Watch enough of the YouTube sailing crowd and you will see a lot of them run refit and maintenance off a Kanban board. It is a horizontal list. Work flows left to right. You start with a backlog on the left, then move cards through ready, in progress, in review, and done. Some people run all of those columns. Some strip it down to planned, in progress, done. Kanban has its own vocabulary for the states, but the idea is simple. A job lives in one column, and it moves right as you make progress on it.
I run mine on GitHub, and it costs nothing. A free account gives you a project board that does far more than a wall of stickies ever could. A card links to the actual repair notes. It holds photos of the problem and the part, so you are not squinting at a fitting trying to remember which one you meant. It carries labels, a priority, a size, a budget, and start and end dates as real fields you can sort and filter, not scribbles you have to read one by one. And it lives on every device you own, so the board is on your phone at the chandlery and on your laptop at the nav station, always the same board. That is a lot of capability, more than most people think they need, and once you have it you tend to use it. None of it is required. Sticky notes on a locker door work just as well, and for some people they work better, because the whole board sits in front of you and you never have to open a laptop to see it. That trade is yours to make. Here is the GitHub project board for Eoti.
The Eoti board on GitHub. Cards flow left to right, from backlog through done.
Kanban is good at one thing. It shows you where a job sits in the flow. What it does not do, at least not on its own, is tell you which job matters most. That gap is where I ran into trouble.
Where the flow board broke down
A post-purchase refit is where the wheels come off a plain Kanban board. I started with the survey on Eoti. It was clean. About a dozen items, all moderately important, nothing on fire. Easy to track left to right.
Then I got to know the boat. The deeper I looked, the more I found, and the sticky notes multiplied. Pretty soon I had a wall of cards, all sitting in the backlog, all shouting for attention, and the flow board had no way to tell me which one to grab first. A card that says the head backflows and could sink the boat looked exactly like a card that says the lack of freshwater flush is annoying. Same column. Same size sticky. That is the priority problem, and Kanban does not solve it.
The grid I built instead
So I stopped running a pure Kanban board and built something that answers the priority question directly. I kept the vertical columns for flow, backlog through done. Then I drew three horizontal lines across the whole board and turned it into a grid. Now every card has a column that says where it is, and a row that says how much it matters.
I named the three rows. The top tier is Health and Safety. The middle tier is Maintenance and Repair. The bottom tier, the loneliest one, is Comfort and Welfare.
The rows map to three priorities: critical, medium, low. Critical is anything that touches the life and safety of the boat. We started with the old line for that one, keep the people in the boat and the water out of the boat. Later we added, and not on fire. Medium is any job you can defer for a month or until the next port without it turning critical in that window. Low is anything that just makes the boat more pleasant to live on.
Here is the split in one example. A toilet that backflows and can sink the boat is critical. A macerating freshwater flush is comfort. Same fixture, opposite tiers. Knowing which is which is the whole game, and the grid forces you to decide it when you write the card.
The cheap way, no technology required
You do not need any of this on a computer. The whole system works with a flat surface, a pad of sticky notes, and a marker. That is the version I would start anyone on, because you can build it in ten minutes and change it with your hands.
Pick your surface. A bulkhead, a hatch board, the inside of a locker door, a hanging locker, even a stretch of painted plywood in the garage while the boat is on the hard. It needs to be somewhere you walk past every day, because a board you cannot see is a board you forget.
Lay out the grid from the slide above. Run painters tape or a grease pencil in three vertical lines for your columns, backlog, in progress, done. Then three horizontal lines for your tiers, health and safety on top, maintenance and repair in the middle, comfort and welfare at the bottom. Now you have nine boxes, and every box means something. A note in the top left is a critical job you have not started. A note in the bottom right is a comfort job you finished.
Color code the tiers if you have colored notes. Red or pink for health and safety, yellow for maintenance, green for comfort. Then you can read the priority of the whole board from across the cabin without reading a single word. If all you have is one color, write the tier in the corner and move on.
Write the card by hand. Job on the top line. Under it, a budget, a rough size as a t-shirt guess, and a start and end date if you know them. That is the whole card. It takes longer to describe than to write.
Move the note as the work moves. Pick it up, slide it one column to the right, press it back down. That physical act is the point. You are not clicking a mouse. You are handling the job, and your hand remembers what your head forgets.
What goes on a card
A single card, opened up. Labels, priority, size, budget, and start and end dates all live on the sticky.
Deciding which row a card belongs in is an art. Writing the card is where the real work sits. On each sticky I put a budget, a rough time to complete as a t-shirt size, and usually a start and end date. (The example board does not show dates, so sue me.)
I also split the budget into parts and vendor. That way I can see at a glance what a job costs if I do it myself versus what it costs if I hire it out. Over a full refit that number adds up, and it is the difference between a lot of decisions.
The card carries one more thing that saves real money: parts status. On the sticky I mark whether a job needs parts, and whether those parts are already on hand. Do that across the whole board and you can batch your orders, pregame the jobs, and cut a serious amount off your shipping bill because you are not paying to rush one bolt at a time.
This is where the free tool earns its keep over the wall. On GitHub those budgets and part flags are structured fields, so I can sort the whole board by cost, filter to just the jobs waiting on parts, or total the parts column to see what the next order runs me. On a wall of stickies you are doing that math by hand, one note at a time. You do not need any of it to get started. But it is sitting there for free the day you decide the wall has gotten too big to read.
Working the board
Two rules keep me sane once the work starts. First, I never let too many cards pile up in the in progress column. Past two or three I lose the thread, bounce between jobs, and burn time doing it. Second, I never move a card backward, out of in progress and back to the backlog or a waiting column. A card that slides backward is a job you have quietly given up on, and it tends to stay dead.
Some people look at all this and see abysmal overhead. They would rather just start turning wrenches. I get it. But the reading is the doing. Three vertical columns crossed with three horizontal rows is fast. Call it a hasty solution if you want. It will save you money and put you well ahead of the boats still sitting in the yard wondering what to fix first.
I’ve included the powerpoint if you wanted to play with the formatting a bit too.
About the Author
