REVOPS

Revenue intelligence that actually moves your forecast.

Three-scenario forecasting, 1,000-run Monte Carlo revenue simulation, and explainable deal scoring — wired directly into HubSpot and Salesforce. No black-box numbers, no hype metrics, no fabricated win-rate lifts.

View Documentation
2
CRMs at launch
3
Forecast scenarios
1,000
Monte Carlo iterations
13
Reasoning methodologies

The forecast is wrong. Everyone knows it.

Reps sandbag. Leaders pad. Finance applies a haircut on top of a haircut. By the board meeting, nobody trusts the number and the spreadsheet has become an argument about emotions instead of a model of the pipeline.

What a typical CRM tells you

  • — A single weighted-pipeline number
  • — Fixed-rule deal scores with no explanation
  • — Stale deals you have to query for manually
  • — Forecasts that change when you reload the page

What RevOps tells you

  • — Conservative / Realistic / Aggressive, always visible
  • — Deal scores with the full signal breakdown attached
  • — Stale-deal detection as a first-class signal
  • — A Monte Carlo distribution you can cite, not a point estimate

What you get

Six capabilities, all backed by production code. Nothing on this page is a stub or a roadmap promise.

3-Scenario Forecasting

Every forecast runs Conservative, Realistic, and Aggressive paths in parallel. You see the spread, not a single number with false precision.

Monte Carlo Revenue Simulation

1,000-iteration Monte Carlo simulation over monthly new-customer and churn distributions. Real np.random draws, not three hardcoded multipliers.

Deal Scoring, Explained

Methodology #213 scores every open deal on win probability with the signal breakdown attached. No black box — you see which inputs moved the number.

Stale-Deal Detection

Pipeline health checks surface deals that have stopped moving, deals past their expected close, and deals missing decision-maker contact.

HubSpot + Salesforce, Native

Direct client integrations to HubSpot and Salesforce. OAuth via Nango, incremental sync, SOQL + CRM API. Pipedrive, Close and Zoho on the roadmap.

Constitutional Guardrails

The AI will not recommend aggressive closes on cold accounts or flag risky discount approvals. Rules written in plain English, enforced at the reasoning layer.

Three employees on the RevOps desk

Every forecast is a consensus between Scout, Finn, and Maestro — each independently scoring, each showing their work before the ensemble picks the final answer.

S
Scout
Market Intelligence

Analyses every deal in context. Researches the account, the competitor, and the segment before scoring.

F
Finn
Financial Modelling

Runs the Monte Carlo. Owns the 3-scenario forecast. Talks in ARR, ACV, and cycle days — not hype.

M
Maestro
Pipeline Operations

Watches pipeline health. Surfaces stale deals, raises flags on slip risk, orchestrates the CRM sync layer.

RevOps Pricing

Target launch pricing. Billed separately from your AI Employee subscription. Locked for the first year of any waitlist signup.

REVOPS STARTER
From $299/mo

For one sales team connecting a single CRM.

  • 1 CRM connection (HubSpot or Salesforce)
  • Up to 500 open deals tracked
  • 3-scenario forecasting
  • Stale-deal detection
  • Weekly pipeline health report
  • Email support
RECOMMENDED
REVOPS PRO
From $999/mo

For revenue leaders who run the forecast themselves.

  • Both HubSpot + Salesforce
  • Unlimited open deals
  • Monte Carlo revenue forecasting
  • Scout + Finn + Maestro squad
  • Methodology #213 deal scoring
  • Priority support + Slack channel
REVOPS ENTERPRISE
From $2,999/mo

For multi-region RevOps orgs with custom workflows.

  • Unlimited CRM connections
  • Custom deal-scoring signals
  • Constitutional guardrail authoring
  • Dedicated model retraining
  • SSO + audit log export
  • Named CSM
Contact Sales

POWERED BY THE COGNITIVE MESH

13 reasoning methodologies, working together

RevOps is not a wrapper around one model. Deal scoring, forecasting, and pipeline health each flow through the same 13 cross-cutting methodologies that make the reasoning explainable and the output stable.

#1
Tree-of-Thought
Explore multiple deal strategies, pick the best path
#3
Graph-of-Thought
Multi-dimensional deal scoring (budget × authority × need × timing)
#7
Self-Consistency
Multiple forecast paths, take the majority
#8
Constitutional AI
Refuse aggressive closes on cold accounts
#9
Self-Refine
Re-score deals on every new CRM activity
#12
Episodic Memory
"This account looks like one we won in Q2 — here's what worked"
#13
Semantic Knowledge Graphs
Account-contact-deal-touch network, multi-hop queries
#16
Theory of Mind
Model what the buyer is actually blocked by
#49
Mixture of Agents
Scout + Finn + Maestro each score independently, ensemble
#50
Mixture of Expert Agents
Route mid-market to Scout, enterprise to Kenna
#58
Tool-Using Agents
HubSpot and Salesforce API calls inside the reasoning loop
#176
Agent Workflow Memory
Reuse plays that worked on similar past deals
#188
Markovian Thinking
Compress a 12-month account history into stable state

Plus Methodology #213 (REVOPS Deal Analysis) and #199 (Revenue Forecasting).

INTEGRATIONS + COMPLIANCE

Integrates with HubSpot and Salesforce today. Pipedrive, Close, and Zoho coming.

WHAT'S REAL TODAY

  • — HubSpot client (OAuth via Nango, incremental sync)
  • — Salesforce client (SOQL + REST, OAuth via Nango)
  • — Envelope encryption on all CRM credentials at rest
  • — Zero Trust Architecture (Methodology #121) for all access
  • — No model training on your CRM data, ever

PRE-CERTIFICATION STATUS

  • — SOC 2 Type II: not yet audited (targeting post-launch)
  • — GDPR deletion workflow: in design
  • — HIPAA: out of scope for RevOps
  • — Data residency: US + EU (post-launch)
  • — Audit trail: every AI decision logged with citations

Questions a skeptical CTO asks

Real. The simulation lives in src/methodologies/skills/financial_modeling.py and runs 1,000 iterations by default, drawing new-customer and churn rates from normal distributions around your historical monthly data. You can raise or lower the simulation count per request. It is not three hardcoded multipliers dressed up with a Monte Carlo label.

HubSpot and Salesforce, both with production client code. Pipedrive, Close, and Zoho are on the roadmap and will show up as "Coming Soon" badges inside the app until their clients ship. We would rather say no honestly than ship a stub you have to work around.

Native CRM scoring uses fixed rule weights inside the CRM. RevOps runs Methodology #213 — a reasoning pass that scores win probability and attaches the signal breakdown, risk factors, and next-step recommendations. Under a feature flag, an XGBoost ML model (Methodology #216) takes over scoring with explainable features. You see why a deal scored the way it did, not just the number.

No. We use APIs only. Your CRM data never becomes training data for a shared model. If you opt into per-org ML scoring, the model is trained on your data and persisted only for your workspace.

CRM OAuth is handled by Nango, so connecting HubSpot or Salesforce is a guided OAuth flow — usually under 5 minutes. The first historical sync depends on pipeline size but runs in the background; you can start reviewing forecasts as soon as the initial sync completes.

Stop arguing about the forecast.

Join the RevOps waitlist. Launch pricing locked for the first year.