# RevOps Desk

> Paste how leads move from marketing to sales today - the lifecycle stages, what makes an MQL, the scoring rules and points, how leads are routed to an owner, the response SLA, what happens to a rejected lead, the pipeline stages and the funnel numbers - plus the business, the motion and the CRM. The browser reads every scoring rule, checks the MQL needs fit and engagement, reads the routing, the SLA and the speed-to-lead, holds lead-to-MQL, MQL-to-SQL, SQL-to-opportunity and win rate against the revops skill's benchmarks and raises 25 flags, free; then an audit says what to fix and a spec writes the lifecycle, scoring, routing, SLAs and reason codes to implement.

URL: https://revops-desk.skillsafe.ai/
API tutorial: https://revops-desk.skillsafe.ai/api.html

## Who it is for

Founders, marketing leads, sales leads and revenue operations people whose sales team calls the MQLs junk, whose leads sit in a shared inbox, whose reps respond in days, whose rejected leads vanish, whose deals stall in Proposal - or who have no lifecycle or scoring yet and want to know what to build first.

## The free scan (in the browser, no model)

- The lifecycle stages named (Subscriber, Lead, MQL, SAL, SQL, Opportunity, Customer, Evangelist) and the MQL definition, and whether one download makes an MQL.
- Every scoring rule with its points, sorted into fit, engagement and negative; the MQL threshold; the most a content action and a buying-intent action earn.
- The routing method (round-robin, territory, named account, skill-based, or a shared queue), a fallback owner, PTO and capacity rules.
- The first-contact SLA, the stated speed-to-lead in minutes, escalation, and what happens to rejected leads.
- Pipeline stages, stalls and stale-deal alerts; data spread across systems.
- Funnel rates: lead-to-MQL (5-15%), MQL-to-SQL (30-50%), SQL-to-opportunity (50-70%), win rate (20-30%), and the weakest stage.
- Flags RO-01..RO-25, from engagement-only MQLs and missing negative scoring to shared-inbox routing, two-day follow-up and dropped rejections.

## The paid lanes (one system prompt, routed on `task`)

- `audit`: the process (motion, CRM, a one-line read), the six areas (lifecycle, scoring, routing, handoff, pipeline, data: good, weak, bad or unknown, each with a finding, quotes and a fix), the funnel (the page's rates, speed-to-lead, the weakest stage), issues as finding, impact and fix with the page's flag ids, 3 to 6 fixes, and a verdict: sound, fix first or rebuild.
- `spec`: the RevOps spec - lifecycle stages with entry and exit criteria and owners, an MQL definition needing fit and engagement, a scoring model (fit, engagement with decay, negative, threshold, weighting, recalibration), routing rules with a fallback, SLAs with escalations, rejection reason codes, pipeline stages with exit criteria and required fields, metrics with the skill's targets, and CRM automations. It can carry the audit's findings by a handoff button.

Every reply is checked against the input: quotes word for word, funnel rates equal to the page's arithmetic, flags carried, and no invented count, rate, price or window beyond the skill's constants.

## Source

Derived from the agent skill @coreyhaines31/revops (https://skillsafe.ai/skill/@coreyhaines31/revops), by Corey Haines, in coreyhaines31/marketingskills (MIT): https://github.com/coreyhaines31/marketingskills/tree/main/skills/revops - its principles, lead lifecycle framework, MQL definition and handoff SLA, scoring models (references/scoring-models.md), lifecycle definitions and reason codes (references/lifecycle-definitions.md), routing rules and speed-to-lead data (references/routing-rules.md), pipeline hygiene and RevOps metrics.

Runs on gpt-terra. Part of SkillSafe (https://skillsafe.ai/).