An owner walked me through her plan to roughly double the business over eighteen months. Eleven people on payroll at the time, three full-time and eight part-time, every one of them on a different availability pattern that she personally accommodated.
The revenue plan was sound. The thing I thought would break first was not sales, capacity or cash. It was the schedule.
Doubling a team does not double the coordination. In my experience it is closer to four or five times, and the reason is worth understanding, because it determines whether the growth plan survives contact with a busy month.
Why it multiplies
Coordination cost does not scale with the number of people. It grows with how many things have to be true at the same time.
Each part-time person carries their own availability window, their own preferences, their own standing commitments and their own tolerance for short-notice change. Each job carries its own timing requirement, its own skill requirement and its own customer expectation. Building a week means finding an arrangement that satisfies all of them simultaneously.
Add one more person and you have not added one more thing to check. You have added another set of conditions that collides with every one already there. That is why the tenth person feels so much heavier than the fourth, and why owners describe hitting a wall rather than a slope.
The practical version: if you are spending five hours a week on the schedule at eleven people, do not budget ten hours at twenty-two. Budget considerably more, and budget it for somebody other than yourself.
What actually breaks
Not the schedule itself. The schedule survives, because the owner absorbs it. What breaks is everything after it:
Jobs get missed. Not through negligence. Through a handoff that depended on somebody remembering.
Quality drifts. People get placed by availability rather than by fit, so work goes to whoever is free rather than whoever is right.
Customers feel it. Reschedules, unfamiliar faces, inconsistent standards. Everything that goes wrong anywhere else ends up in front of a customer, which is why they notice the strain before anyone outside the business does.
The owner disappears into it. Scheduling is unbounded work, distributed across every hour of the day. Ten hours of it consumes more capacity than fifteen hours of production, because it fragments everything around it. You cannot do owner-level thinking in the gaps between availability texts.
And the good people leave first. Reliable staff notice when unreliable colleagues are accommodated, and they draw conclusions about the standard.
The flexibility trap
Here is the part that makes this hard, and it is not an operations problem.
Owners who build accommodating schedules are usually doing it on purpose. They want the business to serve the people who serve it. They remember being on the other side of an inflexible employer. Flexibility is often genuinely why they have retained good people through a period when they could not compete on wages.
That is a real asset and I would not be quick to trade it away.
But there is a jump that has to happen, and it is this: the flexibility that works at eight people is not available at twenty. Not because you stopped caring. Because the arrangement was always being subsidized by the owner’s own hours, and there is a finite supply of those.
The business can still serve its people well. It just cannot do it through unlimited individual accommodation, because that particular form of generosity has a hard ceiling and the ceiling arrives without warning.
What replaces it is structure, not strictness. Predictable schedules are themselves a benefit — people can plan childcare, second jobs and study around a known block far better than around a variable one. Owners consistently underrate this. They experience adding structure as taking something away, and staff frequently experience it as finally being able to plan.
What to build instead
1. Availability blocks, not full-time mandates
The instinct is to solve this by requiring everyone to go full-time. That is expensive, usually unnecessary and will cost you good part-time people you should keep.
The requirement is not more hours. It is predictable hours.
“I need to know you are available eight to one, Monday to Friday” is a completely different ask from “I need you full-time.” The first is schedulable. The second is a pay rise you may not need to make.
Define a small number of standard blocks and have people commit to one. Two or three patterns covering most of your operating window is usually enough. The gain is not the hours, it is that the schedule becomes solvable in advance rather than negotiated weekly.
Tradeoff: you will lose some people who genuinely cannot commit to a block, and a few of them will be good. That is a real cost. It is smaller than the cost of a schedule nobody can run but you.
The exception: a genuinely exceptional person with an awkward pattern can be worth an exception. Just make it an exception you decided, priced and can name, rather than a default that quietly becomes the policy.
2. Change requests need notice, and notice needs teeth
A rule with no consequence is a preference.
Set a notice period for schedule changes — a week or two, depending on how far ahead you build. Then hold it, including the first time it is inconvenient to hold it. The first enforcement is the one that establishes whether the rule is real.
3. Keep the generosity, add a boundary
This is the piece I would most encourage owners not to skip.
