I work with owner-managed businesses on two things, using AI to remove the low-value work slowing your team down, and rebuilding how the business runs so the founder can step back without it falling apart.
The spreadsheet was the right answer at five people. At twenty-five, it is the bottleneck. Four signs to watch for, and a proportionate migration sequence that does not retire the spreadsheet on day one.
Delegating AI to a capable operator is the right call. This post names what you cannot hand off with the mandate, and why it matters.
When your board wants AI results by next quarter and your operator keeps explaining why that is not how it works, the founder's job is to hold a realistic timeline with enough authority that realism reads as command.
How to externalise the decision logic that makes your business run without giving away the judgement that made it worth having.
When a SaaS business comes to market, acquirers pay for defensibility, not feature count. This is what a genuine data moat looks like and why it moves the multiple.
Early in an AI programme, 'no results yet' is the honest answer. Learn to tell a credible update from one that is simply buying time.
Construction AI governance sits at the intersection of physical safety obligations, multi-party contract liability, and IP ownership of AI-generated designs.
The four governance gates every clinical AI delegate must clear before any AI system touches patient data or informs a clinical decision.
You handed off the AI mandate, so why does the team still route everything through you?
The real asset in a specialist manufacturer is rarely on the balance sheet. It lives in the heads of your most experienced operators, and when they retire, it goes with them unless someone has captured it.
© 2026 Larocca Consulting Ltd