When your team says "I'm at capacity"

Nonprofit and public sector leaders have heard "do more with less" for so long that it barely registers as a request anymore. It's just the climate we work in. The data backs that up. In a national survey of more than 1,600 nonprofits, three quarters reported job vacancies, and half named stress and burnout as a reason they couldn't recruit or keep staff (National Council of Nonprofits, 2023).

Many of us hoped AI would give some of that time back. In some ways it has. But the early research also shows that it’s made things harder. In one survey, 96% of executives expected AI to boost productivity, while 77% of employees using it said it had added to their workload (Upwork Research Institute, 2024). A Berkeley study that followed a 200-person company for eight months found that employees using AI worked faster, took on a wider range of tasks, and let work spill into more hours of the day, often without anyone asking them to (Ranganathan & Ye, 2026). When work gets faster, expectations rise to match it. The bar moves, and nobody announces that it moved.

So when someone on your team says they don't have capacity, it's worth slowing down. The instinct is to reassure them or to push through. Both skip the diagnosis. And as you'll see, an honest diagnosis can lead somewhere most of us would rather not go.

What Capacity Concerns Are Really Telling You

I coached a deputy director, I'll call her Noor, whose program manager kept saying she was underwater. Noor's first impression was that the manager needed better time management. When we talked it through, it turned out to be three different problems stacked on top of each other. The manager was spending hours each month polishing a report no one above her read closely. She'd been handed a new data system with no real training. And her caseload had grown by a third after a colleague left and their position wasn’t filled. Each of those needed a different response, and time management training wouldn't have fixed any of them.

That's why I find it useful to ask three questions, in this order.

Is it a clarity question?

Sometimes people are working hard on the wrong things, or putting lots of effort into work that only needs to be good enough, because no one has told them what matters most. Gallup found that only 46% of U.S. employees strongly agree they know what's expected of them at work, down ten points since 2020 (Gallup, 2025). In a stretched organization, everything gets labeled urgent, and staff end up sorting priorities on their own. This one is yours to fix: name what matters most, what can be good enough, and what can stop.

Is it a skills question?

Sometimes the priorities are clear, but the person doesn't yet have the tools, training, or support to do the work in the time they have. This is especially live with AI. In the same Upwork survey, nearly half of employees using AI said they didn't know how to reach the productivity gains their employers expected (Upwork Research Institute, 2024). The job demands-resources model helps here. Demands become draining when they aren't matched by resources like training, feedback, and support (Bakker & Demerouti, 2007). A skills gap is a resource gap, and it's one you can close.

Is it a design question?

And sometimes the work simply isn't built to fit in the hours available, no matter who's doing it or how skilled they are. Christina Maslach and Michael Leiter identify workload as one of six areas where a mismatch between a person and their job drives burnout, and it's one of the strongest predictors of exhaustion (Leiter & Maslach, 1999). The key word in their research is mismatch. Burnout comes from the fit between the job and the person, not from the person alone. If you'd expect anyone in that role to struggle, clearer priorities and more training won't solve it. The job itself has to change: fewer responsibilities, more people, longer timelines, or an honest conversation upward about what won't get done.

Why the Third Question Is the Hardest

Clarity and skills gaps are mostly what Ronald Heifetz calls technical problems. They're real, but the path forward is known: set priorities, provide training, add support (Heifetz, 1994). The design question is different. It's adaptive, because answering it usually means a new way of doing things, and a new way of doing things means someone has to give something up.

Heifetz and Marty Linsky put it plainly: people don't resist change so much as they resist loss (Heifetz & Linsky, 2002). When the work doesn't fit the hours, every honest option carries a loss. A program you care about gets scaled back. A funder hears that a deliverable will slip. A board learns the organization can't keep promising everything it has promised. And leaders often have to let go of part of how they see themselves, as the one who always finds a way to make it work. That's hard to admit. It's even harder to do, because once you name a loss out loud, you have to carry it and help other people carry it too.

That's also why the design question gets skipped. A few weeks ago I wrote about the fundamental attribution error, our habit of explaining other people's struggles by who they are rather than the situation they're in (Ross, 1977). It's easier to hear "I'm at capacity" as a clarity or skills problem, because those feel like they're about the other person, and they don't ask the organization to lose anything. But when staff flag capacity, they're handing you information you can't get any other way. Treat it as data, not a complaint, even when the data points to a loss you'd rather not take.

Want these insights sent straight to your inbox? Sign up for the newsletter below!

Previous
Previous

How to stand behind a decision when it gets hard

Next
Next

The one question that turned a hard no into a yes.