Odokai team member holding a notebook in an office

Pay for the Workflow.

Priced around your workflow.
From £500 a month, ex VAT.
See a simulation or example running in the first 45-minute call.

You pay for a workflow that runs in production, on your data, with your team trained to use it. You see it working before you spend anything. Once we've built one together, you have a fixed fee for the build and a monthly fee to keep it running, both in writing, before you commit to either.

What You Pay For

See it work. Get it built. Keep it running.

The discovery call is free. The build of your first workflow is one fixed fee. After that, a monthly fee keeps the platform running for every workflow your team adds. Nothing else is billed unless you ask us for more.

  1. Step 1

    Discovery Call

    Free ~45 minutes

    Bring a handful of your own documents. On the call we build an agent that handles one of your workflows, live, on your data. You leave with a working agent and a clear view of what production-ready looks like for your team. Your quote is built on that.

  2. Step 3

    Platform Fee

    From £500 a month, ex VAT

    Everything your workflows run on once you're live: model usage up to a fair-use volume we agree with you, model upgrades as they ship, the infrastructure, and the support. Your team builds the next workflows on it without paying for another build.

    From £500 a month on the shared platform, or £1,000 a month for your own instance. In your own AWS account, you use your own model accounts and pay the provider directly. Fair-use policy applies.

Where It Runs

Two ways to run it.

The platform fee depends on where your instance runs. Either way, your team builds the next workflows on it with no rollout fee.

Shared

Your team on Odokai, hosted and run by us alongside other customers. The fastest route to live, with no cloud infrastructure to stand up first.

  • Hosted alongside other Odokai customers
  • Build, run, and govern agents on your own data
  • Role-based access, full audit trail, policy controls
  • Model usage included up to a fair-use volume
  • Model upgrades included as they ship
  • Your team builds the next agents on it, no rollout fee

If keeping your own AWS environment matters, choose Dedicated instead.

Prices exclude VAT. The platform fee starts when your instance goes live.

What Shapes Your Quote

Your number depends on your setup.

Two firms the same size can need very different builds. We work through the factors below with you on the discovery call, so the quote fits the job in front of you.

  • The systems you use

    An inbox and a shared drive is a smaller build than a CRM, a case system and a finance package. Each connection is built once and reused by every workflow after.

  • Your checks and sign-offs

    How many ways the job can go wrong, and who needs to approve the output before it leaves the building. More checks mean a bigger build.

  • Where you want it to run

    Shared Odokai infrastructure is the lowest-cost route. Your own instance in your own AWS account, run by us or by you, costs more each month and keeps everything inside your walls.

  • How much goes through it

    Documents, cases or reports each month. This sets the model allowance in your monthly fee, so you are not paying for capacity you don't use.

  • Who's on it

    How many of your team will be building and running agents once the first workflow is live.

  • Whether you want us to build the next ones

    Your team can build the next workflows themselves. If you'd rather we did, each one is a small piece of work priced well below the first build, because the platform is already there.

Before You Book

What you'll want to know first.

When will I know what it costs?

The platform fee starts at £500 a month on the shared platform, or £1,000 a month for your own instance, ex VAT. The build fee for your first workflow comes in writing straight after the discovery call, alongside the acceptance criteria it is tied to. Nothing is committed until you say so.

What do I get for the build fee?

A workflow you can put in front of clients: edge cases handled, connected to your systems, every action logged, access controlled, your team trained. The platform and the governance it runs on are set up as part of it, once.

Can the fee change once you've started?

No. You sign off the acceptance criteria before kickoff and the fee is fixed against them. If you decide you want more once we've started, we scope and quote that separately, and you choose.

What does the monthly fee cover?

Model usage up to a fair-use volume agreed with you, model upgrades as they ship, the infrastructure, and the support to keep it running. If you go past the allowance, the overage rate is already in your contract. In your own AWS account, you use your own model accounts and pay the provider directly.

What about the second workflow?

Your team builds it on the same instance, using the connections and approval patterns already in place. No second build fee. If you'd like us to build it, it's a small piece of work.

What if we want to stop?

Your data and everything you built export in documented formats on request, on any plan, whether or not you renew. Source code escrow is available where you need it and arranged during contracting.

What does the discovery call cost me?

About 45 minutes and a few of your own documents. No charge, no obligation, and you keep the agent we build on the call.

We want advice, not the platform.

That's a separate engagement: fractional AI leadership on a retainer, embedded delivery on a day rate, or a short independent advisory piece. See Services for what each involves.

Why not just use ChatGPT, Copilot or Claude?

Our comparison sets all four side by side on cost, capability and security controls. Have a look.

Find out what yours would cost

Pick the job your team dreads most and bring a few documents. In about 45 minutes you'll see it working, and you'll have a number in writing shortly after.

From the Blog

Notes on governed models, operational automation, and how teams move from AI-assisted pilots to broader adoption.