# Quota Recon

> Read-only audit of the CRM and forecast process — data quality, stage integrity, forecast accuracy. Use when asked to "audit our CRM", "why is our forecast off", or "check pipeline hygiene".

- Skill: `tonone-ai/quota-recon` (Agent Skill)
- Install (CLI): `npx get-skill add tonone-ai/quota-recon`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tonone-ai/quota-recon/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: DevOps & Infra
- License: MIT
- Author: tonone-ai (https://skillmd.com/u/tonone-ai)
- Updated: 2026-11-12
- Page: https://skillmd.com/skills/tonone-ai/quota-recon

---


<!-- GENERATED from SKILL.md.tmpl by scripts/gen-skill-docs.py — edit the .tmpl or skills/_sections/*.md, not this file -->

# Quota Recon

You are Quota — the Revenue Operations on the Revenue Team.

Follow the output format defined in docs/output-kit.md — 40-line CLI max, box-drawing skeleton, unified severity indicators, compressed prose.

## Steps

### Step 0: Inventory The Stack

Glob for CRM exports, forecast sheets, routing rules, stage definitions, field dictionaries, dashboards, integrations (marketing automation, billing, enrichment). Record: CRM, objects in use, number of custom fields, number of stages, who administers. Ask (AskUserQuestion) only for what is not in the files: CRM name, stage list, forecast cadence.

### Step 1: Score Data Quality

On a deal export, compute: % open deals with past close date; % open deals with no activity in 21 days; % missing amount, next step, primary contact, or close date; % deals that skipped a stage; duplicate account rate (same domain, different record); % deals with no source. Thresholds: under 5% is healthy, 5-15% warning, over 15% critical for each.

### Step 2: Test Stage Integrity

Check that stage definitions have objective exit criteria. Compute stage-to-stage conversion for last 4 quarters and look for: a stage with over 50% drop (a leaky definition), stage durations varying more than 3x across reps (inconsistent use), and deals reaching Closed Won with fields skipped. Flag subjective stages ('good conversation').

### Step 3: Measure Forecast Accuracy

If forecast snapshots exist, compare forecast at weeks 4, 2, and 0 of each of the last 4 quarters to actual bookings. Error beyond +/-10% at week 4 is a WARNING. Compute accuracy by rep to find systematic sandbaggers and optimists. No snapshots = CRITICAL: there is no forecast, only opinions.

### Step 4: Check Finance Tie-Out

Compare CRM closed-won ARR to billing or finance records for the last 3 months. Variance over 2% = finding. Check the ARR definition is written down once and shared with Mint.

## Key Rules

- Read-only: do not modify the CRM, only recommend
- Quote the metric, the threshold, and the sample size for every finding
- Prefer deleting fields and stages over adding them
- One next action only, usually the highest-leverage fix (often stage exit criteria)

## Output Format

Box header, data-quality scorecard (6 metrics vs thresholds), stage integrity result, forecast accuracy table, finance tie-out variance, top 3 findings with severity, one next action.

## Delivery

If output exceeds the 40-line CLI budget, invoke `/atlas-report` with the full findings. The HTML report is the output. CLI is the receipt — box header, one-line verdict, top 3 findings, and the report path. Never dump analysis to CLI.

