Priced in the open. Scoped in writing.

Because a free consultation is a sales call, and it produces what sales calls produce: a proposal shaped like whatever the consultant already wanted to sell you.

A real assessment means interviewing the people who do the work, mapping systems that were never documented, and quantifying opportunities against your actual volumes. That's a week of engineering effort. Charging for it means the findings belong to you regardless of what happens next, including the finding that you shouldn't build anything yet, which is not a conclusion a free consultation has ever reached. For what other firms publish, checked at the source and set beside these numbers, see what AI consulting costs in 2026.

No, and that's deliberate. Crediting it would quietly turn the assessment back into a sales call, because I'd have a financial stake in recommending a build.

The assessment is a standalone deliverable with standalone value. If the honest recommendation is to fix a process rather than automate it, or to wait six months, I still did the work and you still have a report you can act on without me.

Because the model is the easy part. The majority of the work in a production AI system is integration into systems that weren't designed to be integrated with, plus the verification and monitoring layers that tell you whether it's still working.

Those layers are exactly what gets cut to hit a lower number, and cutting them is the single most common reason AI pilots never reach production, or reach it and quietly degrade. A build priced below this is either a prototype or a system nobody is watching.

No. The assessment gives you a ranked list of separately scoped, separately priced projects. You buy the smallest one, and it has to prove its value in your numbers before we talk about the next. I re-earn each phase. That's the deal.

What you're never asked to buy is a fraction of something bigger. Each build is a complete, running system: integrated, verified, monitored, and yours. So if you stop after the first one, you still own a system that pays for itself. Your exposure at any moment is one small build, not the whole vision.

No. It's month to month. Cancel whenever and you keep the system, the accounts, the documentation, and the evaluation harness.

The retainer has to be re-earned every month. If I have to trap you in a contract to keep you, I've already failed at the part I'm claiming to be good at.

You do. Accounts are created in your name during onboarding, credentials are yours, and every build ships with documentation and a handoff session for your team.

No proprietary runtime, no black-box middleware, no hostage situations. If you want to take it in-house or hand it to another firm, you can, and I'll help with the transition.

Fair question, and the honest answer is that it's a real risk you should price in. Here's what mitigates it: everything runs in your accounts, not mine. Documentation ships at delivery, not on request. There's no proprietary layer that only I can maintain, and no credential only I hold.

A system I built can be picked up by any competent engineer, which is a deliberate design constraint. What you'd lose is the operator, not the system.

As engineering requirements, from the design stage. Prompt injection defense, credential handling, tenant isolation, least-privilege access, and audit logging are scoped into the build rather than added before launch.

For regulated environments, I design to the compliance constraints that apply to you and document the controls so they survive a review. I'll sign your NDA and work inside your access policies rather than asking for exceptions.

Billed to your accounts, in your name, visible to you in full. My pricing covers my work: the diagnosis, the build, the operations, and the reporting.

I'll estimate the run-rate before you commit, and I'd rather you see those invoices directly than have me mark them up quietly. Part of what the retainer buys is moving you onto cheaper models when cheaper models are good enough.

One assessment at a time. Two weeks is my attention, not my calendar, and running two at once would make both worse. Operate is capped at eight systems under management for the same reason.

The two-week clock starts from the latest of three things: countersignature, payment, and the access described in the scope being in place. It does not start from our first conversation, and the most common reason an assessment runs late is the third one.

I'm not going to publish a lead time here, because it would be stale the week after I wrote it and a stale number is worse than none. Ask me and you'll get a real date. If I can't start when you need me to, I'll tell you that instead of holding the slot.

Yes. I'm based in Lindale and work regularly in Tyler, Longview, and Dallas, and I like being able to sit in the room during discovery. But the work is remote-first and the build is identical anywhere.

For engagements outside driving distance, the assessment interviews happen by video and I'll travel on site if the scope warrants it.