RevOps Tech Stack Design
Overview
Scott Brinker's annual MarTech Landscape has tracked vendor count from roughly 150 in 2011 to more than 15,000 by 2025 — a 100× proliferation that puts every B2B company under constant pressure to add tools, not consolidate them. The average B2B company uses 15+ GTM tools; most are underutilized or redundant. This skill prevents the most expensive outcome of tool sprawl: a stack where every tool functions in isolation and the revenue motion is unmeasurable because data lives in a dozen disconnected places.
The design principle, from Winning by Design's Revenue Architecture, is that every tool in the stack must instrument a stage of the bowtie funnel — acquisition, activation, close, onboard, expand, renew. Tools that cannot be mapped to a documented process stage are candidates for elimination. Gartner's guidance for advanced-maturity RevOps teams includes inventorying all sales, marketing, and CS technology and eliminating redundant tooling as step one of any ops improvement program.
When to Use
- "Audit our GTM tech stack"
- "Design RevOps stack"
- "Sales tool consolidation"
- "GTM tool budget review"
- "Stack cost optimization"
- "Revenue technology architecture"
- "Approve new GTM tool purchase"
- "Ramp / vendor spend governance"
Authoritative Foundations
- Scott Brinker — MarTech Landscape (chiefmartec.com, 2025). The named justification for why consolidation discipline is a structural problem, not a one-time cleanup. 15,000+ vendors in 2025 means the default answer for any GTM need is "there's a tool for that." The audit in Phase 1 applies Brinker's consolidation logic: inventory and score on utilization and overlap before evaluating any new vendor.
- Gartner — Revenue Operations Research. Gartner's maturity guidance for RevOps includes: inventory all GTM tech as step one; eliminate redundant tooling before adding headcount; every retained tool needs a documented process owner and integration path to the CRM. Advanced-maturity RevOps teams are 2× more likely to exceed revenue goals — tech rationalization is one of the measurable components separating advanced from intermediate maturity.
- Forrester Decisions — Revenue Operations. "Deliver value through revenue technology" is a named priority in Forrester's RevOps responsibility model. Forrester's caution: revenue tech decisions must trace to the operating model, not tool fashion. Phase 3's target-state design enforces this by requiring every tool to map to a bowtie stage before it earns a place in the target architecture.
- Winning by Design — Revenue Architecture. The bowtie funnel — acquisition, activation, close, onboard, expand, renew — is the operating model the stack exists to instrument. Each bowtie stage gets exactly one category-owner tool in the target state; tools without a bowtie-stage assignment are cut.
Step-by-Step Process
Phase 1: Stack Audit
Inventory every tool in the GTM motion. For each tool, record:
| Field | Description |
|---|---|
| Tool name | Vendor + product name |
| Category | CRM, enrichment, sequencing, CI, forecasting, etc. |
| Annual cost | Full contract value |
| Owner | Named individual responsible for the tool |
| Monthly active users | Unique users in last 30 days |
| Utilization % | MAU ÷ total licensed seats |
| CRM integration | Yes / No / Partial |
| Bowtie stage | Which funnel stage the tool instruments |
| Renewal date | Contract renewal month |
Categories to audit: CRM, enrichment, visitor identification / intent (company vs person deanonymization), sequencing, conversation intelligence, forecasting, compensation, territory planning, ABM platform, chat/scheduling, content management, analytics/BI, data warehouse, iPaaS/integration, partner management, project management.
Visitor ID category owner: Load website-visitor-identification/references/visitor-id-vendor-comparison.md.
Consolidation rule: one company-ID vendor (Clearbit/Breeze, Leadfeeder, 6sense,
Demandbase) + at most one person-ID vendor (RB2B, Warmly, Koala) until pilot
proves ROI. Person-level tools require privacy checklist
(visitor-id-privacy-gtm.md) before stack approval.
Full audit scorecard template: see references/framework-notes.md.
