Ask someone on your team how their day went and listen for this answer: “Busy.”

Then ask what they actually got done.

Watch the pause.

That pause is one of the most useful diagnostics you have, and most leaders walk right past it. Someone worked nine hours, ended the day drained, and can’t name a single thing they finished. We tend to read that as a workload problem. Usually it’s something else entirely.

We’ve been blaming the wrong work

When people burn out, the instinct is to assume the load was too heavy. Too many projects. Too many hours. Too much pressure.

But hard work, on its own, rarely burns people out. Hard work has an arc. It has a finish line. It produces something you can point at on Friday and feel a little proud of. People will happily go home tired from that.

What actually grinds people down is the easy work. The small, endless, low-judgment stuff that asks nothing of them except that they be available.

It’s the least demanding work in your organization, and it’s the most expensive.

Why easy work costs more than it looks

The problem isn’t the task. It’s what the task does to everything around it.

• It never finishes. There’s no version of “done” — just a queue that refills overnight.
• It arrives as interruption, never as a scheduled block, so it lands in the middle of the work that needed focus.
• It doesn’t develop anyone. Nobody gets better at their job by answering the same question for the four-hundredth time.
• It’s invisible. It doesn’t show up in a review, a report, or a win. It just quietly eats the day.

And then there’s the switching cost, which is where the real money goes.

A question that takes ninety seconds to answer doesn’t cost ninety seconds. It costs ninety seconds plus the ten or fifteen minutes it takes to climb back into whatever required actual thought. Do that eleven times before lunch and you’ve spent the entire day at work without ever getting to work.

That’s the feeling behind the pause. Not “I did too much.” It’s “I never got started.”

The tell

Every organization has its own version of this, but it usually sounds like a handful of questions on a permanent loop:

• What are your hours? Are you open on the holiday?
• Where’s my order, and has it shipped yet?
• Do you do this? Do you cover my area?
• What does it cost, and what’s included?
• How do I cancel, reset, or change my details?
• Can you send me that link again?

If you read that and thought “constantly,” here’s the reframe: that isn’t a customer service issue. It’s a capacity leak. And it’s draining out of your team’s best hours, because questions arrive during business hours, which are the same hours your people would otherwise be doing the work you hired them for.

Count it before you decide anything

Don’t guess at this. Leaders are terrible at guessing at this, and the guess is always low.

Take one week. Have a few people keep a rough tally of the questions they field, in two columns. First column: how many. Second column, and this is the one that matters — how many had an answer that already existed somewhere. On a page. In a document. In a policy someone wrote two years ago.

That second number is the uncomfortable one. Because that portion isn’t work at all. It’s retrieval. You’re paying skilled people to be a search function.

Most leaders expect the answer to be about twenty percent. It’s routinely double that.

Three moves that actually shift it

First, write the answer down once. If three people on your team answer the same question three different ways, you don’t have an answer — you have folklore. Folklore is also why nobody can hand the task off; it only lives in whoever’s been there longest.

Second, put the answers where the questions actually get asked. Not in a shared drive nobody opens. The reason customers ask is almost never that the information doesn’t exist. It’s that finding it took more effort than typing a message and waiting.

Third, hand off the retrieval layer. This is where a lot of smaller teams now put an assistant on their website that answers instantly from the pages and documents they’ve already written, at whatever hour the question arrives. The point isn’t headcount. It’s that a whole category of interruption stops reaching a human at all.

What has to stay human

Be strict about this line, because getting it wrong costs more than the problem you were solving.

Anything where the honest answer is “it depends” stays with a person. So does every complaint, every frustrated customer, every judgment call, every conversation where money or a relationship is genuinely at stake. Anything where being heard matters more than being answered.

And test for the one real failure mode before you trust anything: an assistant that guesses. A system that says “I’m not sure — let me get someone” is far more valuable than a cleverer one that fills the silence with something plausible and wrong. Confident and wrong is worse than no answer, especially about a price or a policy. So ask it the awkward, undocumented questions on purpose and see what it does.

Two honest warnings

This won’t fix a broken process. If your shipping is genuinely slow, answering faster just delivers bad news more efficiently. Speed on top of a real problem makes the problem more visible, not smaller — which is useful, but it isn’t what most people are hoping for.

And it will expose how little your organization has written down. Most of your operation is sitting in three people’s heads, and you’ll find that out the moment you try to put the answers in one place. That’s uncomfortable. It’s also a risk you already had — you just hadn’t named it. If those three people leave, so does the answer.

A note for the people who fix this for a living

Some of you don’t run the organization with this problem. You advise it. You consult, you coach, you run an agency, and you walk into a different version of this leak every few months on somebody else’s payroll.

If that’s you, there’s a decision worth making deliberately rather than by default. You can keep pointing clients at whatever tool you happen to like, or you can fold it into what you deliver under your own brand, so the thing your client uses every day has your name on it instead of a vendor they’ve never heard of. The first option is a recommendation you give away. The second is part of your actual offer, and it keeps you in the relationship.

One caveat, and it’s the whole thing: only do that if you’re genuinely willing to support it. Putting your name on software you won’t stand behind is a fast way to own every problem it causes without any of the credit when it works.

You can’t hire your way out of this

The usual response is to add a person to absorb the volume. It works, for a while. Then volume grows back to fill the new capacity, and now you’ve got two people doing shallow work instead of one.

Worse, the new hire’s entire day is the interrupt layer. So they burn out too, and your capacity problem quietly turns into a turnover problem — which is the same problem, only more expensive and harder to see coming.

Your best people didn’t join to answer the same six questions. They joined to do the work only they can do.

Protect that, and a surprising amount of what you’ve been calling a burnout problem turns back into a capacity problem — which is the kind you can actually solve.