Weekly Priorities by Subtraction: Choosing What Your Team Will Not Do This Week
Most weekly planning starts with additions. Someone reads the backlog, pulls in the items that feel important, and ends up with a list that describes everything the team hopes to finish. By Wednesday that list has quietly become a record of good intentions. A subtraction-first approach reverses the order. Instead of asking what the team should do this week, you begin by asking what the team will not do, then build priorities from the work that remains. This article describes how to run that exercise on a shared board, what questions to ask, and how to tell whether the result is realistic.
Start with the full candidate pool
Before removing anything, collect every item that someone has argued belongs in the coming week. This includes cards already in progress, carry-over work from last week, requests from stakeholders, and follow-ups that were promised in meetings. Do not filter during collection. The goal is to see the true size of the pool, because teams that skip this step tend to discover their real backlog only after the week has started.
Once collected, group the candidates by the kind of commitment they represent. Some are work already underway, some are new requests, and some are maintenance tasks that keep existing systems running. Grouping makes the next step easier, because a new request and an in-progress card usually deserve different treatment even when they look similar on the board.
Apply the subtraction test
For each candidate, ask whether the week would be meaningfully worse if the item were deferred to next week. Be specific. A vague sense that something matters is not enough. Ask who would notice, what decision would be delayed, and what downstream work would stall. If the honest answer is that nothing much changes, the item moves to a deferred column rather than the weekly plan.
A second question helps with items that survive the first test: what does finishing this item unblock? Work that unlocks other people's progress usually ranks above work that only benefits its owner. This is where a board with visible dependencies helps, because you can see which cards other teammates are waiting on rather than relying on memory.
Limit the active set
After subtraction, count what remains. Many teams find that a short list of three to five priorities per person is workable, while longer lists tend to become decorative. The right number depends on the team's typical interruption rate. If a team regularly loses a day to urgent support, a list of five items may already be optimistic. Treat the number as a hypothesis to test over several weeks rather than a rule.
Write the final priorities in plain language that someone outside the team could understand. A priority like "ship the onboarding email sequence to the first cohort" is more useful than "onboarding," because it signals what done looks like.
A short checklist for the review
Use the following questions as a checklist during a thirty-minute planning session. Does every priority have a named owner? Does each item have a stated definition of done? Have you identified at least one dependency for each priority, and confirmed that the other party knows they are waiting? Is the deferred column visible to the team, so that nothing disappears silently? Have you compared the active list against the number of working hours each person actually has, after meetings and known interruptions? If any answer is no, revise before the week begins rather than during it.
Finally, review the previous week's deferred items. Items that have been deferred two or three weeks in a row deserve a direct decision: either commit to them with a date, or remove them from the board entirely. Keeping stale deferred work visible without a decision creates the same overload that subtraction was meant to prevent.
Subtraction does not guarantee a calm week. It does make the trade-offs explicit, so that when interruptions arrive, the team already knows what gives way. That shared understanding is often the difference between a plan that survives contact with reality and one that is abandoned by Tuesday.