Most businesses hiring an AI consultant for the first time have no idea what they’re buying. That’s not a knock — it’s a structural problem. AI consulting is new enough that there’s no shared vocabulary for what “good” looks like, and vendors know it.

The result is a lot of expensive slide decks, a lot of “AI maturity assessments,” and very little that actually ships.

Here’s what a real engagement looks like.

Week One Is Listening, Not Pitching

A consultant who shows up with a pre-baked solution on day one learned nothing about your business. The first week should be almost entirely intake: understanding your current workflows, where time is actually lost, which processes are rule-based versus judgment-based, and which pain points your team complains about most.

If your consultant is already suggesting tools before they understand your operations, that’s a bad sign. Good consultants are tool-agnostic until they understand the problem. Then they pick the right tool — not the one they already know how to demo.

The Deliverable Is Working Software, Not a Strategy Doc

There’s a place for written strategy. A one-pager that defines scope, success metrics, and rollout sequence is genuinely useful. But if your engagement produces a 40-slide deck and a “roadmap to AI adoption” with no working implementation, you bought the wrong thing.

At minimum, a productive AI consulting engagement should produce at least one thing that runs. A working prototype of an automated workflow. A tested prompt chain that handles a real task. A deployment you can actually use.

If your consultant can’t show you something running by the end of week three, ask why. “We’re still in discovery” is not an answer past a certain point.

They Should Know What They Don’t Know

The AI landscape changes fast enough that no one can be current on everything. A consultant who claims otherwise is either not paying attention or not being honest with you.

What you should hear instead: “I’ve built this specific type of integration several times” and “I haven’t worked with that tool in production, but here’s how I’d evaluate it.” Specificity and honesty are the signals. Vague confidence is not.

They Should Be Running AI Themselves

This is the one I’d put most weight on. A consultant advising you on AI workflows should be operating them, not just theorizing about them. Ask directly: How do you run your own business? What’s automated? What are you using in production right now?

At NexHam, we run multiple coordinated AI sessions daily. Automated operations run overnight. Client work ships without manual intervention at each step. When we recommend an approach to a client, it’s because we’ve already proven it under real conditions — not because it looks good in a demo environment.

That operational experience is what separates a practitioner from someone who read the documentation.

What You Should Walk Away With

By the end of a well-run engagement, you should have:

  • A clear inventory of which processes in your business are good candidates for automation (and which aren’t yet)
  • A working implementation of at least one workflow
  • A team that understands what was built and how to maintain it
  • An honest assessment of what comes next and roughly what it costs

You should not walk away with an AI strategy that requires a six-month implementation before anything is visible. You should not have a recommendation that requires a tool your team has no capacity to learn. And you definitely should not walk away without a working demo.

The Question Worth Asking Before You Sign Anything

One question cuts through a lot of noise: “Can you show me something you built for a previous client that’s still running today?”

Not a screenshot. Not a case study. An actual live system, or a demo environment from real work. If the answer is no — or if the answer is a lot of hedging — keep looking.

There’s no shortage of people who will take your budget to tell you that AI is transformative. Find someone who can show you how.

NexHam is an AI consulting and software company. We run a fully AI-operated business and help other companies do the same. If you’re evaluating AI consulting options, we’d like to talk.