Project management, right-sized.
Also called: Project setup · Cadence & gates · Lightweight delivery · Running the build
A light, deliberate way to run a small product project, so the work moves and the key decisions get caught, without drowning in process.
Pick a cadence, set a few decision gates tied to the ten stages, and track the decisions, not just the tasks. For a two-person build, that is the whole job. Over-processing a tiny team is as costly as no process at all.
What project management is here
Project management here is not a Gantt chart, a sprint board with forty cards, or a weekly status deck nobody reads. For a product built by two or three people, all of that is overhead pretending to be progress. What you actually need is small: a rhythm that keeps the work moving, a handful of points where you stop and decide before spending real money, and a written record of the decisions that shaped the product.
The trap most small teams fall into is binary. Either they run no process at all, in which case the project drifts and the same questions get re-argued every fortnight, or they import a heavyweight framework built for a thirty-person team and spend more time grooming the board than building the thing. Both fail the same way: the actual product work gets crowded out, by chaos in one case and by ceremony in the other.
The right size for a lean physical-product project is three moving parts. A cadence so the team knows when it syncs. A set of decision gates that line up with the ten stages, so you do not commit to tooling before the design is settled. And a decision log, because in my experience the thing that sinks small projects is not a missed task, it is a settled decision quietly coming undone six weeks later because nobody wrote down why it was made.
Track decisions, not just tasks
Tasks tell you what is being done. Decisions tell you what the product is. “No app” is not a task; it is a decision that shapes the firmware, the price, the support burden and the marketing. If the only thing you track is a task list, that decision lives in someone’s head, and heads forget. Write the decision, the date, and the one-line reason. That single habit is worth more than any board.
How you run it
Here is the setup we used on the proofing box. It fits on one side of paper, which is the point. Anything that needs a second page on a two-person project is usually solving a problem you do not have yet.
Notice what is missing. No story points, no velocity, no burndown, no role nobody needed. Five rows did the entire job of running a project from idea to launch, because the team was small enough that information moved between two people without a system to carry it.
Over-processed against right-sized
The same project, run two ways. The left column is what happens when a small team copies a big team’s playbook, or runs on vibes. The right is the version that fits the headcount.
- A board with forty cards, groomed more than the product is built.
- Status decks written for an audience of two people who already know.
- Or the opposite: no rhythm, decisions re-argued every fortnight.
- Tooling money spent before the design was ever frozen.
- Key decisions living in someone’s memory, undocumented.
- One short weekly call, a task list for the live stage only.
- A handful of gates tied to the stages, money held behind each.
- A decision log: what, when, why, one line each.
- Clear lanes, so most calls never need a meeting at all.
- Tools you already own, nothing that needs a tutorial.
The honest claim, and the one I will stand behind: for a project this size, every hour spent on process beyond that right-hand column is an hour stolen from the product. The discipline is keeping the process small on purpose, not letting it grow because growing it feels productive.
How it fits the bigger picture
Project management is activity 01.05, the last activity of Stage 01 Idea. It sets the rhythm and the gates that carry the project through everything that follows, starting with Stage 02 Discover, where the research that tests your idea begins. Everything in Idea was about deciding what to build; this activity is about how you will run the building.
What it can do
It gives a tiny team just enough structure to keep moving and to stop spending money at the wrong moment. The gates protect the budget; the cadence protects momentum; the decision log protects the choices you already made from quietly unravelling later.
What it can’t do
It cannot make a weak idea work, and it will not save you if you skip the decisions the gates are meant to test. A gate is only as good as the work behind it. Process organises judgement; it does not replace it, and no cadence rescues a project that has nothing worth building.
See the full 10-stage process →
Try it yourself
Before Stage 02, spend twenty minutes setting up. Pick one fixed weekly slot for a short sync. Write down three or four gates tied to the stages where real money or commitment is at stake. Open a single decision log and put your first decision in it, with today’s date and the one-line reason. Resist adding anything else until a real problem demands it.
Or let the guided version frame it for you. Start the Free Sprint → and it will help you name the decisions worth tracking before you build.
Your project-setup checklist
Project notes: running it light from Stockport
▸ From the notebook · optional reading
How Dan and Anna Hartley ran the whole proofing-box project on one weekly call, four gates and a single decision log, from a kitchen table in Stockport.
3 min read · click to open
When we first sat down with Dan and Anna in Stockport, Dan wanted to buy project software. He had used a heavy board at his old job and assumed that was what running a project meant. I asked one question: “There are two of you. Who is the board for?” He paused, and we agreed to start with nothing and add only what hurt to be without.
The whole system
It came down to three things. A 30-minute call every Monday morning, before the day started. Four gates, written on a sticky note above Dan’s bench: Discover done, design frozen, prototype through its BS EN 61010 bench checks, tooling committed. And a decision log in a shared doc, one line per decision.
That last one earned its keep faster than I expected. “No app, no Wi-Fi, no account” went in on day one, with the reason: keeps the bill of materials at £38 to £55, protects the £149 price, no support burden. We worked that single line hard. Three separate times across the year, once Anna’s, twice a contractor’s, someone proposed adding connectivity. Each time the answer was already written, dated, and reasoned. The conversation closed in minutes instead of becoming a Monday-call argument.
Where the gates did their job
The tooling gate was the one that mattered most. A Stoke-on-Trent ceramic supplier offered a discount if the team committed to the mould early. Tempting, and on a vibes-run project they would probably have taken it. But the gate said design frozen first, and the design was not frozen: the wall thickness was still moving as the Manchester PCB heater layout settled. We held. Two weeks later the thickness changed by 3mm. Committing early would have scrapped a mould.
What it cost
- Cost. Twenty minutes of setup, then thirty minutes a week. No software, no licence, no learning curve.
- Saved. One scrapped ceramic mould avoided, three connectivity debates closed before they started, and a price band that held from idea to launch.
The temptation, always, is to feel that a small process means you are not taking it seriously. The opposite is true. The process was small because the team was small, and keeping it that way was the most disciplined thing they did.
— Idea stage, project notes, 2026
— Next stage → Stage 02 · Discover