onboarding-academy · deliverable

Onboarding Academy — Stakeholder Pack (four one-page contracts)

Four roles make or break this program. Each contract below is one page, ends with “what good looks like by day X”, and is meant to be signed — a contract, not a memo.

1. New hire — “What my first 90 days look like”

You are not expected to be good on day 1. You are expected to be loud about what you don't know.

Days	What happens to you	What we ask of you
1–5	Explainer before dock entry, buddy full-time, first physical win on day 3	Ask anything. Silence is the only wrong behavior in week 1.
6–20	Full flows with buddy nearby; 3 borderline cases/week rehearsed	Log every error — the recovery ritual protects you, it does not mark you.
21–35	Solo with a radio; certification by d28; d30 skills gate	Use the call-the-lead script card. Asking is scripted, not shameful.
36–41	Net comes off; parity confirmed from data	Run the error loop unaided: flag → log → fix → debrief.
42–90	Cross-dock rotation, masterclass, mentor candidacy, day-90 review	Tell us at d90 whether this job still has a future for you — honestly.

Your protections: no-blame error ritual from day 6 · your buddy is the same shift pattern as you · no gate is used against you without a recovery plan first.

What good looks like by day 35: you handle a chilled load solo, your error rate sits inside the veteran band, and the last time you hid a mistake was never.

2. Buddy — “20 minutes a day, heavy in weeks 1–3”

You are the program's nervous system. You see the drop-risk signals before any dashboard does.

Time budget: full-time shadowing d1–5 (this is scheduled, not stolen from your quota), then ~20 min/day d6–20 (debriefs, scan-log review), then on-radio availability d21–35. Released at parity.
Week 1–3 heavy lifting: run the day-3 first-win task; rehearse borderline cases from the card deck; log dock-shock signals (silence at huddle, late breaks, scan-gun avoidance).
You are not the assessor. Gate decisions belong to the shift lead + L&D. Your job is coaching and signal-logging, never scoring.
Your own protection: buddy duty counts toward your performance review as a named contribution; you get the masterclass seat first when you rotate off.

What good looks like by day 20: your new hire runs a full flow with ≤1 prompted intervention, and you have filed at least one drop-risk signal (or an explicit all-clear) — either is a success, silence is not.

3. Shift lead — “Gate decisions + coaching moments”

You own the two moments this program lives or dies on: the d30 gate and the no-blame ritual.

Week 1: publicly log one of your own errors in the huddle. The recovery ritual is believed only if leadership goes first. This is non-negotiable and takes 3 minutes.
Weekly: one coached flow per new hire from d10; weekly scan-log vs. error-log audit (catch silence, not people).
d20 flow audit + d30 skills gate: you score jointly with L&D. Gates are pass/recovery-plan, never pass/fail-and-forget.
Parity sign-off (d36–41): two independent confirmations from dashboard data. No vibes-based parity — if the dashboard says no, the answer is no.
Escalation: cohort gate pass-rate <80% → you and L&D trigger the intervention (buddy-hours +50%), you do not absorb it silently.

What good looks like by day 30: every new hire in your shift has a dated gate decision, your own logged error is still on the huddle board as proof of the ritual, and nobody on your shift has gone quiet on a mistake.

4. L&D owner — “Dashboard, interventions, and the honest numbers”

You own the metric, the instruments, and the discipline of not massaging either.

Instruments you maintain: d30 skills gate rubric · solo error-rate dashboard (daily, vs. veteran band ≤2.1/100 lines) · cohort dashboard (TTP median, gate pass-rates, retention).
Intervention triggers you act on, in writing: d30 cohort pass <80% → buddy-hours +50% for that cohort · any individual >5 days off the parity ceiling → named recovery plan with owner + date · week-2 quit → exit interview within 5 working days, findings to the next program review.
Attribution discipline: TTP is reported with its confounds (cohort size, season, dock staffing). You publish the honest number even when it misses target — the dashboard's credibility is the program's credibility.
Cadence: weekly cohort pulse (15 min with shift leads) · d30/d60/d90 checkpoint reports · quarterly program review with the failure-mode register updated.

What good looks like by day 90: the cohort's TTP median and retention rate are on management's desk with confounds attached, every triggered intervention has a paper trail, and the failure-mode register has at least one entry updated from real data.

Fictional disclosure: VanDijk Logistics and all roles/figures are a fictionalized composite for this portfolio. Stakeholder pack drafted with AI assistance; final judgment and copy are human-owned.

Prefer the source? Download the raw markdown.