Operations · Founder Leadership

Not Everything Should Escalate to You. Some Things Absolutely Should.

·10 min read ·OperationsFounder Leadership
A business owner sits at her desk with a laptop and notebook, calmly considering several documents, a phone and a tablet being held out toward her from both sides.

You have managers now: a sales manager, an operations lead, someone running the office. The org chart looks like a company that has grown up. And still, by Thursday afternoon, your desk holds a strange collection.

A scheduling question someone in operations could have answered. A dispute between two departments that now needs you to “weigh in.” An employee problem a manager has been circling for a month. A serious customer situation. And every so often, the thing you should have heard about three weeks ago, arriving now, fully grown.

It’s tempting to read that as one problem: too much still comes to me. It isn’t. Some of what reaches you absolutely should. Some shouldn’t be there at all. And some should have arrived much earlier.

The useful question isn’t how much is reaching you. It’s what is reaching you, how it gets there, and when it arrives. If you suspect the deeper issue is that authority never really moved when you hired managers, start there. This article is narrower: reading what still lands on your desk, and what it says about the management around you.

Escalation isn’t the problem. Mis-calibration is.

Some escalation is exactly what you want. The economist Luis Garicano described hierarchy as a way of matching problems with the people who can solve them: routine problems get handled close to the work, and rarer, harder ones move up. That isn’t failed delegation. It’s what the structure is for.

So zero escalation isn’t the goal. A founder who never hears about problems doesn’t necessarily have a well-managed company. They may simply have one that stopped telling them things. The goal is calibration: ordinary management work gets absorbed where it belongs, and what leadership genuinely needs to know about or decide arrives early enough to matter.

Too much of the wrong thing

Over-escalation is the version founders feel. Questions a manager could have answered travel upward. Discomfort travels upward dressed as risk. Problems come around managers instead of through them. It accumulates until the founder is the default resolver of everything nobody else wanted to own.

Too little, too late

Under-escalation is the version founders rarely see, almost by definition. A job slipping, a customer relationship cooling, a margin eroding: something material builds without reaching leadership until it has become a much bigger problem—or doesn’t reach them at all.

Both failures can exist in the same company at once. Passing up a routine question costs a manager little; passing up bad news about something they own can cost a great deal. A team can be quick with the small things and slow with the serious ones.

A stream of paper notes pours into a copper funnel on a desk, while a single warning document sits apart, off to the side.

Being informed isn’t the same as being handed the problem

One distinction matters more than volume. A manager who tells you a project is slipping and what they’re doing about it is keeping you informed. A manager who tells you it’s slipping and asks what you want done has handed you the problem.

Healthy teams often send more information upward, not less. Amy Edmondson’s research found that teams where people felt psychologically safe did more asking for help, seeking feedback, and discussing errors. What deserves a closer look is responsibility moving upward that should have stayed put.

Not everything on your desk is asking for a decision

The things reaching you are not all the same kind of thing, even when they arrive in the same tone of voice. Some are genuine decisions that need authority only you hold. Some are risks: they may not need a decision yet, but they need you to know, because the exposure is ultimately yours. Some are referee requests, where two parts of the business disagree and you’re the only tiebreaker both will accept. And some are hand-backs, where a manager has something that’s theirs to handle, and it comes back to you anyway.

The first two belong with you, at least some of the time. The second two usually don’t. Real issues are often mixed, and the point isn’t to sort them into boxes. It’s to stop treating everything that arrives as the same request.

When two departments need a referee

Some of the most persistent escalations I see have little to do with any one manager’s ability. They start at the boundary between two departments.

In one business, sales believed that once a quote or proposal went out, following up with the prospect became the administrative team’s job. Admin saw it differently. The disagreement wasn’t about whether follow-up should happen. It was about who owned it, and the follow-up sat in the gap between two assumptions.

In another, sales treated a signed contract as the finish line. Once it was handed to operations, responsibility for the customer and for completing the job well belonged to operations alone.

Hands gesture toward a stack of wooden blocks sitting on a table between a Sales sign and an Operations sign.

The customer doesn’t experience two departments. They experience one company. When problems surface after a handoff like that, they tend to become arguments about whose problem it is, and those arguments go to the person above both departments. In many founder-led companies, nobody else sits there, so the founder becomes the standing referee, settling a boundary nobody drew clearly.

One distinction I find useful: sending work to another function is not the same as that function accepting it. “I sent it to operations” describes an action; operations formally accepting ownership describes a transfer of responsibility. Until that happens, the work is in transit, and work in transit tends to belong to nobody.

In both examples, each department knew its activities; it was far less clear who owned the outcome. A repeat dispute over ordinary work marks a boundary nobody owns. Not every referee request is a symptom. When two functions pull apart because the business faces a real trade-off between priorities, settling it is your job.

When the hard part is the conversation

Here is a pattern I’ve seen more than once. A manager knows an employee’s behavior or performance needs addressing, but avoids the direct conversation or never formally corrects it. The issue lingers until it reaches the owner or a senior leader, who ends up having the conversation the manager ordinarily would have.

That looks like a decision being escalated. Often, though, the manager doesn’t need the founder’s authority to address ordinary performance or behavior. What travels upward is discomfort: the interpersonal risk of the conversation itself.

Problems drifting back up the chain are not a new idea. William Oncken Jr. and Donald Wass captured it in “Management Time: Who’s Got the Monkey?” (Harvard Business Review, November–December 1974): a subordinate raises a problem in the hallway, the manager promises to think about it, and the “monkey” moves to the manager’s back. What I’d add from operating experience is that what gets handed up isn’t always a decision. Sometimes it’s the hard conversation itself.

