The strategy-deck problem

Most AI strategy decks describe a market opportunity, list popular use cases, and recommend a roadmap of pilots. They could be applied to almost any business. That's the problem — they aren't actually about your business.

A strategy that could be cut-and-pasted into another company isn't a strategy. It's a market summary with a budget attached.

The result is predictable. Companies invest in three or four pilots, see scattered returns, and quietly conclude that AI "doesn't work for them." The technology wasn't the problem. The diagnosis was missing.

What a real diagnosis looks like

A real diagnosis is a written document. It names the specific places in your business where AI saves real money — and the places where it doesn't.

It comes from time inside your business, not from external research. We sit with your QS for a morning. We watch your bid manager send a tender. We follow a callout from booking to invoice. The findings are specific: "Your bid cycle takes 11 days. Seven of those days are administrative. Here's what AI removes, what it doesn't, and what it costs."

The diagnosis includes a sequence (which workflow first, why, and what it has to prove before the next one starts), a number (expected pounds saved over twelve months, with assumptions stated openly), and what we'd recommend against. The last part matters most.

What you should expect from a diagnostic-first engagement

A diagnostic engagement is short. Usually three to four weeks. The deliverable is a written document, not a slide deck.

The cost is meaningful but not large — typically less than what a single bad pilot would cost. The point of the diagnosis is to make sure your bigger investments don't go to the wrong places.

It's also the only honest way to know whether the company you've hired understands construction. After four weeks, you'll know whether they've engaged with your business or repackaged generic advice.

When a diagnosis says "no"

Sometimes the diagnosis concludes that AI isn't the right next investment. The bigger lever might be CRM data, process documentation, or hiring one more PM.

That outcome is rarer than buyers expect — there's almost always something worth doing — but it's not impossible. And the willingness to say "no" is the difference between an advisor and a salesperson.

A diagnostic-first approach makes that "no" possible. A pilot-first approach makes it almost impossible to say.