Approach
Two things make the work different: twenty years of being responsible for programs from start to finish, and a careful, evidence-based way of finding out what is really going wrong. I apply both to aircraft cabin programs and to strategy work for companies.
MIT's 2025 NANDA study found 95% of enterprise generative-AI pilots deliver no measurable P&L return.
The reason is almost never the technology. MIT NANDA's research points to a missing problem definition: teams pick AI first and then look for something to use it on. This site does not lead with AI. It leads with twenty years of judgment from running programs, and AI comes in only once the real problem is clear. Tools change. Knowing which problem is worth solving does not.
Advice usually comes from one of two sides: analysts who have never been responsible for the result, or managers who know only the one program they ran. This work brings both together: many programs led from start to finish, so the patterns are familiar, and a careful, evidence-based check of every conclusion.
Experience
Twenty years of managing programs and being responsible for the result, not advising from the outside. From the contract to delivery to the return of leased aircraft, in airline cabin programs and in building a new defense plant from scratch. Enough variety, and enough repetition, to see how programs really run and where they quietly start to fail.
Approach
The problem everyone sees is rarely the real one. A stalled cabin project, a slipping production line, aircraft that cannot fly: the real cause usually sits one level deeper, in how the program is run, in the contract, or in the supply chain. That is where the work looks first, before anyone spends money on the symptom.
On the owner’s side
Judgment formed as the person responsible for the result, or as the buyer checking a supplier, never as the seller. That is the view an honest review needs.
Evidence, not opinion
Every figure has a source and a date, and is checked. Where sources disagree, both numbers are shown, not averaged into one convenient figure.
A decision, not a slide deck
AI does the research at scale; people make the judgment. The result is something you can act on, with its assumptions in the open.
I look at the whole system
I have led programs across their entire life cycle: from the first contract and business case, through design, suppliers and production, to delivery, years of operation and the return of leased assets. Having been responsible at every stage, I know where things tend to fail, and that they usually fail where two parts of the system meet.
So I do not start with one department or one problem. I look at how everything works together: sales, purchasing, operations, finance, suppliers and customers. The question is always the same: what holds the whole business back?
One weakest link
Weakest link
Every business works like a chain. At any moment, one weakest link limits what the whole chain can deliver. Improving everything else changes little; improving that one link moves the whole business.
- 1Find the one thing that limits results today.
- 2Get the most out of it, and line everything else up behind it.
- 3When it is no longer the limit, look for the next one.
What everyone sees
The problem everyone sees: a stalled cabin project, a slipping production line, grounded aircraft
How the program is run: schedule, structure, who decides
The contract: terms, incentives, return conditions
The supply chain: the part, the supplier, the bottleneck
Where the real cause sits
Root cause
AI reads large amounts of data; every figure has a source and a date, and is checked. Each step down is backed by evidence, not opinion.
Where Do You Really Stand?
Why an outside-in look comes before any decision: three questions, hypotheses instead of verdicts, and AI in its proper place.
Read the article →Whether it is an aircraft cabin program or strategy work for a company, the job is the same: find where things will really fail, prove it with checked data, and come back with a decision that holds.