Every COO has heard it. “We’re at capacity.” “The team is stretched.” “We can’t take on anything else.” And yet somehow, when a new priority lands — a board request, a sales commitment, a product emergency — the team finds a way to absorb it anyway. They work nights. They skip the lower-priority sprint items. They cut corners no one will notice until six months later.
The result is a slow, invisible erosion: burnout, attrition, quality debt, and a culture where “we’ll manage” becomes the default answer even when it really shouldn’t be.
The 120% capacity problem is one of the most universal and most damaging patterns in modern operations — and it almost always comes down to one thing: the way teams plan capacity doesn’t reflect how work actually flows through the organisation.
Why Planning at 100% Is Already Wrong
When you assign someone to a project at “100% capacity,” you’re assuming all of their working hours are available for that project. But that’s not how people work. Meetings, 1:1s, hiring interviews, ad hoc requests, Slack messages, context-switching — these consume 20–40% of most knowledge workers’ time before they’ve written a single line of code or processed a single task.
Experienced operations leaders know to plan at 70–80% utilisation. But most teams are still capacity-planned at 100%, which means every project they’re assigned to is already under-resourced before it begins.
The Visibility Problem
The second issue is aggregation. A COO typically sees project status at the portfolio level — RAG (Red/Amber/Green) status, milestone dates, budget consumed. What they don’t see is the person-level reality underneath: that three of your senior engineers are each allocated to five concurrent projects, or that your only UX designer is a single point of failure across six product initiatives.
Without visibility into team-level capacity, all you can do is respond to fires rather than prevent them.
The Allocation-Without-Timeline Problem
Many organisations track who is allocated to what, but not for how long. Knowing that Sarah is “on the platform migration” tells you very little. Is that 20% for two weeks? 80% for six months? The absence of time-bounded allocation is what causes the compounding effect — more work gets committed to the team because it looks like there’s slack that doesn’t exist.
What Good Capacity Planning Looks Like
Effective capacity planning requires three things working in concert:
1. Baseline capacity, not theoretical capacity. Know how many available hours per person per quarter exist after meetings, leave, and overhead — not just raw working days.
2. Time-bounded allocation. Every person’s allocation to every project should have a start and end date and a percentage. “Sarah: Platform migration, 60%, Q3 2025” is a capacity plan. “Sarah: Platform migration” is a hope.
3. Demand vs supply visibility. You need to see, at the quarter level, whether the total demand being placed on each team is greater than or less than their available supply. If it’s greater, something gives — and it’s better if you’re the one deciding what, not gravity.
The Compounding Effect of Chronic Overload
Teams that operate at 120% for a quarter can recover. Teams that operate at 120% for three years cannot. The long-term effects are well documented: increased attrition, reduced quality, slower velocity over time as technical and process debt accumulates. The people who leave first are always your best ones — they have options.
The COO’s job is to ensure the organisation is structured and resourced to deliver the strategy. That’s impossible if the teams executing the strategy are structurally unable to succeed.
What to Do This Quarter
Start with a capacity audit. For every team, answer: what is their realistic capacity this quarter (in person-days or hours, after overhead)? What work are they currently allocated to, and what does that total? The gap between those two numbers is your capacity deficit — and naming it is the first step to addressing it.
From there, you have a choice between four responses: increase supply (hire or contract), decrease demand (deprioritise work), extend timelines, or improve throughput. Each has implications. Only one of them — continuing to absorb overload silently — has no good long-term outcome.
