Add organizational context to your AI stack

Pricing is based on three components: a platform fee, the number of active data source integrations (Slack, Teams, Calendar, HRIS, etc.), and API usage volume. This means you pay for what you actually connect and consume, not a flat per-employee fee. We work with organizations from 500 to 50,000+ employees.

Yes. Every engagement starts with a sandbox environment where you can test BehaviorGraph with your own data and integrations. Pilots typically run 4 to 8 weeks with a limited set of teams and data sources before expanding org-wide.

BehaviorGraph operates exclusively on metadata: collaboration patterns, calendar signals, presence data, and survey responses. It never reads, processes, or stores message content, email bodies, or document text. This is an architectural constraint, not a configuration option.

API, MCP, and RAG endpoints return coded identifiers (e.g. LEG-07, ENG-01) instead of employee names. Your internal systems handle identity resolution with your own access controls. This prevents misuse for individual performance evaluation while keeping the context fully usable for routing and escalation.

Yes. Enterprise deployments support cloud-hosted, single-tenant, and on-premises options. We work with your security and infrastructure teams to determine the right deployment model for your organization.

Traditional ONA produces periodic reports for HR. BehaviorGraph turns the same behavioral signals into continuously updated, queryable infrastructure that AI systems can use at runtime. It is not a dashboard for people teams. It is a context layer for enterprise AI.

No. BehaviorGraph works three ways: (1) As an overlay on top of your existing enterprise search, enriching every result with people context and routing. (2) As a standalone API that your AI agents query directly for routing, escalation, and governance decisions. (3) As a full behavioral intelligence platform with its own chat, dashboards, and automation. If you already have enterprise search, we make it better. If you don't, we still cover the organizational context that search alone can't provide.

Your agents call our API or MCP (Model Context Protocol) tools before taking action. BehaviorGraph returns who to route to (based on behavioral relevance), who's available as backup (when the primary is overloaded or on leave), and when to escalate (with the full decision path and approval chain). Works alongside platforms like LangChain, CrewAI, or any MCP-compatible agent. Or use our API directly without any other platform. We provide the part others don't have: organizational behavioral context.

Yes. Most companies start with one entry point: search enrichment (if you already run enterprise search), agent routing (if you're deploying AI agents), or bottleneck detection (if you want to find where work stalls). Each use case connects to the same behavioral graph, so expanding later requires no additional data setup or integration work.