In a laboratory study published in Science, people were less willing to make a choice when the outcome also affected others, a reluctance tied to wanting more certainty rather than to risk aversion. The study didn’t involve managers, so this is my interpretation: a manager facing a conversation that affects someone’s job is in a similar spot, and waiting for certainty, or for someone else to act, can feel safer than acting.

So ask whether the manager needed authority or wanted cover. If they lack the authority to formally correct someone, that’s a structural gap. If they have it and want someone else to carry the risk, the founder who steps in solves today’s problem and confirms the route for the next one.

Not every personnel issue that reaches you is avoidance. New managers may need coaching through their first few hard conversations. Serious matters, such as potential terminations, legal exposure, or conduct concerns, often should involve senior leadership or HR. The pattern to watch is the ordinary performance conversation that keeps arriving with your name on it.

When it comes around your manager instead of through them

Route is a separate question. A decision, a referee request, or a hand-back can reach you through the manager, or around them.

Going over a supervisor’s head has been studied as “leapfrogging,” and researchers caution that it can further strain relationships already under strain.

The clearest illustration comes from Google, which in 2002 experimented with eliminating engineering managers. According to David Garvin’s account in Harvard Business Review, the experiment lasted only a few months, ending when too many people went directly to Larry Page with questions about expense reports, interpersonal conflicts, and other nitty-gritty issues. Remove the layer, and routine issues go straight to the top.

In most founder-led companies, the layer exists; the question is whether people believe it’s the route that works. My interpretation, from operating experience, is that founder responses shape that belief. When going around the manager reliably gets a faster answer, it becomes the route, which connects to how founders teach their teams what to bring them.

Some bypass is legitimate. Concerns about the manager, safety issues, and ethical problems need a route around the manager, and skip-level conversations that keep you informed are healthy. Hearing something directly isn’t taking the problem back; what matters is whether responsibility returns to the manager.

Some things should reach you, and sooner than they do

Some matters genuinely belong at leadership level:

- safety; - legal, regulatory, and licensing obligations, including sign-offs that legally belong to a license holder; - material financial exposure; - major customer risk; - significant personnel actions; - strategic trade-offs between competing priorities; - situations the business hasn’t encountered before.

Where exactly the lines fall in a particular business is a design question of its own. What matters here is whether these reach you at all, and when.

Under-escalation rarely announces itself, and it isn’t always about anyone hiding anything. I’ve seen it show up in operational assessments: incomplete project controls make it hard to identify problems before they become scheduling, customer, margin, or owner problems. The business compensates through direct communication, checking, workarounds, escalation, and owner involvement. When the structure doesn’t surface problems early, the owner becomes the fallback.

Late escalation, in other words, can be structural rather than personal. If nothing makes a slipping job visible until it’s a customer problem, the first escalation will arrive late, however conscientious the people are. Finding where problems fail to surface is diagnostic work before it’s anything else.

Other fields have measured the cost more directly. A systematic review in surgery linked delayed escalation of care to higher mortality, with hierarchy and communication failures among the causes. In aviation, 1,751 crew members reported staying silent in half of the speaking-up situations they recalled, mainly from fear of damaging relationships, fear of punishment, or operational pressure.

Neither study measures founder-led businesses, and neither suggests that half the problems in your company go unreported. They show a principle: important information doesn’t travel up a hierarchy automatically, and the reasons people hold it back are often relational. The aim is for leadership to be informed early enough to act, without the problem being handed upward. A manager who flags a material risk early and keeps owning it is doing exactly what you want.

What the pattern on your desk is telling you

What reaches you, how it gets there, and when it arrives say more about your management environment than the volume ever will. Recurring referee requests point to unowned boundaries. Repeated hand-backs point to managers who hold the title but not yet the comfort, skill, or authority for the harder parts of the role. Traffic around managers points to a route people have learned works. Late arrivals point to a structure that doesn’t surface problems early.

One pattern sits underneath many of these. I regularly encounter roles that exist on the org chart, sometimes with a job description, without a clearly defined deliverable or accountability. People know their activities but not the results they’re responsible for. When something falls between functions or misses a deadline, the question becomes: whose responsibility was this? If nobody can answer cleanly, it defaults to the one person responsible for everything.

That escalation is a symptom, not the original problem. I’ve written before about one reason this happens: when no one has named the priority, no one can truly own the outcome, and decisions drift back to the founder. Much of what reaches you fits that explanation. Some of it, like a handed-back hard conversation or a dispute at a departmental boundary, adds to it. One client situation on this site involved a team that escalated nearly everything, not from lack of judgment, but because no one had defined what they could decide.

Read the pattern fairly. New managers escalate more while they’re learning, and a crisis can pull decisions upward for a while; both should reverse. Sometimes the manager lacks the information to decide, and the fix is getting it to them (economists Philippe Aghion and Jean Tirole argued that real authority tends to follow information). Sometimes your involvement is simply rational, because you hold the relationship, the expertise, or the risk. And sometimes there just aren’t enough people to do the work.

Hiring managers wasn’t necessarily the mistake; in most growing companies it’s the right move. But a management layer isn’t judged by how quiet the founder’s desk becomes. It’s judged by whether ordinary management work stops where it belongs, while what genuinely needs you arrives early enough to act on.

The next time something lands on your desk, ask what it actually is, how it got there, and whether it arrived when it should have. Ask those questions consistently, and a pattern starts to emerge. That pattern will tell you far more than the frustration ever will.

Sources

Not sure what should still be reaching you?

If too many decisions, problems, and handoffs still find their way back to your desk, we can look at where they're coming from—and what the pattern is telling you.

You don't need to diagnose the problem before we talk. That's part of the conversation.

Book a 30-Minute Strategy Call