Most of the people who call me have already decided they should be "doing something with AI." Almost none of them have decided what. That is the actual problem, and it is a smaller one than it looks — the order you do things in matters more than any single answer.
Here is the order, drawn from four short notes published this week.
Before you spend anything, try it for free
Most business owners I talk to don't need an AI strategy. They need their Monday morning back.
There is a prompt you can paste into ChatGPT, Gemini or Claude right now, with nothing installed and nothing signed up for: "Summarize my top 5 emails today into 3 action items with deadlines and owners."
That is the whole thing. It works because it is specific — five emails, three actions, owners and deadlines. Vague prompts get vague answers, which is why most people's first experiment with AI is disappointing and their second one never happens.
There are fifty-odd more like it in a free library on this site, sorted by marketing, customer service, operations, nonprofits and HR. No signup. If that is all you ever take from here, it was worth the visit, and it costs you nothing to find out whether this stuff is useful in your business before anyone quotes you for anything.
Then pick the task — and there is a test for it
This takes about ten minutes and needs no software.
Think back over your week and find the message you sent more than five times in nearly the same words. The "we got your request." The status update. The appointment reminder. The directions to the office. Most people find one without looking hard, and a lot of people find four.
That is the first thing worth automating. Not because it is clever — it isn't — but because you have already written it, already tested it on real customers, and already proven that it repeats. Everything before that point is guessing about what your business needs. This is the one part of the process where the evidence is free and already in your sent folder.
The temptation is to start with the impressive thing instead. The impressive thing is usually the one nobody can describe precisely, which is why it never gets finished.
Know what you are already paying for
The other half of the picture is the bill you already have. A website builder, a scheduler, a chat widget, an email tool and a form builder are five separate bills and five separate logins doing one job: somebody lands on your site, you find out who they are, you follow up.
Worth saying plainly, because it cuts against my own interests: plenty of what you pay for earns its money and should be left alone. Accounting software, payroll, bookkeeping, the industry-specific system your whole trade runs on — none of that is a candidate. If the tools you have are working, the honest advice is to keep paying for them.
What the first conversation should actually be
Fifteen minutes is an odd length for a sales call. That is deliberate. It is long enough to work out whether there is something worth building and too short to turn into a pitch.
The shape of it: you describe one repetitive task. You get told whether it is worth automating. Sometimes the honest answer is no, or not yet, or that you already own a tool that does this and nobody showed you the setting. If it is worth doing, you get a flat number and a written scope before any work starts — a price agreed up front, so there is no invoice to be surprised by.
Nothing is installed, signed or paid for on that call.
Ask the ownership question before the price question
This is the one most people skip, and it is the expensive one.
Plenty of automation gets built inside a vendor's own platform. It works — and it works exactly as long as you keep paying. Stop, and there is nothing to take with you.
So ask three things. Where does it live? Whose account does it run in? What am I left holding if the two of us never speak again?
Either answer can be fine. A rented system with a good price and an honest description of what it is can be the right buy. Buying one and assuming the other is not.
For what it is worth, the work here gets built in your accounts, on tools you already use, and you own what gets built — including the accounts it runs on. One workflow, live in about a week, at a flat price starting at $495. The bar it has to clear before it is called done is that it is simple enough for anyone on your team to use without you.
That is the whole order: try it free, find the message you keep retyping, look at what you already pay for, spend fifteen minutes on it, and ask who owns the result before you ask what it costs.
If you want to walk through your own version of that list, the services and prices are here: tech-d-enterprises.com/services
And the fifteen minutes is here, free, with nothing to sign: tech-d-enterprises.com/appointments