Projects
Freeze: the hardest first step in managing many projects

When an organisation runs many projects or orders at once with shared people, the single most powerful first step is also the most uncomfortable one: freeze a large share of the open work, so that only a fraction is actively worked on at any time.
It feels wrong. Everything is already late. How can stopping work help?
Why freezing works
Freezing means ranking all open projects by priority, keeping the top few active, and putting the rest on hold until their turn comes. With fewer projects open:
- Each project moves in bigger chunks. People’s time is split across fewer projects, so each one progresses faster and tasks finish sooner.
- Less time is lost switching. Fewer switches mean less set-up and less re-learning, so output goes up.
- Work comes together better. Projects progress in step across teams, and integration points stop being bottlenecks.
The limit on active projects is set just high enough to keep everyone busy, so nobody is starved of work.
The active list and the waiting list
After the freeze, each team has:
- an active list: projects currently being worked on, up to the work-in-progress limit; and
- a waiting list: projects frozen until their turn.
A project moves from waiting to active by one rule: one out, one in. A new project waits until an active one is completed.
The waiting list flexes. If projects arrive faster than they are closed, it grows. A list that keeps growing is the signal to add capacity. Until then, adding more projects to the active list would only slow every project down and reduce total output.
How to set the limit
The limit depends on two things: how many people the team has, and how many can usefully work on one project at the same time.
Take a team of 12 people. If up to 4 can work on one project at once, a limit of 3 projects could keep everyone busy. But project content varies: some weeks a project can only use 2 people. So a realistic limit might be 4 or 5.
A practical method:
- List all open projects in priority order.
- Assign people starting from the top project, as many as each can usefully take.
- The point at which everyone is assigned gives the starting limit.
If projects often have natural waiting periods (for client approvals, materials or test results), raise the limit a little, so there is always something ready to work on.
Expect the team to be a bit overloaded or under-loaded from time to time. The limit is a good-enough estimate, refined with experience.
One list or many?
How teams share lists depends on how they work together.
Teams that work in tandem, handing work back and forth frequently, need one common active list and one limit. Otherwise their work falls out of step.
Teams working in parallel on different parts of the same projects, which come together later, can share a list if they can handle the same number of projects. But if one team takes much longer than another, a common list would leave the faster team waiting for the slower one before a new project can start. Then separate active lists, drawn from one shared master priority list, work better.
Teams working in sequence, where one finishes its part and hands over to the next, each keep their own active list. A project completed upstream joins the downstream team’s waiting list. Every project carries a master priority number that stays the same through all teams, so a higher-priority project always moves ahead of a lower one in each waiting list.
Most real organisations mix these patterns. Map how the work actually flows, and set lists accordingly.
When project-level limits aren’t enough
Sometimes, even with the project limit respected, people are still overloaded, because the active projects happen to be unusually heavy. Then add a second level of control: a limit on the number of components being designed at once, or on the number of tasks open within active projects.
What the first month feels like
Picture a design team that freezes half of its 30 open projects. In the first two weeks, clients of frozen projects call, and managers have to hold the line. Output looks no better yet. In week three, the active projects start closing noticeably faster, because people are no longer pulled across fifteen other jobs. By week four, two or three frozen projects have already moved into the active list and are progressing faster than they did when they were “open”. By the second month, total completions per week are usually above the old rate, and the conversation shifts from “why did we freeze?” to “why did we ever run thirty at once?”
The bitter pill
Freezing is the most crucial step in any multi-project improvement, and the easiest to dilute. Every open project carries commitments made to someone outside the organisation, and there is huge pressure to keep working on all of them.
The temptation to keep everything moving is exactly what keeps everything slow.
Holding the freeze takes firmness from leadership. The first weeks can feel uncomfortable. But as frozen projects start moving through faster than open projects ever did before, the benefit becomes visible to everyone, and it lasts.