Use cases
Project management for a small startup: less ceremony, one number
Before you have a process you have priorities. A board, a calendar and a budget of hours are enough to keep five people aligned without inventing rituals.
At three to ten people nobody needs a project management office. What you need is to know what is being worked on, what is blocked and how much time you have left this month.
Every ceremony you add has a cost paid by the same people who are supposed to be building the product.
Backlog first, board second
The backlog is where everything goes before it deserves attention: ideas from users, bugs, the pricing page you keep postponing. The board only holds what someone is working on this week.
That separation is the cheapest form of focus there is. If the board has thirty cards, it is not a board, it is a wish list with columns.
Track the hours left, not the velocity
Velocity charts need stable cycles and enough history. A startup has neither. A simpler number works better: how many hours of work are committed this week, and how many are left.
Estimating by task also gives you the only input you need to decide whether to hire or to cut scope. Without it, “we need one more person” is an intuition.
Keep the team small as long as you can
Before hiring an operations person, look at what is actually repetitive: writing up task lists from calls, moving cards after decisions, summarising the week. That is the work an AI agent can take over through MCP, with access limited to one project.
The rest — deciding what matters, talking to users, saying no — should stay human for a while.
Frequently asked questions
- Do we need Scrum to use the board?
- No. Columns, a backlog and a weekly review work fine. Add cycles and roles when they solve a problem you actually have, not because a framework says so.
- Should everyone log their hours?
- If you sell time or need to know how many hours things consume, yes, everyone. If you are building a product, log hours on the initiatives that consume real capacity and skip the rest.
- Is a tool worth it at three people?
- Worth it when not knowing costs more than the tool: usually when the first client asks for a report, or when two people are working on the same thing without knowing.