AI consulting + implementation for small businesses
Everybody says put AI in your business. Nobody says where.
I find the places AI actually pays off in a small business, install the systems, and hand you the keys.
Best fit: owner-run businesses that want working systems, not a slide deck.
Real work
Named clients. Running systems.
Not stock photos of handshakes. Work you can go look at.
Marketing Director. I run direct response for one of the best-known names in online business education: sales pages, email campaigns, and multiple millions of dollars in profitable Facebook ads. And I build the AI systems behind it all: ad generators, competitor monitors, research pipelines that run every week without me touching them.
Sales pages, product marketing, and photography for their park-model cabin lines and RV park. Small local business, real inventory, real buyers. The kind of company this whole page is for.
A phone line my Amish neighbors call to get their questions answered. No internet on their end, an AI system on mine. Built it, run it, maintain it. If I can make AI useful for people who don't own computers, I can make it useful for your shop.
The tools that run my own businesses: a competitor-ad monitor that reports every Monday morning, automated invoicing, and listing pipelines for an online store of handmade goods. I eat my own cooking.
The first step
One fixed price. One week. Then you decide.
Every engagement starts the same way, and it isn't a contract. It's a work order.
Fixed. In writing. No hourly meter running.
Where would AI actually pay off for us?
Can't I just use ChatGPT myself?
Who owns what you build?
Will my customers know it's AI?
What does this actually cost?
Every engagement starts with the questions you're already asking.
Straight answers
- Can't I just use ChatGPT myself?
- Yes, often. If it's one small, contained task, do it yourself, today, for free. Bring me in when it has to work the same way every time, for the whole crew, without you babysitting it.
- Is this just a wrapper around ChatGPT?
- I build on the same AI models anyone can buy, the way your electrician buys wire from the same supply house you could. What you're paying for is knowing where it pays off, a clean install on your real workflows, and it still working in month three.
- Who owns what you build?
- You do. Code, accounts, logins, documentation, all of it, handed over. If we part ways tomorrow, everything keeps working and everything is yours. No lock-in, no monthly hostage.
- What about my customer data?
- Rules in writing before anything connects: what the system can see, where it lives, who can pull the plug. Your access, revocable any day. And your customer list doesn't get pasted into public tools, period.
- Will callers know it's AI? Can they reach a person?
- Yes and yes. Anything customer-facing announces itself honestly and keeps a clear path to a human. I build these so nobody feels tricked, starting with your customers.
- What if AI just isn't right for my business?
- Then the assessment says so, and if it can't find you at least five hours a week worth taking back, you get the $999 back. Some of the most useful reports I write end with "wait, and here's what to watch for."
Built and running
Nothing on this list is a demo.
Every one of these started as a job I had to do. Some run on a schedule without me. Four of them you can download and run yourself.
My neighbors call a phone number and get their question answered. No internet on their end, an AI system on mine, and I keep it running.
Before the marketing team I direct starts the workday, a run has already read my sources and posted the briefing to Slack. It goes out every morning without me touching it.
A decade of client footage, about sixty thousand files deep in Dropbox, that nobody could search. Now every video is transcribed, indexed, and ranked, so one query finds the exact clip, and editing moves much faster. The same pipeline became a library of every story and hook in that archive, with timestamps, and it makes the sales pages and video scripts better.
It watches what competitors are running and hands me a report every Monday at 8:00 AM.
It writes and designs a complete long-form sales page as one self-contained HTML file, on a 34-step direct-response framework. I built it for the offers I run at work.
It mines competitor reviews and social comments across Reddit, YouTube, TikTok, Trustpilot, and Udemy, then reports what buyers actually complain about. Their words, not my summary of their words.
Hand it any business and it comes back with six complete Meta image ad concepts. Copy for every placement, plus image prompts that work in any AI image generator.
Feed it a sales page and it grades the page: what reads AI-written, what is legally risky around income claims, what conversion elements are missing. The fixes come back in priority order in a styled report.
It reads a marketing email against CAN-SPAM, the FTC rules on income claims, and the acceptable-use lines email platforms actually enforce. You see the problems before you send, not after.
I type one sentence, "invoice Acme 450 for the March shoot," and get a PDF invoice, a ledger entry, and a dashboard of what is still unpaid.
Four of these are yours if you want them. They are the actual files I use, with my private details taken out. No email gate, no catch. Take them here.
The math
Do the arithmetic before you buy anything.
Here's an example with made-up numbers, because I don't know your business yet. Say writing a quote takes 25 minutes and you send eight a week:
× 50 weeks = ~167 hours / year
× $85 / hour = a $14,000-a-year line item
Getting that first draft down to five minutes doesn't take magic. It takes somebody setting it up right once, on your templates, in your voice, connected to your price book.
Now run yours
Your numbers, your math. Not a promise, a measurement. The assessment does this for your whole business.
Find my numbersYour numbers will be different. That's the point of the assessment: measure first, build second, and skip anything where the arithmetic doesn't clear.
How it goes
Everything I build has to move one of three numbers: more customers in the door, more from each customer, or fewer hours burned on work a system should do.
The assessment
Find where AI pays off in your business, in writing, with the math shown. Work Order No. 001 above.
The build
I build the systems the assessment picked: follow-up that sends itself, quotes that draft themselves, a phone line that answers when you're on a job, paperwork that fills itself in, and past customers who haven't been back in a year, found and followed up automatically. Most builds take weeks, not months. You own everything: code, accounts, logins, documentation.
Before anything gets built, we fill in four blanks together: which of the three numbers it moves, the exact thing we'll measure, where that number sits today, and where it should sit after 60 days. If we can't fill in the blanks, I don't build it.
Only if useful
Two working sessions a month on whatever's next, plus a direct line to me with an answer inside a business day. Plenty of clients don't need it, because what I hand over comes with instructions and keeps working after I leave.
The next step is a call, not a contract.
Twenty minutes. Bring the part of your week you'd most like to stop doing by hand. I'll tell you honestly whether AI can take it off your plate, and what I'd charge to make that happen. Worst case, you leave with the names of one or two tools you'd never heard of, and we never talk again.