Why the busiest month is often the least profitable one

Every business has a month where everybody worked flat out and the numbers came back worse than the quiet month before it. It is disorienting enough that owners usually blame timing or an accounting artefact. It is normally none of those, and there are four usual causes.

First, the work you take at the peak is different work

In a quiet month you take the jobs you want. In a busy one you take what is in front of you, which means smaller jobs, unfamiliar jobs, jobs with awkward specifications and jobs for customers you would normally decline. Each of those carries more setup, more supervision and more rework per dollar than the work you would have chosen.

This shows up nowhere in the monthly accounts, because the accounts do not know which jobs were which. It shows up immediately if you sort the month's jobs by size and look at the small end — and in most businesses that we look at, the smallest quarter of jobs by value consumes a wildly disproportionate share of the setup, the supervisor's attention and the mistakes.

Second, setup does not scale and it is rarely priced

Almost every small business prices with an hourly figure applied to the work, and almost every one of them under-counts setup. Setup is the changeover, the reading of the drawing, the collection of the materials, the telephone call to clarify, the first attempt that establishes how this one is going to go. On a large job it is a rounding error. On a small one it can be most of the job.

In a busy month you do more jobs, so you do more setups, so the proportion of the month spent on unpriced work goes up. The business is not less efficient; the mix has moved towards the part that was never being paid for.

The test is simple and slightly painful. Take the smallest ten jobs of your busiest month and the largest ten. Work out, for each group, the total hours actually spent — including the setup, including the chasing — against what was invoiced. Most owners doing this for the first time discover that the small end of the business is being funded by the large end, and that nobody had any way of knowing.

Third, overtime is not a linear cost

A busy month buys capacity at a premium, and it buys it at exactly the moment the business is least able to supervise it. The premium is visible in the payroll; what is not visible is the second-order effect — tired people make more mistakes, mistakes create rework, and rework consumes the capacity the overtime was bought to provide. That loop is the reason a fifteen per cent increase in throughput can cost a great deal more than fifteen per cent.

Fourth, the peak is when the informal system fails

Every small business runs partly on things one person knows and holds in their head. That works well at normal volume and it fails suddenly, not gradually. In the busy month the person who knows is interrupted forty times a day, the informal check they always did quietly does not happen, and the errors that system was silently preventing all arrive at once.

Owners usually experience this as the month everything went wrong. It is more useful to read it as the month the business found out where its real capacity limit was, which is not the machine and not the floor space — it is the attention of two or three people.

What to do with the answer

The instinct is to add capacity, and the peak is the worst time to make that decision because the business is at its least readable. The more useful move is to make the month legible afterwards.

• Sort the month's jobs by value and compare the top and bottom quarters on hours actually spent. That comparison is the whole argument, and it takes an afternoon.
• Price setup separately from the work, so a small job carries its own preparation instead of being subsidised by a large one.
• Set a minimum job value, which is a decision about who you are rather than a pricing tweak, and expect to lose some work — that is what it is for.
• Write down what the two or three people at the centre actually do when they are interrupted, because that is the real capacity limit and nobody has ever measured it.

A word about the number you are looking for

There is not one. We are not going to tell you that businesses of your size typically lose a certain percentage on small jobs, because we have not measured a population of businesses and a figure presented as if we had would be invented. The comparison that matters is between the top and bottom of your own month, and it is available to you this week without anybody's benchmark.