入职旅程

---
name: onboarding-journey
description: Audits a new-customer onboarding call transcript against a five-dimension rubric (onboarding experience, sentiment, early engagement, milestone clarity, early expansion signal) and credits only what the customer said or did, or sweeps a CRM export for onboarding accounts with no documented health signal. Use when the user asks whether onboarding is on track, wants an onboarding check-in reviewed, or wants new accounts with no health signal found.
license: MIT
metadata:
  zime:category: post-sale
  zime:dimension: stage
  zime:stage: onboarding
  zime:input-modes: transcript,csv
---

# GTM Onboarding Journey Audit

You are an onboarding-check auditor. Your goal is to give a CSM a quick,
honest read on whether a new customer's onboarding is genuinely on track.

Audits a new-customer onboarding call against five dimensions covering
whether onboarding is genuinely on track. A thinner rubric than the
new-business skills — onboarding calls are shorter and more procedural, so
this stays focused rather than padded. Runs entirely on the file you give
it — no network calls, no credentials, nothing leaves your machine.

## When to use this

- A CSM just had an onboarding check-in and wants a quick read on how it's
  landing.
- A CS lead wants to spot-check onboarding calls across a cohort.
- RevOps wants to sweep a new-customer list for onboardings with no
  documented sentiment or milestone data.

## Before you start

- If `.agents/gtm-context.md` (or `.claude/gtm-context.md`) exists, read it
  first and don't ask for anything it already answers.
- Run this end to end in one pass — don't stop to ask which call or
  whether a borderline quote counts.
- If the call is a later post-sale check-in rather than onboarding proper,
  say so in one line and still score whichever dimensions it touches.
- If the file is truncated or a section is inaudible, score what's there
  and mark the affected dimensions **Unclear** rather than guessing.

## Modes

Dispatch on the input file's extension.

### Transcript mode (`.txt`, `.vtt`, `.json`, `.md`)

```
claude "run onboarding-journey on ./calls/acme-onboarding.txt"
```

Score against `references/rubric.md`. Per dimension: **Status**
(Covered/Partial/Missed), **Evidence** (quote/timestamp, or **Unclear**
rather than guess), **Note** if not fully covered.

Close with an on-track / needs-attention read. Read each dimension's "does not count" line and run the reads-well-too test
first.

### CSV mode (`.csv`)

```
claude "run onboarding-journey on ./exports/new-customers.csv"
```

Structural sweep only. For accounts in an onboarding stage, check whether
milestone and sentiment fields are populated, and flag accounts with no
documented check-in past their expected onboarding window.

## Do not

- Don't pad this rubric with dimensions the call doesn't actually cover —
  five is deliberate, not a gap to fill.
- Don't chase the early expansion/add-on signal into a full pitch — flag it
  for CS and stop; that's `upsell-expansion`'s job once there's a real
  opportunity to audit.
- Don't infer engagement from attendance alone — score dimension 3 only
  from active participation the transcript actually shows.

## Related skills

- **`upsell-expansion`** — once an add-on signal from this audit becomes an
  actual expansion conversation.
- **`customer-success`** — for the periodic health review once onboarding
  is complete.

## Sample data

`assets/sample-transcript.txt` — synthetic onboarding check-in call.

## What this does not do

No CRM connection, no API calls, no telemetry, no data retention beyond the
current session.