Most expensive mistakes I’ve seen weren’t failures of execution. They were companies spending real money to solve the wrong problem — quickly, confidently, and in the wrong direction.
I find the real one first.
Across enterprise telecom, contact centers, sales, operations turnarounds, and now AI systems, the pattern has been the same: read past the stated problem, identify the constraint underneath it, then design the simplest move that fixes the structure instead of just treating the symptom.
At a refinishing business sliding from $600K toward an estimated ~$450K, the issue was not simply “sales.” It was demand structure. Repositioning the company toward recurring commercial accounts drove it to roughly $900K that year and ~$1.1M the next. With Stethoscope, the AI-assisted SaaS product I built and deployed, the original problem was not model quality. The deeper issue was input quality: the workflow was not capturing the context that actually drives business decisions.
Stethoscope is now production-hosted and in pilot. It helps small businesses capture operational context through structured AI-guided huddles across areas like HR, legal, people, strategy, financials, and CRM-lite workflows. Building it gave me a grounded view of where AI actually works, where it breaks, and why workflow design matters as much as model selection. My opinion about AI is not built on consensus or commentary; it comes from designing around the limits I ran into while shipping.
I’m building toward consulting, product, or solution-strategy work where this diagnostic ability is central to the role: finding the real constraint before companies commit time, money, or people to the wrong solution.
If you know a situation where that would be useful, I’d welcome the conversation.
