What this actually costs
The AI audit is free. Builds are quoted against a defined scope before you commit, and ongoing support is a flat monthly fee. We do not publish fixed prices because the same request costs very different amounts depending on the systems it has to touch, so below is exactly what moves the number and how to get a real figure for your situation.
How engagements are structured
AI audit
FreeYou answer questions about your business and get a written automation plan for your industry and your tools. No call required first, and no obligation after.
Who it's for: Anyone working out whether automation is worth it at all, and where it would pay off first.
Build engagement
Scoped per project, quoted before you commitWe design and build the system: the agents, the integrations, the interface your team actually uses. Priced once against a defined scope, not hourly, so the number does not move while we build.
Who it's for: Teams with a specific workflow that is costing real hours every week.
Ongoing support
Flat monthly feeHosting, monitoring, model and API costs, fixes, and continued improvement once the system is live. Flat, so a busy month costs the same as a quiet one.
Who it's for: Anything running in production. Software that nobody maintains stops being an asset.
Products
Per-seat or per-agency subscriptionProducerHQ and Book Analytics are products rather than builds, so they carry standard subscription pricing instead of a scoped quote.
Who it's for: Independent P&C insurance agencies.
What actually moves the number
Five things, in roughly the order they matter. If you can answer these, you can estimate the shape of an engagement before you speak to anyone.
How many workflows
One well-defined workflow is a materially smaller engagement than five. The most common way a project gets expensive is scope that grows during the build, which is why we scope and quote before starting rather than billing hourly.
What it has to connect to
Systems with a real API are straightforward. Systems without one need a different approach, and that difference is usually the single largest line in a quote. If your stack is modern and documented, you are at the cheaper end.
How much judgement is involved
Moving structured data between systems is well-trodden. Reading unstructured documents and making a call a human would otherwise make costs more to build and more to run, because it needs evaluation and a human-in-the-loop path.
Volume, for the monthly fee
The build cost does not change with volume. The ongoing fee does, because model and API costs scale with how much you run through it. A team of five and a team of eighty do not pay the same monthly.
Whether it touches regulated data
Healthcare and insurance work carries requirements around retention, auditability and access that add real engineering. Worth saying up front rather than discovering it mid-project.
Pricing questions, answered
Is any of it free?
+
Yes. The AI audit is free and has no obligation attached. You answer questions about your business and tools, and you get a written automation plan back covering where AI would pay off first, what it would take, and what to do in what order. Nothing else on this list is free: builds and ongoing support are paid engagements.
How much does AI automation cost?
+
It depends on scope, and we quote it before you commit rather than billing hourly. The variables that actually move the number are how many workflows are in scope, whether your systems have usable APIs, how much human judgement the AI has to replicate, your monthly volume, and whether regulated data is involved. The fastest way to get a real figure for your situation is the free AI audit, which produces a scoped estimate.
Why do you not publish fixed prices?
+
Because the same request costs very different amounts depending on your systems. Automating document intake for an agency on a modern platform with an open API is a fraction of the work of the same automation on a system with no integration path. Publishing one number would be wrong for most people who read it. We quote against a defined scope before any work starts, so the figure you agree is the figure you pay.
How do you charge: hourly, fixed, or retainer?
+
Builds are fixed-price against a defined scope, so the cost does not drift while we work. Ongoing support is a flat monthly fee covering hosting, model and API costs, monitoring and continued improvement. We do not bill hourly, because hourly billing rewards the wrong thing.
What is the smallest engagement you take on?
+
A single well-defined workflow that is costing a team real hours every week. That is deliberately the starting point: it proves the approach on something measurable before anyone commits to a larger programme. Most of our longer engagements began that way.
How long before it is live?
+
Most builds go live in two to three weeks from a defined scope. Longer for work involving regulated data or systems without an integration path, which is exactly why scoping happens before quoting.
Do I need to switch systems to work with you?
+
No, and we would push back if you suggested it. Our work sits on top of the systems you already run, using the exports and APIs they already produce. Replacing your system of record is a much bigger and riskier project than automating the work around it.
Get a real number for your situation
The audit is free and produces a scoped plan. If it turns out automation isn't worth it for you, that is a useful answer too.