Work In Progress Limits Explained With A Kitchen, Not A Factory

Your team is busy on nine things and has finished none of them. Every update sounds like progress and nothing reaches anyone.

Work in progress limits are a cap on how many things a team is allowed to have started at once. Starting less finishes more, because unfinished work does not help anybody and every extra started item makes the others slower. The usual explanation uses a factory, which is a poor picture for knowledge work. A restaurant kitchen on a Friday night is much closer to the truth.

The kitchen

Picture a small kitchen with three cooks and one pass where finished plates go out.

Twelve orders arrive at once. The kitchen has a choice. It can start all twelve, with pans everywhere, three cooks moving between dishes, sauces cooling while somebody plates something else. Or it can start five, finish them, send them, and start the next five.

Start all twelve and nothing leaves the pass for a long time, then everything leaves at once, half of it lukewarm. Start five and food goes out steadily from early on.

The total amount of cooking is the same either way. What changes is when customers get fed, and how much gets ruined while waiting.

Why the factory picture misleads

Factories are usually explained in terms of machines running at capacity. That makes managers think a person at eighty percent busy is a person being wasted.

Kitchens make the real cost visible. When a cook is running six dishes, they are not cooking slower, they are switching, remembering, checking what stage each pan is at. That switching is invisible in a report and expensive in reality. Same with your team. Somebody with four active projects spends a real part of the day just working out where they left off.

There is a second thing kitchens make obvious. A half cooked dish is worth nothing. Neither is a half finished project. Value arrives when it goes out, not when it is started.

What a limit actually looks like

A work in progress limit is a number written against a stage of work. Three per person, or six for the whole team in the doing column, or four in review. Which stages you can put a number against depends on how many columns the board has.

When the stage is full, you cannot start something new. You have to help finish something that is already there.

That last sentence is the entire mechanism, and it is the part people resist, because helping finish someone else's work feels less productive than starting your own.

Picking your first number

Do not calculate it. Observe it.

• Count how many things are genuinely in flight right now. Most teams are startled by the number, and a board that cannot tell you is showing one of the signs it has become fiction.
• Set the limit slightly below that. Not half, slightly below, or people will simply ignore it.
• Run it for a month and see what happens at the boundary.

The value shows up at the boundary. When someone cannot start their next task, the reason why becomes a conversation, and that conversation is usually the first honest look at a bottleneck the team has had in months.

What managers should expect to feel

Uncomfortable. That is normal and worth naming in advance.

A team under a limit looks less busy. Some people will appear idle for stretches. Your instinct will be to find them something to start, and that instinct is exactly what created the original problem.

The right response to an idle person is to ask what is blocking the nearest unfinished thing and whether they can help clear it. In a kitchen, a cook with a free minute helps plate the dishes that are nearly ready, they do not start a new order.

When limits are the wrong tool

Limits assume the work flows through shared stages. If everyone on your team works alone, end to end, with no handoffs, a team limit will not do much and a per person cap is the useful version.

They also do not fix genuine understaffing. If the queue is long because there are three people doing the work of six, a limit will make that painfully visible, which is useful, but it will not create capacity. Be ready for the honest conversation that follows.

The one sentence to remember

If you take nothing else from this, take the number that matters most: not how much your team started this month, but how much left the pass.

Count finished things. Then look at how many are in flight and ask whether cutting that number would make the first number go up. For most teams it will, and finished things are also the only number worth quoting when you are reporting to people who never open the board.