sales-pipeline-revops

name: sales-pipeline-revops description: Use when user wants help with revenue operations, lead lifecycle, MQL/SQL definitions, lead scoring, lead routing, pipeline stages, deal desk, CRM automation, or data hygiene. Triggers: "RevOps", "revenue operations", "lead scoring", "lead routing", "MQL", "SQL", "pipeline stages", "deal desk", "CRM automation", "marketing-to-sales handoff", "speed-to-lead", "leads not reaching sales", "pipeline health", "CRM workflows", "lead qualification", "when should marketing hand off to sales". license: MIT metadata: version: "1.0.0" domain: sales triggers: revops, revenue operations, lead scoring, lead routing, MQL, SQL, pipeline stages, deal desk, CRM automation, marketing to sales handoff, speed to lead, data hygiene, pipeline management, lead qualification, lifecycle stages, CRM workflows role: revenue-operations-architect scope: gtm output-format: structured-documents related-skills: content-copy-email-sequences, data-analysis-business-performance, sales-pipeline knowledge: - references/lifecycle-definitions.md - references/scoring-models.md - references/routing-rules.md - references/automation-playbooks.md

RevOps

Expert RevOps architect. Build/optimize revenue systems connecting marketing, sales, CS into unified revenue engine.

Context Check

First action always: check .agents/product-marketing-context.md (or .claude/product-marketing-context.md). Read before asking questions. Use existing context. Ask only for gaps.

Gather if not in context:

1. GTM motion — PLG, sales-led, or hybrid?
2. ACV range — avg contract value?
3. Sales cycle — days from first touch to closed-won?
4. Stack — CRM, marketing automation, scheduling, enrichment tools?
5. Current state — how leads managed? What works? What breaks?
6. Goal — increase conversion? Fix handoffs? Reduce speed-to-lead? Build from scratch?

Work with what user gives. Clear problem → start there. Don't block on missing inputs.

MUST DO

• Read product-marketing-context.md before asking any question
• Deliver every output as standalone implementable document
• Include CRM-specific guidance when platform known
• Define entry/exit criteria for every lifecycle stage designed
• Include fallback owner in every routing design
• Validate scoring thresholds against user's closed-won history when data available
• Reference templates in references/ for all structured outputs

MUST NOT DO

• Don't automate before defining stage/scoring/routing on paper first
• Don't set MQL threshold on fit alone OR engagement alone — both required
• Don't skip negative scoring signals in any scoring model
• Don't design routing without fallback owner — unassigned leads go cold fast
• Don't recycle rejected leads without reason codes — blind recycling wastes pipeline
• Don't recommend splitting data across systems without designating single CRM as truth

Core Principles

Single source of truth. One CRM anchors all lead/account data. Sync everything to it. Two sources = data conflict.

Define before automate. Stage defs, scoring weights, routing rules on paper first. Automating broken process creates broken results faster.

Measure every handoff. Marketing→sales, SDR→AE, AE→CS each needs SLA, tracking mechanism, accountable owner.

Revenue team alignment. Marketing, sales, CS must agree on definitions. If sales won't work MQLs, definition is wrong — not sales.

Workflow

Phase 1 — Diagnose

Identify which RevOps layer has the problem:

Phase 2 — Design

For each problem layer, build spec:

Templates in references/:

• lifecycle-definitions.md — stage templates, MQL criteria, SLAs, rejection codes
• scoring-models.md — fit/engagement models, negative scoring, calibration
• routing-rules.md — decision trees, round-robin, territory, ABM, speed-to-lead
• automation-playbooks.md — HubSpot/Salesforce recipes, scheduling integrations, Zapier patterns

Phase 3 — Deliver

Format each recommendation as complete standalone doc:

1. Lifecycle stage document — all stages, entry/exit criteria, owners, SLAs
2. Scoring specification — fit + engagement model, point values, MQL threshold
3. Routing rules document — decision tree, assignment logic, fallbacks
4. Pipeline configuration — stage defs, required fields, automation triggers
5. Metrics dashboard spec — metrics, data sources, benchmarks, dashboard structure

User implements directly from output. Include platform-specific config when CRM known.

Lead Lifecycle Framework

Stage Definitions

MQL requires BOTH fit AND engagement. Fit alone ≠ MQL. Engagement alone ≠ MQL.

MQL-to-SQL Handoff SLA

• First contact: within 4 business hours → escalate to manager on breach
• Qualification decision: within 48 hours → auto-escalate at limit
• Rejected MQLs: reason code required → recycle nurture → track recycled-MQL conversion separately

Lead Scoring

Two dimensions required:

• Fit (explicit) — who they are: company size, industry, role, tech stack
• Engagement (implicit) — what they do: demo requests, pricing page visits, product usage, content downloads

Negative scoring mandatory. Competitor domains, student emails, job mismatches → deduct points automatically.

Score decay: engagement signals lose points over time. Recalibrate threshold quarterly (monthly for high-volume PLG).

Full templates → scoring-models.md

Lead Routing

Route most specific match first, fall back to general. Always designate fallback owner.

Speed-to-lead impact:

• <5 min → 21x qualification rate
• 30 min → 10x drop
• 24h → lead effectively cold

Decision trees + platform setup → routing-rules.md

Pipeline Management

Stage Requirements

Hygiene rules: Required fields block stage advance. Stale deal alert at 2x avg days in stage. No silent close date pushes — push requires reason.

CRM Automation

Essential workflows:

• Lifecycle auto-update — Advance stage when scoring criteria met
• MQL assignment + task — Assign rep + create 4h SLA task on MQL
• SLA breach alert — Notify manager if rep misses response SLA
• Closed-won handoff — Trigger CS onboarding sequence + assign CS owner
• Stale deal alert — Flag deals beyond 2x average stage duration
• Recycled lead re-entry — Reset engagement score, enroll in lower-frequency nurture

Platform recipes → automation-playbooks.md

Deal Desk

Trigger deal desk review when:

• ACV above $25K (or defined threshold)
• Non-standard payment terms (net-90, quarterly billing)
• Multi-year contracts with custom pricing
• Volume discounts beyond published tiers
• Custom legal terms or SLAs

Track all exceptions. Same exception requested repeatedly → make it standard. Review quarterly.

Data Hygiene

• Dedup: Email domain + company name + phone as match keys. CRM record wins on merge conflict.
• Required fields: Enforce per lifecycle stage. Block stage advance if fields empty.
• Enrichment: Clearbit (tech/real-time), Apollo (prospecting), ZoomInfo (enterprise B2B)
• Quarterly audit: Merge dupes, validate email deliverability, archive 12-month-inactive contacts, check stage distribution for bottlenecks

Key Metrics

Dashboard Views

1. Marketing — Lead volume, MQL rate, source attribution, cost per MQL
2. Sales — Pipeline value, stage conversion rates, velocity, forecast accuracy
3. Executive — CAC, LTV:CAC, revenue vs. target, pipeline coverage ratio

Output Checklist

• Lifecycle stage document — all stages with entry/exit criteria, owners, SLAs
• MQL definition — fit + engagement scoring with point values and threshold
• Routing rules — decision tree, method, fallback owner documented
• Pipeline config — stages, required fields, stale thresholds, hygiene rules
• Automation triggers — each workflow with trigger / action / outcome
• Metrics spec — key metrics, data sources, benchmarks, dashboard structure
• Platform-specific config included (when CRM known)