What Every Business Owner Should Ask Before Hiring Anyone for AI

TLDR: Small businesses consistently lose money in five specific workflows, and hiring the wrong AI consultant is how they compound that loss. Before any money changes hands, you need to know your own numbers and ask the right questions.

Summary:

There is a version of AI adoption that is boring, unglamorous, and genuinely profitable. It is not the version where someone sells you a transformation. It is the version where you fix one leaking workflow, measure the result, and move to the next one. The newsletter makes this case with unusual clarity, and I find myself agreeing with most of it.

The five workflows it identifies are not arbitrary. Speed-to-lead is the gap between a form submission and a human response, and most businesses sit at fourteen hours when the data shows under sixty seconds doubles close rates. Document processing means invoices, intake forms, and contracts that a person handles in fifteen minutes but a properly built system handles in two, running overnight without anyone watching. Follow-up sequences are the deals that die because nobody chased them, and personalisation that stops the moment a prospect replies is what makes the difference. Database reactivation is money sitting in every CRM in the form of dormant contacts, and the gym example pulling mid five figures from four thousand lapsed members with zero ad spend is the kind of number that makes you want to open your own CRM right now. Internal reporting is the Friday evening work that could land in your inbox before breakfast.

What I appreciate here is the self-diagnosis framing. Before you talk to any consultant, sit down with last month's numbers. Write out your response time to new leads, the hours your team spent copying data between systems, how many past customers you have not contacted in ninety days, who builds your weekly reports by hand, and where your team complains most about manual work. If you cannot answer those questions in writing, no builder can help you because there is nothing to measure yet.

The harder point the newsletter makes is about screening consultants. The market in 2026 is full of people who pitch well and cannot ship. The consultant who loses is the one selling AI transformation. The one who wins is the one who says they can get your response time under five minutes by Friday. That specificity is how you separate someone who understands your problem from someone who is selling you their solution.

The paywalled section apparently contains the actual interview questions, email scripts, and trial structure for engaging a consultant safely. What is freely available is enough to make you dangerous in the room, which is probably the point.

Key takeaways:

Five workflows cover almost every place service businesses bleed money: speed-to-lead, document processing, follow-up sequences, database reactivation, and internal reporting
You cannot evaluate an AI consultant without first knowing your own baseline numbers in writing
Specificity from a consultant is a signal of competence; vague transformation language is a red flag
Database reactivation against existing contacts often outperforms new ad spend for immediate revenue lift
Picking the right workflow first matters as much as picking the right builder

Why do I care: As someone thinking about tooling and systems, the framing here translates directly to how engineering teams should evaluate internal tooling proposals. The principle of measuring the bottleneck before choosing a solution is the same whether you are a gym owner evaluating a CRM automation or an engineering manager evaluating a new observability tool. The consultant-screening heuristic, favouring specificity over vision, also maps cleanly onto how I evaluate new frameworks or vendors. If someone cannot tell me what breaks when their tool fails, I do not trust them yet.

What Every Business Owner Should Ask Before Hiring Anyone for AI