A time-off policy of “everything is approved, just give me some notice” sounds generous and usually is, right up until the schedule cannot absorb it. Then every refusal becomes a personal exception rather than a policy, which means the owner almost never refuses, and the schedule takes the strain instead.
Retain the right to decline, and you gain the ability to approve. Once declining is a normal operational answer rather than a rejection of the person, you can say yes to most requests without the yes costing you a week of rework. The generosity survives. What changes is that it becomes affordable.
4. Give scheduling to one person, with a number
This is the actual fix, and everything above is preparation for it.
Scheduling should be owned by one person who is not you, with real authority to say no and a measure they are held to. Something like schedule density and same-week reschedule rate.
The change this produces is bigger than the hours it saves. It stops the schedule being emotional.
When you own the schedule, every request is a relationship. A long-tenured employee asks for Friday and you are weighing their circumstances, your history and how you will feel saying no.
When someone else owns it against a real standard, the same request is arithmetic. How many people are available, how many jobs need covering, did the request arrive with enough notice. If it did not, the answer is no, and it is not a referendum on anyone.
That is not coldness. It is what makes consistency possible, and consistency is what staff actually want from a schedule.
Tradeoff: decisions will be made differently than you would make them, and some will be worse. That is the price of the hours you are buying back. There is more on what a real handoff requires in delegation is not distribution.
5. Know how many people one person can actually watch
Once there are enough people to warrant a layer of oversight, plan the shape rather than discovering it.
A single person can typically oversee somewhere in the range of fifteen to twenty people across six to eight active sites. That means this layer stays small for a long time — two people can often cover a business several times your current size.
That range is a rule of thumb from businesses I have worked with rather than a benchmark. It varies with how geographically concentrated the work is and how well documented your standards are. Tight geography and good documentation push it up; scattered sites and tribal knowledge push it down hard.
Sequence it before the growth, not during
The reason to do this while the team is small is that every one of these changes is a negotiation with existing staff, and negotiations are much easier when you are not simultaneously trying to deliver a record month.
If you are planning meaningful growth, the schedule work belongs about two quarters ahead of the hiring, not alongside it. You need the availability structure to exist before you onboard people into it, or you are teaching new hires a system you are still arguing about.
A rough order:
- Define the availability blocks and the notice rule. Write them down.
- Migrate existing staff onto blocks, one conversation at a time, with a lead time. Some will need a few months.
- Apply the rules to new hires from day one. Every hire made under the old arrangement is another conversation you will have to have later.
- Hand scheduling to one owner with authority and a number.
- Then hire into the growth.
Owners want to run this in reverse, because hiring feels like progress and scheduling policy feels like admin. The reversed order is how you end up with twenty people, no structure and a schedule only you can run.
Frequently asked
Do I have to make everyone full-time? No, and it is usually the wrong fix. Part-time works well at scale when the availability is predictable. What breaks is not part-time hours, it is per-person negotiated patterns.
What if I lose people when I introduce blocks? Some will go, and it is worth being honest that a few may be people you value. Give a long lead time, involve them in choosing which block, and make the reasoning explicit. In my experience most people accommodate a predictable structure more easily than owners expect, because predictability is itself worth something to them.
Is scheduling software the answer? It helps, and it does not solve this. Software applies rules; it does not decide what the rules are. Digitising a set of individually negotiated arrangements gives you the same problem with a subscription attached. Set the structure first, then buy the tool.
How do I know when I have hit the wall? Two signals. You cannot build next week’s schedule without several conversations, and a single person calling off requires you personally to rebuild the day. Either one means the schedule is being held together by you rather than by a system.
Who should own scheduling if I only have a few people? Below roughly six or seven people, the owner holding it is usually still reasonable. It is worth writing down how you do it well before you hand it over, because the handoff is much easier from a documented process than from memory.
Should the scheduler also be the field manager? Early on they often are, and it works. They separate as you grow, because scheduling is a planning function and field oversight is a quality function, and the person who is good at one is not reliably good at the other.
The schedule is usually the first thing that breaks in a growth plan, and it is almost never the thing the plan accounts for. If you want help sequencing this against your own hiring timeline, here is how I work.