Kit 5 — “Onboarding Academy: 90 days to productive” · TTP-driven curriculum

One-line: a 90-day onboarding curriculum for chilled-goods receivers, architected around one metric (time-to-productivity: 41 → 26 days) and the five failure modes that actually kill onboarding — not around a calendar.

Why this kit exists

A 90-day onboarding program is the most commodified artifact in L&D portfolios — usually a Gantt chart with “buddy assigned” in week 1. This kit is the counter-argument: evidence-driven architecture. Three commitments make it different:

1. TTP as the spine. Every phase names which of the four TTP drivers it attacks (knowledge fluency / environment navigation / confidence / error-recovery). A week that cannot name its driver does not belong in the program.
2. Failure-mode-first design. The program is built around the 5 most common new-hire failure patterns from (fictional) VanDijk exit data (figures such as “error-hiders quit at 3× the rate” are illustrative premises) — quit-in-week-2 dock-shock, certified-but-frozen, silent errors, paperwork drift, post-parity coasting — each mapped to a designed counter-measure and pinned to its risk moment on the timeline.
3. Artifact continuity. Kit 1's certification module and explainer return as named, scheduled components (explainer pinned to d1-before-dock-entry; module attempts scheduled, not ad-hoc), with an explicit “what changed since Kit 1” note. The portfolio reads as one consulting engagement arc.

Artifacts

Design decisions worth naming

• The failure-mode overlay is the kit's signature visual. Toggle it on and the timeline stops being a plan and becomes a diagnosis: every pin is a way this company historically lost a new hire, sitting exactly where the program defends against it.
• Retention is treated as a TTP driver, not an HR afterthought. Every week-2 quit resets a cohort median and burns a hiring cycle — so the growth track (d42–90) is part of the productivity architecture, protected in the rota.
• Gates are go/recovery-plan, never go/fail. The d30 skills gate miss triggers +50% buddy-hours for the cohort — a pre-committed, budgeted intervention, not a blame event.
• The baseline's flaws are published with the baseline. 5 of 19 pre-program starters never reached parity and are excluded from the 41-day median; the honest read (“41 days for survivors, 26% never made it”) ships alongside the headline.

Market mapping

Honest limitations

• The 41-day baseline is a retrospective reconstruction from (fictional) logs; the first live cohort is the real starting gun, and the dashboard says so.
• TTP claims for the first two live cohorts are association, not causation — season, staffing and cohort size confounds are published next to every number.
• L4 (incident reduction, throughput) is explicitly not claimed in year 1; the sample would be too small.
• The interactive blueprint is a design-communication tool for stakeholders; it is not an LMS and does not track real learners — that packaging decision belongs to the implementation phase.

Build notes

Single-session build. Quality gates: node --check on all script blocks, Chromium headless render smoke-test, zero placeholder markers left in any file, every phase + failure pin + checkpoint renders a detail panel. AI-assisted drafting disclosed per artifact; architecture, thresholds and final copy are human-owned.

by I Wayan Usada · runs fully offline · VanDijk Logistics is a fictional client (Kit 1 universe)

STATUS: DELIVERED 2026-08-29 — final artifact published as part of the six-kit portfolio suite (kits 1–6).

Process deliverables

• storyboard.md — Text-based storyboard — the scene plan written before build
• visual-mockups.md — Visual mockups & style guide — mood, type, key screens, component decisions