Phase 2: Consolidation Analysis
Score each tool on five dimensions to build a keep/consolidate/cut recommendation:
- Utilization: MAU ÷ licensed seats. Below 50% is a consolidation signal.
- Overlap: Two or more tools in the same category without documented differentiation is a red flag.
- Integration: No CRM integration means data lives outside the single source of truth — a structural liability.
- Cost per active user: Annual cost ÷ MAU. Above $200/month/user without documented ROI is a cut signal.
- Switching cost: Deeply integrated tools (Salesforce, Outreach) cost 3-6 months to replace; point solutions (Calendly) cost one week.
Consolidation target: 15-25% reduction in GTM tool spend, computed from the
actual audit numbers — not asserted before the audit is run. See
references/framework-notes.md for the consolidation decision matrix and
category-owner table.
Phase 3: Target-State Architecture by ARR
Every retained tool maps to exactly one bowtie stage. Each category has exactly one owner tool in the target state.
< $1M ARR — Minimum viable stack:
- CRM: HubSpot (free) or Attio
- Enrichment: LeadMagic, Apollo
- Sequencing: Smartlead or Instantly
- Analytics: Google Analytics + CRM reports
- Calendar/booking: Calendly
$1–5M ARR — Scaled stack:
- CRM: HubSpot (paid) or Salesforce Essentials
- Enrichment: LeadMagic + Clay waterfall
- Sequencing: Outreach or Salesloft
- Conversation intelligence: Gong or Chorus
- Analytics/BI: CRM dashboards + Looker or Metabase
- iPaaS: n8n or Zapier for custom integrations
$5M+ ARR — Full RevOps stack:
- CRM: Salesforce
- Enrichment: Clay waterfall + dedicated enrichment ops
- Sequencing: Outreach or Salesloft
- Conversation intelligence: Gong
- Forecasting: Clari or BoostUp
- Territory/compensation: Fullcast, CaptivateIQ, or Spiff
- Data warehouse: Snowflake or BigQuery + dbt
- BI: Looker or Tableau
Phase 4: Integration Architecture
The CRM is the single source of truth. Every data flow reads from or writes to the CRM; no closed loops between non-CRM tools.
- Enrichment → CRM: Clean, enriched records flow automatically within 48 hours of creation.
- CRM → Sequencing: Accounts and contacts assigned to sequences carry CRM context (stage, persona, industry).
- Sequencing → CRM: Activity logged back — emails sent, calls made, replies received, meetings booked.
- CRM → Analytics: Pipeline and activity data feeds the BI layer.
- iPaaS layer: Bridges tools without native CRM connectors; every integration has a named owner and a documented data contract (source, destination, field mapping, sync frequency).
Phase 4b: Customer & Prospect Data Boundaries (GTM)
RevOps owns where customer and prospect data may live in the stack — not encryption design.
| Rule | Why |
|---|---|
| CRM is system of record for GTM-sourced contacts | Sequencers and enrichment are spokes |
| No customer PII in outbound tools beyond lawful prospecting fields | Sequences are not file shares |
| Bulk exports require owner + retention date | Audit trail for enterprise buyers |
| Enrichment credits gated per role | Prevents shadow list hoarding |
When sales or CS exchanges files with customers (imports, POC data), route
through references/gtm-data-exchange-playbook.md — not ad hoc Slack/email.
Checklist → deal-desk/templates/customer-data-exchange-checklist.md.
Rep hygiene → references/gtm-security-hygiene-basics.md.
Phase 4c: Stack Migration & Rollout Projects
Tool changes are GTM projects — not side quests between tickets.
- Charter —
gtm-operations→skills/gtm-ops/gtm-operations/templates/gtm-project-charter.md - RACI —
skills/gtm-ops/gtm-operations/templates/raci-matrix-template.mdExample 2 (CRM Migration) or Example 4 (Vendor Evaluation); RevOps Lead Accountable on mapping, integrations, and cutover - Milestones — audit complete → sandbox UAT → training → cutover → day-30 adoption
- PM tool — ClickUp RevOps Projects Folder + Gantt (
clickup-gtm-workspace.md) - Spend —
gtm-spend-managementapproval before POC contracts
Do not add vendors from Phase 1 without a named project DRI and bowtie-stage owner.
