The task board: the minimum that works

Why a small team needs one

So everyone can see who's doing what and what's stuck, without asking.

Without a board, a team of three loses the picture within two weeks: someone builds what another already built, something is forgotten, and nobody can answer "what's left?".

The minimum set of columns

Four, and that's enough:

To do. Everything agreed but not started.

In progress. What people are working on right now. Every task has a name against it.

In review. Done, but somebody needs to look. Optional for a small team, but useful: it shows work is stalled through no fault of its author.

Done. Finished and checked.

What isn't needed at the start: "in analysis", "in testing", "ready to release", "released". Those appear when the process grows complicated, not in advance.

What to put in a task

A comprehensible title. Not "the form" but "expense entry form: fields, saving, amount validation".

What counts as done. A line or two. It spares you the classic exchange of "I thought this needed an export too" — "I thought it didn't".

Who's doing it. A name. A task without a name in the in-progress column is being done by nobody.

Nothing more is needed at the start. Detailed descriptions, subtasks, labels and priorities appear if and when required.

The rule that pays most

Limit how many tasks are in progress. One per person, two at most.

The reason: when everyone has taken three things, everything sits half-finished. There's a sense of furious activity and no output. When everyone takes one and finishes it, work moves faster, though it feels slower.

It's the most underrated rule in organising work, and the most frequently broken.

Not drowning in the tool

The temptation to configure everything elaborately is strong. Resist: a team of three doesn't need seven task types, automated transitions and fifteen custom fields.

The sign you've overdone it: people stop moving the cards. The tool has become more expensive than the benefit.

The minimum viable option is any free board, or even a shared document with four lists. The tool is secondary; the habit is primary.

What kills a board

Cards nobody moves. If a card sits in progress for a week while the person is doing something else, the board stops reflecting reality and people stop trusting it.

A dumping ground in "to do". A hundred tasks, half of them obsolete. Review and discard monthly.

Tasks with no owner. "Someone will do it" means "nobody will do it".

Only the lead using it. If one person moves all the cards, it isn't the team's board but their personal list.

Why start one on your own project

Not for tidiness, but because a board is the simplest way to see a project stalling. An empty "done" column after two weeks and three cards stuck in progress tell you the project's state more honestly than any feeling.

And it gives you something to say at interview: how you organised the work, how tasks were divided, what you did when someone got stuck.