Task suite in development

RevOpsEval

An evaluation benchmark for AI agents on Revenue Operations tasks. Public leaderboard. Open methodology. Real workflows.

What it measures

How accurately AI agents perform the recurring analytical work of a RevOps leader. Tasks scored against expert practitioner judgment, not synthetic intuition.

Sample tasks

Diagnose a Salesforce org's pipeline hygiene from raw CRM data
Prepare a weekly forecast call memo with rep-level variance analysis
Pressure-test a stuck $750K deal and recommend forecast category
Triage activity capture failures across reps and identify systemic issues
Stress-test a comp plan against the prior year's actuals

Who it's for

AI model evaluators, agent framework builders, and RevOps leaders comparing tools. Free and public. Models submit results via API or self-hosted run.

Status

Four of the five sample tasks already have working reference implementations, shipped as open-source Agent Skills in revops-skills. The formal task specification, scoring harness, and public leaderboard are in development.