Somewhere between your fifteenth and your fortieth employee, your calendar quietly becomes the company’s bottleneck. Nobody announces it. A team lead leaves and you absorb their reports “for now.” An AI agent quietly absorbs the status-reporting work that justified a coordinator role, so you don’t backfill. Twelve months later you have eleven direct reports, you’re still doing real work yourself most of the week, and execution somehow feels slower with thirty people than it did with eight. That is a span of control problem, and right now it is being created faster than most founders are solving it.
The instinct is to treat it as a headcount question — hire a VP, push people down a layer, redraw the org chart. Sometimes that’s right. Often it’s expensive and premature. The more useful question is narrower: what actually breaks when a manager’s team gets wider, and what holds it together?
The span of control math has already changed
This isn’t a projection. It’s already happened. Gallup’s most recent workplace research found the average number of people reporting to a manager rose from 10.9 in 2024 to 12.1 in 2025 — close to a 50% increase in average team size since Gallup began measuring in 2013. The median has stayed flat at roughly five to six, meaning the average is being pulled up by a growing tail of managers running very large teams.
The direction of travel is deliberate. Gartner predicted that through 2026, 20% of organizations would use AI to flatten their organizational structure, eliminating more than half of current middle management positions — and named the side effects plainly: managers overwhelmed with additional direct reports, and broken mentoring and learning pathways for junior staff.
For a venture-stage company, the flattening rarely arrives as a restructuring memo. It arrives as drift. You automate a workflow, decide not to hire the manager you’d planned, and inherit the coordination load personally. The cost doesn’t show up on the P&L. It shows up in decisions that take three weeks instead of three days.
Attention, not headcount, is the real ceiling
The most useful finding in Gallup’s span of control research — a study of 92,252 teams across 104 organizations and 46 countries — is that there is no universal right number of direct reports. What predicts whether a wide span works is how much of the manager’s own time is consumed by individual contributor work.
Gallup found that 97% of managers still carry individual contributor responsibilities, spending a median of 40% of their time on their own hands-on work. Below that 40% line, manager engagement stayed steady at roughly 37% regardless of team size. Above it, engagement fell as teams grew — down to 32% for managers running 25 or more people. Same headcount, very different outcomes, determined by what else was on the manager’s plate.
This reframes the problem. A founder with nine reports who spends 70% of the week writing code or closing deals is over their span. A well-supported operator with fourteen reports and a clear mandate may not be. The constraint is attention, not the org chart.
Three questions worth more than a reorg
- What percentage of my week is my own delivery work? If it’s meaningfully north of 40% and you have more than five reports, you have a structural problem, not a discipline problem.
- Which reports exist because of the work, and which exist because of history? Reporting lines calcify. A person who joined when you were nine people often still reports to the founder at forty for no reason anyone can articulate.
- Where is decision latency actually accumulating? Instrument it. Track how long approvals, hiring decisions, and scope calls sit. That number tells you where to add a layer far more honestly than a headcount plan does.
The habit that makes a wide span survivable
The same research isolated one manager behavior that outperformed nearly everything else: meaningful feedback delivered to each person at least once a week. Gallup reports this practice nearly triples the share of engaged employees. Across seven studies covering 44,025 responses, employees who strongly agreed they had received meaningful feedback in the past week were highly engaged — about seven in ten — regardless of team size. Those who didn’t strongly agree? Roughly one in four.
Read that again, because it’s the whole argument. Team size stopped mattering when weekly feedback was real. Gallup also notes these conversations don’t need to be long — 15 to 30 minutes, done consistently, is enough.
For a founder with a dozen reports, that’s roughly four to six hours a week of one-to-one time. That sounds like a lot until you price the alternative: a $200K executive hire brought in a stage early, which is a far more expensive way to buy back the same attention. If you can’t find those hours, the honest conclusion isn’t “I need a VP.” It’s that you are personally doing too much delivery work — see the previous section.
What this means for hiring in Canada and the U.S.
Widening spans collide with a genuinely tight market for experienced operators on both sides of the border. In Canada, ManpowerGroup’s 2026 research found 71% of Canadian employers struggle to find skilled talent. Canadian scale-ups also compete directly with U.S. multinationals for the same senior people, often on compensation they can’t match — which is why “just hire a strong VP” is a slower and less certain plan in Toronto or Vancouver than it looks on a slide in San Francisco.
Two practical consequences:
- Treat internal capacity as the first lever, not the last. Restructuring who reports to whom, killing meetings that exist to compensate for unclear ownership, and freeing a founder from delivery work are all faster than a four-month executive search.
- Protect the training ground. Gartner flagged it and it’s the least-discussed cost of flattening: if you delete the layer where people learn to manage, you are borrowing from your own leadership bench two years out. If you flatten, replace the lost apprenticeship deliberately — structured mentorship, real ownership on real scope, senior operators who coach rather than approve.
How to audit your own span this quarter
A concrete sequence you can run in a week:
- Log two weeks honestly. Categorize every hour as delivery, management, or strategy. Most founders discover the split is worse than they’d have guessed.
- Map current reporting lines against current work. For each direct report, write one sentence on why that line exists today. Anything you can’t justify in a sentence is a candidate to move.
- Check the feedback loop. Ask three people when they last had a conversation with you that changed how they were working. If the honest answer is “a while,” the span is already too wide for how you’re running it.
- Measure decision latency for 30 days. Then decide whether the fix is a hire, a delegation of authority, or a deletion of the decision entirely.
- Only then write the headcount plan. A hire made after this exercise has a defensible mandate. A hire made before it usually gets a title and no scope.
Most founders who think they need a VP need four hours a week back and a clearer set of decision rights. Some genuinely need the VP. The audit is what tells you which one you are.
Where we land on it
We take a view on this because we live inside it. Protocol 42 structures, builds and operates a small number of companies each year — and we deploy capital selectively into a few of them — which means we carry the operating consequences of every org decision rather than handing over a recommendation and leaving. Wide spans are survivable and sometimes preferable at early stage. What isn’t survivable is a wide span run by someone who has no attention left to spend on it. That combination doesn’t announce itself; it just slowly degrades the quality of every decision the company makes.
The discipline is unglamorous: know where your hours go, know why every reporting line exists, and protect the weekly conversation that makes the rest of it work. That’s most of the job. You can read more about how we build and operate companies, and if you’re wrestling with this at your own company, partner with us is the place to start the conversation.