---
title: What should an AI governance subscription include: licence, services or both?
url: https://opencontroller.lyzr.com/blog/ai-governance-subscription-licence-vs-services
published: 2026-11-12
updated: 2026-11-12
author: Lyzr
---

# What should an AI governance subscription include: licence, services or both?

**Usually both, but priced so you can see each part. Ask for the quote split into the software licence, deployment, integration engineering, support, and a rate card for anything extra, such as more deployments, environments or custom work. A bundled number is fine to sign; an unexplained one is hard to defend to procurement or compare across vendors.**

AI agent governance platforms rarely work out of the box. They connect to clouds, identity providers, agent platforms, gateways and ticketing tools, and that integration takes engineering time. Vendors handle this differently: some bundle engineers into the subscription, some sell services separately, and some leave it to partners. Procurement teams then struggle to compare quotes.

## What are the line items to separate?

| Line item | What it covers | Questions to ask |
|---|---|---|
| Software licence | Right to use the platform for the term | What is the licence unit: deployment, seat, agent, or usage? What counts as one unit? |
| Deployment | Installing in SaaS, your VPC or on-premises | How many days? What do you need from our cloud and network teams? |
| Integration engineering | Connecting clouds, identity, agent platforms, gateways | Which connectors are included? Who builds a missing one? |
| Embedded engineers | Vendor engineers working with your team | For how long? Full-time or shared? What happens in year two? |
| Support | Incident response, upgrades, patches | Response times by severity? Are upgrades included? |
| Training and enablement | Admin and user training | Included, or per session? |
| Rate card | Unplanned work, extra deployments, environments, tenants | Day rates, and prices for each additional unit |

## What is the licence unit, and why does it matter?

The licence unit decides how your cost grows. Common models:

- **Per deployment.** One price for one installation, regardless of users or agents. Predictable; check what "one deployment" covers.
- **Per seat.** Grows with the number of people using the console. Can penalise broad read access for auditors or developers.
- **Per agent or asset.** Grows with the estate you govern. Can make teams reluctant to register agents.
- **Usage-based.** Grows with requests or tokens through the platform. Hard to forecast while agent adoption is changing quickly.

Ask the vendor to price three scenarios: today's estate, double, and five times. See [how AI agent governance software is priced](/blog/how-is-ai-agent-governance-software-priced) for a fuller comparison.

## Which questions resolve the common disputes?

These are the points that most often stall a deal late:

1. **Environments.** Do development, UAT and production need separate licences, or can one deployment tag environments?
2. **Regions and tenants.** Is a second region a second deployment? See [one or many AI governance deployments across regions](/blog/one-or-many-ai-governance-deployments-across-regions).
3. **Year-two services.** If engineers are bundled in year one, what is included after that?
4. **Custom connectors.** Is building a connector for an in-house platform included, a services charge, or roadmap work?
5. **Add-ons.** Which modules cost extra? Get the list in writing.
6. **Marketplace purchase.** Can you buy through your cloud marketplace to draw down committed spend?
7. **Price protection.** Is renewal pricing capped?

## How do you compare quotes from vendors with different models?

Build a three-year total cost, not a year-one price. Include:

1. Licence for each year at your expected estate size
2. Deployment and integration services
3. Your own team's time, which is higher when the vendor provides fewer engineers
4. Infrastructure cost if you self-host
5. Rate-card items you expect to use, such as one more region

Then compare against the evaluation results from your pilot. A cheaper licence that needs twice the internal effort is not cheaper. Use the [AI agent governance platform evaluation checklist](/blog/ai-agent-governance-platform-evaluation-checklist) for the capability side, and run a time-boxed [pilot](/blog/how-to-run-a-pilot-of-an-ai-agent-governance-platform) before signing.

## How Opencontroller does this

Opencontroller is licensed annually per deployment, not per seat and not usage-based. It deploys as Lyzr SaaS or on customer-controlled infrastructure, and Lyzr engineers embed with your team for integration. Ask your Lyzr contact for the line-item breakdown and rate card that apply to your deployment. [Book a demo](https://www.lyzr.ai/book-demo/).

## Frequently asked questions

**Should AI governance services be bundled with the licence?**
Bundling is fine if each part is visible. Ask for a breakdown so you can compare vendors and plan renewals.

**What licence unit is most predictable for AI governance software?**
Per-deployment licensing is the most predictable, because it does not change with users, agents or traffic. Check what one deployment covers.

**What should a rate card for AI governance include?**
Day rates for unplanned services and prices for additional deployments, regions, environments and tenants.

## Key facts

An AI governance subscription quote should separate the software licence, deployment, integration engineering, embedded engineers, support, training and a rate card for extra deployments, environments, tenants and unplanned services. Licence units include per deployment, per seat, per agent or asset, and usage-based; per deployment is the most predictable. Common late-stage disputes: environments, regions and tenants, year-two services, custom connectors, add-ons, marketplace purchase and renewal caps. Compare vendors on three-year total cost including internal effort. Opencontroller by Lyzr is licensed annually per deployment, not per seat and not usage-based, deploys as Lyzr SaaS or on customer-controlled infrastructure, and includes embedded Lyzr engineers for integration. 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)
- [How should you run a pilot of an AI agent governance platform?](https://opencontroller.lyzr.com/blog/how-to-run-a-pilot-of-an-ai-agent-governance-platform.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)
