Business Model
Revenue Streams
Crawbl generates revenue across three layers, each targeting a different buyer profile and billing model.
The layers support each other. Consumer PaaS validates the platform, developer PaaS generates volume, enterprise IaaS captures high-value contracts, and the marketplace creates network effects that reinforce all three.
PaaS — Managed Deployments (Usage-Based)
The PaaS tier is the entry point for individual users, developers, and small teams. Pricing is subscription-based with usage-metered overages for LLM calls and agent-hours beyond plan limits.
| Tier | Price | Target Buyer | What They Get |
|---|---|---|---|
| Free | $0/mo | Developers, hobbyists | Limited agents, capped LLM usage, default skills only. Enough to evaluate the platform and build a proof of concept. |
| Pro | $9.99/mo | Power users | More agents, stronger LLM model access, expanded integration support. The daily-driver tier for individuals who rely on their swarm. |
| Business | $29.99/mo | Teams | Larger swarms, advanced integrations, team-level coordination, and priority support. Designed for small teams collaborating through shared agent workflows. |
The free tier serves two purposes: it removes friction for developer adoption, and it generates usage data that improves the platform for paying customers.
The Pro-to-Business upgrade path is driven by team features and integration depth. When a user needs their swarm to coordinate with teammates, they upgrade.
IaaS — Enterprise Infrastructure (Contract-Based)
The IaaS tier is the high-value play. Enterprises deploy the complete Crawbl stack into their own infrastructure under annual contracts.
| Revenue Type | Description |
|---|---|
| Enterprise licenses | Annual contracts for platform deployment rights. Priced per-cluster or per-seat depending on deployment model. |
| Deployment fees | One-time or phased fees for integration, customization, and migration services. Covers the work of deploying Crawbl into a customer's existing cloud environment. |
| Support contracts | SLA-backed tiers (standard, premium, dedicated) with defined response times, escalation paths, and dedicated engineering support at the highest tier. |
| Custom runtime integration | Professional services for integrating proprietary or specialized runtimes beyond the standard ZeroClaw/PicoClaw/NanoClaw/OpenClaw lineup. |
Enterprise IaaS solves the top three enterprise objections to AI platform adoption:
Data residency
The platform runs in the customer's infrastructure, so data never leaves their control.
Vendor lock-in
Cloud-agnostic Kubernetes deployment keeps customers from depending on a single cloud provider.
Compliance
Enterprise isolation, RBAC, audit logging, and BYOK support cover the main security and regulatory requirements.
The IaaS model also creates a natural upsell path: enterprises start with a single-cluster deployment, then expand to multiple clusters, regions, or business units as they see results.
SaaS — Platform Services (Marketplace)
The marketplace layer creates network effects and a revenue-sharing ecosystem around the platform.
| Revenue Type | Description |
|---|---|
| Agent marketplace | Revenue share on third-party agent and skill sales. Developers build agents and skills; Crawbl handles distribution, billing, and trust verification. |
| Paid skill packages | Premium first-party integrations sold as add-ons. Enterprise-grade connectors for specialized systems (SAP, Salesforce, ServiceNow). |
| Platform commission | Percentage on all marketplace transactions. Standard marketplace economics — the platform captures value from facilitating the exchange. |
The marketplace is not planned until Phase 3 (skills) and Phase 4 (agents). This is deliberate: the trust model, sandboxing infrastructure, and verification pipeline must be production-proven before third-party code runs in user swarms.
Unit Economics
The cost structure of Crawbl is designed around ZeroClaw's extreme efficiency. Traditional AI agent platforms run Python or Node runtimes that consume hundreds of megabytes per agent. ZeroClaw runs at approximately 5 MB per agent, enabling multi-tenancy densities that are orders of magnitude higher.
| Metric | Value | Impact |
|---|---|---|
| Agent memory footprint | ~5 MB per agent | Hundreds of agents per node instead of tens |
| Scale-to-zero | 10 ms cold start | Idle agents cost nothing; resources are reclaimed immediately |
| LLM cost target | ~$5/mo per paying user | Platform-managed cost controls keep inference costs predictable |
| Inference model | Hosted provider APIs (Anthropic, OpenAI, Gemini) | No GPU infrastructure to manage; costs scale linearly with usage |
| Enterprise revenue | Annual contracts | Predictable recurring revenue with high retention |
The unit economics work because the infrastructure cost per user is radically low.
A single Kubernetes node that would host 10-20 Python-based agents can host hundreds of ZeroClaw agents.
This density advantage compounds: lower costs enable a more generous free tier, which drives adoption, which generates data, which improves the platform, which justifies higher enterprise contract values.
Scale-to-zero is the other critical lever.
Most agent platforms charge for idle compute because their runtimes cannot cold-start fast enough to be practical.
ZeroClaw's 10 ms cold start means agents can be fully suspended when inactive and restarted instantly when needed. Users experience no latency penalty, and the platform eliminates idle resource waste.
Growth Strategy
The growth strategy follows a four-stage flywheel where each stage generates the momentum for the next.
Consumer app validates the platform
The mobile app is the proving ground. Real traffic shows whether the orchestrator, runtime engine, integration layer, and cost controls hold up in production.
Key metrics: onboarding completion, daily usage, swarm provision time, LLM cost per user, and integration reliability.
Developer PaaS attracts volume
Once the platform is hardened and a public API exists, developers become the volume channel. Lower hosting costs and fewer moving parts are the wedge.
Key metrics: developer signups, API call volume, time-to-first-agent, and 30/60/90-day retention.
Enterprise IaaS captures contracts
Enterprise sales start once the platform has a reliability story. The pitch is simple: the same stack developers use, but deployed in the customer's own infrastructure with an SLA.
Key metrics: enterprise pipeline value, contract size, deployment success rate, and expansion revenue.
Marketplace and federation create network effects
Skills, agents, and federation connections make the network more valuable over time. The moat is not just data; it is the relationships and workflows that span organizations.
Key metrics: marketplace transactions, third-party developer revenue, federation connections per enterprise, and cross-org workflow completion rates.