Phase 5: Cost Optimization
- Consolidation: Replace 3-4 overlapping point solutions with one platform. HubSpot's paid tiers can replace 5-7 point solutions at $1-5M ARR.
- License audit: Remove departed-employee seats; downgrade active-but-not- admin seats to viewer/read-only.
- Contract negotiation: Multi-year commits typically produce 20-30% discounts; ramp pricing (lower year one, scales with usage) protects early runway.
- Savings baseline: Compute 15-25% savings from audit actuals (cost per active user × redundant or unused seats) — commit to a target only after the audit data supports it.
Output Format
RevOps stack blueprint containing: full audit inventory with utilization
percentages and bowtie-stage assignments, consolidation analysis with
keep/consolidate/cut recommendations and a computed savings estimate, target-
state architecture by ARR stage with one category-owner tool per bowtie stage,
integration architecture with named data owners and directional data-flow
contracts, and cost-optimization plan with negotiation levers and a
baseline-derived savings range. See references/framework-notes.md for the
audit scorecard template and category-owner matrix.
Quality Check
Before delivering, verify:
- Every retained tool has a documented bowtie-stage assignment — no tool without a process home
- Each category has exactly one owner tool in the target-state architecture
- Utilization is measured from actual MAU data, not self-reported or assumed
- Consolidation savings estimate is derived from audit actuals, not asserted upfront
- Every integration has a named data owner and a documented direction (source → destination)
- Stack is scoped to the company's current ARR stage, not aspirational tooling
Common Pitfalls
- Adding tools before auditing existing ones. With 15,000+ vendors (Brinker 2025), the default GTM motion is "there's a tool for that." Fix: run the Phase 1 audit before evaluating any new vendor; close a tool before opening one.
- Tool decisions driven by tool fashion, not operating model. Forrester's warning: revenue tech that doesn't trace to the operating model creates data fragmentation. Fix: every tool in Phase 3 must map to a bowtie stage; if you cannot name the stage, the tool does not belong in the stack.
- CRM disconnected from the integration hub. Data that lives only in sequencing or enrichment tools is invisible to leadership. Fix: enforce the hub-and-spoke model in Phase 4 — every tool either pushes to or pulls from CRM; no closed loops outside the CRM.
- Asserting savings targets before running the audit. "We'll save 20%" stated before utilization is measured is fiction. Fix: compute the savings figure from Phase 2 audit actuals; commit to a target only once the data supports it.
Execution Artifacts
references/framework-notes.md— Audit scorecard, consolidation matrix, savings mathtemplates/output-template.md— Deliverable shellscripts/check-output.pywebsite-visitor-identification/references/visitor-id-vendor-comparison.md— Intent/deanonymization vendorswebsite-visitor-identification/templates/visitor-id-vendor-eval-scorecard.md— Pilot scorecardreferences/gtm-data-exchange-playbook.md— Customer data exchange SOP (repo root)gtm-operations/templates/raci-matrix-template.md— Migration + vendor eval RACI (Examples 2, 4)gtm-operations/templates/gtm-project-charter.md— Rollout chartergtm-operations/references/gtm-project-management-playbook.md— Tool rollout milestonestemplates/stack-audit-scorecard.md— Tool inventory worksheettemplates/target-state-architecture.md— Category-owner target stack
Related Skills
- gtm-operations, gtm-tool-cost-model, gtm-spend-management, crm-integration, campaign-governance, analytics
- website-visitor-identification — canonical visitor ID / intent vendor selection
- deal-desk, customer-onboarding — customer data exchange at sale and onboarding
- revenue-team-onboarding — rep access and hygiene gates
- gtm-role-descriptions — GTM Engineer vs RevOps ownership (
gtm-engineer-hiring.md) - clay-toolkit, n8n-automation — skills to test in GTM Engineer interviews