected it to span.

Confidence: High. Language note: the driver result is an association. The claim says "explain variance," not "cause."

Recommended moves (pack MGR-B2, ACC-R3): a 90-day supervisor operating cadence (daily 15-minute huddle, weekly 1:1 with a structured agenda, removal of three named administrative reports); supervisor capability sprint for the 214 supervisors; span review in Plants B–E.

12.4 Burning finding 3 — The shared-services redesign moved cost, not work

Claim. Cross-unit handoffs fell from 58 (2024, imported, mapped items) to 41 (2026), a significant decline (d = 0.41). 212 respondents describe handoff failures; the theme is 1.9× over-represented among maintenance planners and plant controllers. Nine of twelve executives described "tickets with no owner."

So what. The 2025 redesign centralised 140 roles into shared services and replaced named contacts with queues. The plants now absorb the coordination cost informally, which is invisible in the business case and visible in morale at the interfaces.

Evidence

• Metric (Wave compare): 6 mapped items, legacy import labeled "imported · mapped items only," hierarchy mapping 94%.
• Heatmap: decline concentrated at plant ↔ Finance and plant ↔ Procurement interfaces; HQ functions unchanged.
• Theme "handoff failure," confidence medium (validation agreement 78%).
• Quote: "I raise a ticket, it bounces three times, then I just call [name redacted]." — Plant controller, Plant C
• Interview excerpt: "We saved the headcount. I am not sure we saved the work." — Executive
• Documents: 2025 shared-services business case projects savings; contains no service-level or ownership design for plant interfaces.

Confidence: Medium-high. Legacy comparison rests on 6 mapped items only; the appendix states this.

Recommended move (pack OPM-R4 → Offering: Interface Redesign Sprint): name an accountable owner for each of the eight highest-volume plant ↔ shared-services interfaces, publish service levels, and restore a named contact for plant escalations. Re-measure handoff items in the month-9 pulse.

12.5 Bright spot — Plant A runs a local operating system worth copying

Plant A leads the company on Motivation (+11 vs rest of org) and Work Environment (+9), both significant. Its supervisor span is 1:17, and supervisors hold a daily huddle that the plant manager introduced in 2024. Innovation & Learning in Plant A is below average: its practices have not travelled because nobody asked. Recommended move: document and pilot Plant A's cadence in Plant C within 60 days.

12.6 Deck outline (firm PPT master, 22 slides)

Slide 17 is deliberate. Showing what the evidence does not support is the fastest way to earn the room's trust in what it does.

12.7 Diagnostic Room run-sheet (45 minutes)

12.8 Action portfolio seeded from the readout

12.9 Firm-only view (never shown to the client)

• Commercial next step: Supervisor capability sprint (Offering: Frontline Leadership, $380k) and Interface Redesign Sprint ($260k). Retainer Pulse, quarterly, from month 3.
• Risks to the relationship: CFO sponsored the 2025 shared-services redesign. Pre-wire finding 3 with the CFO one week before the steering committee.
• Pattern Library candidates: "Strategy understood, not usable" (4th occurrence in the book); "Shared services moved cost, not work" (2nd).