When Growth Creates More Work Than Revenue

You have more customers than last year, more revenue than last year, and less room in the week than you had before either of those things was true.

Bigger, and worse to run

The specifics vary; the shape does not. Sales are up. So is the number of things that only you can resolve. You are back doing work you stopped doing two years ago, because it is quicker to do it than to explain it. Month-end takes longer than it used to. There are more conversations about who was supposed to do what, and more of them happen after something has already gone wrong.

The margin, meanwhile, has not moved much. Possibly it has moved the wrong way. And there is a quiet suspicion, usually unvoiced, that the last twenty customers were worth less than the twenty before them.

That suspicion is normally correct, and it is measurable.

Why hiring is the wrong first answer

Hiring solves capacity. It does not touch the reason capacity ran out. If the administrative load rises with every customer you take on, an extra pair of hands buys perhaps twelve months — after which you are in the same position with a larger payroll and one more person who needs coordinating. The coordination cost, note, has just gone up too.

This is not an argument against hiring. It is an argument for knowing which of the two problems you have before you spend forty thousand a year answering the wrong one.

Two cost lines, one of them invisible

Every business has costs that scale with revenue — materials, delivery capacity, the people doing the billable work. These are visible. They sit in your cost of sales, you priced for them, and when revenue doubles you expect them to roughly double.

Every business also has costs that scale with customer count — coordinating who does what, handling the case that does not fit, re-entering the same information into a second system, chasing people, and assembling reports by hand. These are invisible, because they are absorbed by salaried people who were hired to do something else, and by the owner, who is not counted at all.

Nothing in your accounts distinguishes these two lines. So when the second one grows faster than gross margin, growth converts itself into overhead, and the business gets busier without getting better. The good news is that you do not need a new accounting system to see it. You need one number, taken twice.

Instrument

Overhead per customer

A rough calculation, deliberately. You are looking for a direction, not a figure accurate to two decimal places — and the direction is legible even when the inputs are estimates.

1. 01Pick two monthsThis month, and the same month last year. Same month, so seasonality does not distort the comparison.
2. 02Count active customers in eachDefine “active” once — billed in that month, or under contract in that month — and apply the same definition to both. The definition matters less than using it twice.
3. 03Estimate non-billable hours in eachAdmin, coordination, chasing, exception handling, reporting, re-entering data. Ask the three people who do most of it for an honest weekly figure, multiply up, and add your own hours. Nobody will be exact. It does not need to be.
4. 04Divide, and compareNon-billable hours ÷ active customers, for each month. Then put the two numbers side by side.

Reading the result

FallingYou have operational leverage. Growth is making the business more efficient, and hiring ahead of demand is a reasonable bet.

FlatYou are renting your growth. Every new customer costs the same to service as the last one, so scale buys you revenue and no margin.

RisingGrowth is actively costing you. Adding customers is adding more overhead than contribution, and more salespeople will make it worse rather than better.

Then take the same hours and sort them into the four types below. Which type dominates decides what actually fixes it — and it is very often not what people expect.

Four kinds of work, and what each one responds to

The trap sits in the last row. Coordination is the loudest of the four — it is what fills the day and what people complain about — so it is what automation projects usually aim at first. It is also the one software helps least, which is why so many businesses buy a project-management platform, adopt it enthusiastically for six weeks, and end up coordinating the platform on top of everything they were coordinating before.

Re-entry is the quietest of the four and the easiest to remove. It is almost never the one anybody asks about.

What good looks like

• Overhead per customer falling year on year, even slightly.
• Information entered once, by the person closest to it, and never typed again.
• Exceptions arriving in one queue with an owner, rather than in an inbox with none.
• A customer able to get an answer without an employee retrieving it for them.
• Month-end producing itself, and being trusted without a manual check.
• The owner able to be away for a fortnight without the week being reconstructed on their return.

Where to start, cheapest first

1. Remove one re-entry. Find the piece of information typed into two systems most often, and connect them. Smallest project, largest immediate return, and it proves the case internally better than any business case would.
2. Automate the chasing. Payment reminders, document requests, approval nudges. These need no intelligence, only reliability, and they return hours from people who currently give them up quietly.
3. Make exceptions visible. One place where the cases that do not fit are queued, owned and counted. Counting them is the point — the count tells you which exception is worth designing out.
4. Then consider self-service. Only once the first three are done. Giving customers a portal on top of a manual back office does not remove work; it adds a channel through which more of it arrives.