No small business designs its structure. It accretes one — each hire answering an urgent need, each reporting line defaulting to the owner, until one day there are twelve direct reports, three people doing overlapping jobs, and a "structure" that exists only as the pattern of who interrupts whom. Structure problems masquerade as people problems ("nobody takes ownership") and owner problems ("I'm in everything"), which is why they survive so long. Here's what the chart actually needs to look like at three sizes — and the transitions between them, which is where the pain lives.
At 10 people: one team, real seats
At ten, you don't need managers — you need clarity. The structure is honest about being a single team around the owner, but three disciplines change everything. First, seats, not people: write down the functions the business actually contains — sales, delivery/operations, admin/finance — and who owns each, even where one person holds two seats and the owner holds three. The chart's job at this size is exposing exactly which hats you're wearing, because the hat list is your future hiring plan in disguise. Second, one owner per outcome: every recurring outcome (quotes out, jobs scheduled, invoices raised, customers answered) has one name, and shared ownership is recognised as the synonym for no ownership that it is. Third, a deputy behind each critical seat — the cross-cover we bang on about, drawn onto the chart so it's a design rather than a hope. The classic ten-person failure isn't missing management; it's blurred seats, where three capable people half-own everything and the owner remains the routing layer for all of it.
At 20: the layer you can't skip
Somewhere in the teens, the maths breaks: an owner can genuinely lead perhaps six to eight people; past that, "everyone reports to me" means nobody is actually led — feedback gets thin, decisions queue, and your best people start managing informally without the authority to do it well. Twenty people requires the first real management layer: typically a delivery/ops lead and one other (sales lead or office/finance manager, depending on your shape), each owning their function's outcomes and its people.
Two truths about this transition, both learned expensively. Promoting your best doer is the default and often the mistake — the skills are different, and you risk losing a great technician to create a struggling manager. Sometimes the right doer can grow into it (with actual support: training, your time, explicit expectations that the job is now the team's output); sometimes the honest answer is hiring management experience and letting your best doer stay brilliant. Decide on the person, not the precedent. And the owner must actually vacate the layer. The structure fails not on paper but in the hallway — when staff keep bringing things to you and you keep answering, you've made your new managers decorative. The discipline is the warm redirect, a hundred times: "that's Sarah's call now — what does she say?" Your job narrows to direction, standards, the managers themselves, and the outside world. It will feel like loss for a quarter. It's the trade that buys every subsequent stage.
At 40: functions, rhythm, and the chart as strategy
By forty, the business is a structure of teams: delivery probably split (by service line, region or crew), sales and marketing distinct from operations, finance/admin a real function with its own controls, perhaps the first people-and-culture ownership. The owner now leads a leadership team of four or six — and the binding constraint shifts from spans of control to coordination: the structure only works if information and decisions flow between functions on a rhythm (the weekly leadership meeting, the monthly numbers, the quarterly planning — the operating cadence we've written about isn't decoration at this size; it's the nervous system the chart implies). New failure modes appear on schedule: silos (sales promising what delivery can't), the second layer's quality (your managers are now hiring managers — your standards survive only if the leadership team carries them), and the owner who skips levels, undermining the very structure they built every time they "just quickly" fix something three layers down.
Draw it for the business you're becoming
The practical instruction that ties it together: once a year, draw two charts — today's actual structure (who really reports where, hats and all) and the structure the business needs at its size eighteen months from now. The gap between them is your hiring and development plan, sequenced and dated. Structure drawn in arrears is archaeology; drawn ahead, it's the cheapest strategy document you'll ever produce — because most growth ceilings I've watched businesses hit weren't market ceilings or capital ceilings. They were charts that nobody redrew.
Owner-dependency and structural bottlenecks show up clearly in our free Business Health Check — five minutes to see which transition you're actually in.
If this is the problem you are working on, Valont’s People hub runs employer advice line and team culture & insights as part of a connected back office, and the award finder shows where to start.
About the author

Andrew Northcott
Founder & Chairman, Valont
Andrew Northcott is the founder and chairman of Valont and of its parent group, Wattlestone, a private Australian holding company that acquires businesses to own and operate for the long term. He has spent two decades building and running Australian SMEs. By 2009 he was running a labour-hire business that went on to become one of the largest privately owned labour-hire companies in its market before it was sold, an experience he describes as learning the weight of payroll, compliance and cash flow first hand. Along the way he completed the Owner/President Management program at Harvard Business School. He founded Valont in 2024 to make that kind of business capability accessible to owners who could never justify building it themselves. He writes about ownership, cash, people, systems and the decisions that compound.
LinkedIn →