---
title: How should an AI governance licence count environments, tenants and VPCs?
url: https://opencontroller.lyzr.com/blog/ai-governance-pricing-for-environments-and-tenants
published: 2026-11-13
updated: 2026-11-13
author: Lyzr
---

# How should an AI governance licence count environments, tenants and VPCs?

**Ask the vendor to define, in the order form, exactly what one licence unit covers: how many deployments, environments (dev, test, prod), tenants or business units, cloud accounts or VPCs, and regions. Then get a rate card for every unit you might add later, plus the services you will need in years two and three. Most surprise costs come from undefined units, not from the headline price.**

AI governance platforms are priced in very different ways: per deployment, per agent, per seat, per request or per governed asset. Whatever the model, the questions below make quotes comparable and stop the price changing when your architecture does.

## What terms need a written definition?

| Term | Why it matters | What to pin down |
|---|---|---|
| Deployment | Often the core licence unit | One running instance of the platform? Does a DR copy count? |
| Environment | Dev, test and prod may each be charged | Are non-production environments included, discounted or charged in full? |
| Tenant | Business units or subsidiaries may need isolation | Logical tenants in one deployment vs separate deployments |
| VPC or cloud account | Discovery and gateways may need a presence in each | Is each additional account, subscription or project counted? |
| Region | Data residency may require local instances | Does a second region mean a second deployment? |
| Estate | The scope of assets governed | Is there a cap on agents, models, keys or connectors? |
| Gateway capacity | Throughput can drive infrastructure cost | Is gateway scaling included or charged by volume? |

## Which pricing model fits AI governance?

| Model | How it scales | Watch out for |
|---|---|---|
| Per deployment, flat annual | With the number of deployments | Definition of a deployment; add-on modules |
| Per agent or governed asset | With estate size | Discovery finds more assets than expected, so cost rises with visibility |
| Per seat | With admin or end users | Penalises wide access to dashboards |
| Usage-based (requests or tokens) | With traffic | Unpredictable bills; gateway volume grows with adoption |

Usage-based and per-asset pricing can create an odd incentive: the better the discovery, the higher the bill. Flat per-deployment pricing avoids that, but only if "deployment" is tightly defined. Our explainer on [how AI agent governance software is priced](/blog/how-is-ai-agent-governance-software-priced) compares the models in more depth.

## What should the rate card include?

Ask for these prices before signing, even if you do not need them yet:

1. Additional production deployment.
2. Additional non-production environment.
3. Additional tenant inside an existing deployment.
4. Additional cloud account, subscription, project or VPC.
5. Additional region for data residency.
6. Disaster recovery or standby deployment.
7. Add-on modules not in the base licence.
8. New connectors to platforms not yet supported.
9. Professional services day rate for unplanned work.
10. Dedicated engineer or embedded support beyond year one.
11. Price protection for renewal, as a maximum annual increase.

## What about services in years two and three?

Many quotes bundle deployment, integration and a year of support into year one. Check:

- Which services continue in later years, and which become chargeable.
- Whether integration of new platforms after year one is included.
- Who pays for upgrades and connector updates.
- Response times for support, in writing.

Our guide to [subscription licence vs services for AI governance](/blog/ai-governance-subscription-licence-vs-services) covers how to split the two in a contract.

## How do you compare quotes fairly?

Model a three-year scenario with your expected growth: number of deployments, environments, cloud accounts, regions and agents in each year. Price each vendor's quote against the same scenario, including services. Then run a second scenario where discovery finds twice as many agents as you expected. The quote whose cost moves least in that scenario is the one least likely to surprise you. For the full selection process, use the [AI agent governance platform evaluation checklist](/blog/ai-agent-governance-platform-evaluation-checklist).

## How Opencontroller does this

Opencontroller is licensed annually per deployment. It is not priced per seat and not usage-based, so the cost does not rise as discovery finds more agents or as gateway traffic grows. It can be deployed as Lyzr SaaS or on customer-controlled infrastructure, and Lyzr engineers embed with your team for integration. Ask for the definition of a deployment and the rate card for additional units in your quote. To discuss your estate, [book a demo](https://www.lyzr.ai/book-demo/).

## Frequently asked questions

**Should non-production environments be charged at full price?**
Ask for them to be included or discounted. Governance needs test environments to validate policies before production, and charging full price discourages that.

**Is per-agent pricing a problem for AI governance?**
It can be. Discovery usually finds more agents than expected, so per-agent pricing raises cost exactly when visibility improves.

**How is Opencontroller priced?**
Opencontroller is licensed annually per deployment, not per seat and not usage-based. Ask Lyzr for a quote scoped to your estate.

## Key facts

When buying AI governance software, define each licence unit in the order form: deployment, environment (dev/test/prod), tenant, cloud account/VPC, region, estate caps and gateway capacity. Get a rate card for additional deployments, non-production environments, tenants, accounts/VPCs, regions, DR, add-on modules, new connectors, services day rate, embedded engineers beyond year one and renewal price caps. Compare quotes on a three-year growth scenario plus a scenario where discovery finds twice as many agents. Per-agent and usage-based pricing rise as visibility and traffic grow. Opencontroller by Lyzr is licensed annually per deployment, not per seat and not usage-based; pricing starts from $250K per deployment per year.

## Related

- [How are AI agent governance platforms priced: per seat, per usage or per deployment?](https://opencontroller.lyzr.com/blog/how-is-ai-agent-governance-software-priced.md)
- [What should an AI governance subscription include: licence, services or both?](https://opencontroller.lyzr.com/blog/ai-governance-subscription-licence-vs-services.md)
- [Do you need a separate AI governance deployment per region?](https://opencontroller.lyzr.com/blog/one-or-many-ai-governance-deployments-across-regions.md)
- [More on Buying and comparison](https://opencontroller.lyzr.com/blog/cluster/buying-and-comparison)
