AI readiness audit
Under $1K
We map your workflows, data, and risks, then rank what is worth building.
You get: A ranked list, costs, and a first project you could start next month.
About 1 to 2 weeks
Pricing
What custom AI actually costs
The short answer
Most projects land in one of four bands: a readiness audit under $1K, a single workflow at $2K to $10K, an agent with integrations at $6K to $15K, and a platform or data mirror priced by data size. Every project is different, so every project gets custom pricing, and you get a fixed quote before work starts.
Single workflow
$2K to $10K
One automation, assistant, or dashboard, connected to your real systems.
You get: Working software in production, documented, with you owning the accounts.
2 to 6 weeks
Agent with integrations
$6K to $15K
An agent that takes actions across systems, with permissions and an audit log.
You get: Practice mode, an approval queue, then measured autonomy.
3 to 8 weeks
Platform or data mirror
Custom pricing
Multi-system software, or your operating data mirrored into a database you own. Priced by data size and the number of systems.
You get: A system your business can run on, delivered in phases.
4 to 6 weeks
Every project is custom. Every project is different, so every project gets custom pricing. These ranges are where most projects start, and you get a fixed quote after a discovery call.
Why ranges and not one number
Four things move the price
- Integrations
- Each system you connect adds build and test time. Two is normal. Eight is a different project.
- Data quality
- Clean, accessible data is cheap. Data trapped in a vendor tool or a PDF is not.
- Autonomy
- An agent that drafts for a person costs less than one trusted to act alone.
- Compliance
- Health, legal, financial, and texting rules add review work. We would rather price that in than skip it.
Run your own numbers
Compare the build to the work it replaces
These are your numbers, not a promise from us. Fifty working weeks. The share is the assumption that matters, so we measure it in the audit before anyone commits.
1 square = 1 hour a week · 36 hours of repeat work
- Hours back per week
- 22
- This work costs, per year
- $57,600
- Recoverable, per year
- $34,560
Questions
Pricing, plainly
How much does an AI agent cost?
A business-grade agent with integrations usually falls between $6K and $15K with us. A narrow single-purpose agent can cost less, and a multi-system platform is priced by scope. Published guides range from about $3K to $450K because they are pricing very different things.
What makes the price go up?
The number of systems to connect, how messy the data is, how much the system can do without a person, and any compliance requirements. A model call is usually the cheap part. Integration and testing are the expensive parts.
What are the ongoing costs?
You pay for model usage and hosting directly on your own accounts, so there is no markup hiding in them. Optional management plans cover monitoring and changes, and they are quoted with your scope.
Who owns what you build?
You do, outright and forever. The code, the repositories, the accounts, and the data are yours from day one. Move it or host it anywhere you like. We offer management plans to run and improve it, but you never need us to keep it running.
Do you do fixed-price work?
Yes. You get a fixed quote before work starts. If scope changes, we quote the change before doing it. Retainers are available for ongoing work.
Is there a bundle discount?
Yes. Combining two or more XL Marketing services, such as a build from us with a website from ResultsXL, automatically earns a bundle discount.
Get a fixed quote
Thirty minutes with a solutions architect, and you will know which band you are in.