The problem
The design asks how a boat’s service history could remain useful across jobs and changes of owner.
Marine service history is attached to whoever did the work rather than to the boat. One yard holds the engine records, another the electrical job from two seasons ago, and the owner holds a folder of invoices that may or may not be complete. When the boat is sold, most of that context is lost, and the next owner and the next yard start over.
The cost is paid twice. Diagnosis repeats work that was already done, because nobody can see what was replaced or why. And the boat’s value at sale rests on assertion rather than record, which is precisely the situation where a documented history is worth the most.
My role
I designed the business, the per-hull record model, the brand system and the financial model. A separate AI-assisted review tested the assumptions and the model.
What was designed
A service business around a permanent per-hull record: intake, estimates, deposits, invoices, and the record carried forward with the boat, with a brand system and a financial model behind the plan. This entry presents the design artifacts. It does not establish a customer deployment.
How the design works
The organizing decision is what the record belongs to. Attaching it to the hull rather than to the customer account is what allows it to survive a change of owner, and it changes the shape of everything downstream: intake identifies the boat first, estimates and invoices attach to it, and the history accumulates in one place across jobs and across owners.
The commercial thinking follows from that. A service business normally treats its records as internal operational overhead. Here the record is part of what the customer is buying, which is a different promise and a different reason to keep it complete. The financial model and the brand system were built around that promise rather than bolted on afterward.
It is worth being clear about what remained unresolved in the design. Nothing here settles who is willing to pay for a record that mostly benefits a future owner, or how a yard is persuaded to contribute history it currently treats as its own. Those are the questions customer validation exists to answer, and they were not answered.
Evidence
A separate AI-assisted review dated August 14 scored the market-sizing checklist at 81.8% and the financial model at 71.9%. Its scripted investment simulation returned Decline at 30.8%; the simulated partner debate preferred More Diligence. These are review outputs, not market validation or an investor decision.
Comparable-company claims that failed re-verification were removed. The project record still lists zero of 26 planned validation calls completed.
Observed results
There is no result to report, because the design was never put in front of customers. The review that exists was a scripted exercise, and its own verdict was to decline the proposal as presented and seek more diligence.
That verdict is left standing in this entry rather than reframed. The work demonstrates a business and product design capability, tested against a structured critique that it did not pass, and the honest conclusion is that the plan needed customer evidence it did not have.
What this does not show
This is business and product design work. The reviewed record contains no customer deployment or completed customer-validation calls. The AI investment simulation declined the proposal as presented; no real investment committee reviewed it. A public brand board would show design work only.
An AI-assisted review is also not an independent one in the sense that matters commercially. It tests a plan for internal consistency and for the gaps a checklist can see. It cannot tell you whether anyone will buy.
Relevant next action
If a recurring task needs a usable application, the application and field-form engagement starts with the person who will use it and what happens when they cannot. The delivered form of this pattern, where field work reaches the office as an operational record, is in field-to-office capture.