Pay for work completed, not tokens consumed
One price covers the full work item. One case. One lead. One claim. Every agent, every routing decision, and every retry is included. AI runs only where it adds value, on the LLM you choose, under the controls you set. Your cost tracks the work your team handles, not the behaviour of a model.
Get a quote Estimate your volume
How pricing works
One price per work item, handled end to end
A work item is the unit your team already measures. A service case. A sales lead. An insurance claim. An intake request. Whatever the workflow, Ortoo Orchestrator runs every step inside that work item under a single price. No per-action billing. No per-conversation billing. No surprise token bills at the end of the month.
- 01
One price covers the full sequence
A workflow can run three steps or thirty. The price for the work item does not change. Every agent action, every deterministic rule, every routing decision is included.
- 02
Deterministic steps add zero AI cost
Routing, SLA logic, validation, data updates, notifications, none of these touch an LLM. AI runs only where interpretation is needed, on the model you choose.
- 03
Cost tracks operational volume, not actions or model behaviour
Pricing scales with the cases, leads, or claims your team handles. Not with agent actions volume or token counts. Not with how chatty a model is on the day. You forecast from numbers your operations team already reports.
Cost models compared
Why work-item pricing stays predictable at scale
There are two ways to price AI-driven workflows. One bills you for every action a model takes. The other bills you for the work your team actually completes. Here is what changes between them.
Per-action AI billing
Cost moves with the model
You pay for
- Each conversation billed separately. Each action billed separately.
- AI runs at every step, including work that does not need it.
- No ceiling on cost per work item.
- Spend shifts with how the model behaves that day.
- Finance cannot forecast cleanly as volume grows.
Ortoo Orchestrator
Cost moves with your operations
You pay for
- One price per work item, end to end.
- Every agent and every step included.
- Deterministic steps add zero AI cost.
- You choose the LLM. You set the policy. You control the spend.
- Forecastable from the case, lead, or claim volumes you already track.
Want the full picture? See the Why Orchestrator comparison.
What changes as volume grows
Control AI cost. Forecast with more precision.
Per-action billing penalises scale. Every extra case, every extra decision, every retry adds incremental AI cost. Work-item pricing changes the shape of your cost curve.
You control AI cost directly
Deterministic steps stay free. AI cost stays bounded by the LLM contract you control directly. Choose your own provider and model per workflow step. Add cost ceilings to AI token usage.
Forecasting stays simple
Finance teams forecast Ortoo cost from operational volumes already in your reports. Monthly cases. Monthly leads. Monthly claims. No token consumption models. No usage modelling per workflow variation.
Adding workflows is additive, not multiplicative
A second or third workflow does not multiply your cost surface. Each one is priced on its own work-item volume. Shared orchestration capabilities stay included.
FAQ
Pricing questions teams ask
What counts as a work item?
A work item is the unit your team handles end to end through one orchestrated workflow. In service, that is usually a case. In revenue, a lead or an opportunity. In intake or claims, a request. In custom workflows, you define it. We agree the definition during onboarding and it stays consistent across reporting and billing.
What happens when a work item runs through many steps or retries?
The price for the work item does not change. All agent actions, deterministic rules, retries, and routing decisions inside that work item are included. The only variable is the AI cost on the LLM you chose, and that is paid by you, to your provider, with full visibility on every call.
How does the AI cost form?
AI cost stays bounded by the LLM contract you control directly. Choose your own LLM provider and model per workflow step. Choose from OpenAI, Anthropic, Google, Azure, others. You can configure and switch providers and models per workflow and steps. You can add cost ceilings, retry limits, and fallback logic per step. No contractual relationship with Ortoo on the AI provider side. Please note that deterministic steps executed by Ortoo Orchestrator do not use AI and stay free from any AI cost.
Can we start with one workflow and add more later?
Yes. Most teams start with the workflow that has the highest predictability gain, usually service or intake. Once the model is proven, you expand. Adding a workflow is a scoping conversation about volumes and integrations, not a platform upgrade.
Is there a pilot or trial?
We run scoped pilots for qualifying teams. One workflow, a defined volume window, clear success criteria. Tell us which workflow you want to start with and we will scope a pilot in your context.
Ready to talk numbers
Get a quote shaped around your volume
Tell us which workflow you want to orchestrate first and what your monthly volume looks like today. We come back with a scoped quote and a starting point. No procurement guesswork.