The Side-Activity Threshold Is a Rate, Not an Amount

Why the turnover cancels out

The condition people think they are testing is whether the money coming in beats the money going out. That comparison is almost always won, because a side activity with no employees and no premises has very little going out — which is why so many of them look profitable and feel like a poor deal. The condition that actually matters includes a third term nobody writes down: the hours, and what they would otherwise have produced.

Once it is there, the algebra does something useful. Revenue is a rate multiplied by hours; the alternative is a different rate multiplied by the same hours. Divide both sides by the hours and they are gone. What remains is a comparison of two rates, and no quantity of turnover appears anywhere in it. That is not a modelling trick — it is the reason the question "from what turnover?" has never had a satisfying answer and never will.

What an hour keeps of what it bills

Two deductions stand between a billed euro and a kept one, and they compound rather than add. Direct costs come off the top — materials, a platform's commission, the marginal cost of delivering the thing. Contributions and tax then apply to what is left. At 30% direct costs and 25% on the margin, a euro billed keeps 52.5 cents, and the required billing rate is the alternative value divided by that fraction.

What the first year does to the answer

Setup hours are real and nobody bills them: registering, learning the platform, making the first samples, writing the terms, the two clients who took a long conversation and bought nothing. On the profile above, sixty of those hours across a first year of 416 billable ones drop the realised rate from 25 euros an hour to 21.85 — a shortfall of 13% that lands exactly when people are deciding whether the thing is worth continuing. Judging the activity on its first year judges it at its worst, and abandoning it there is the commonest way a viable side business ends.

The alternative value is not always a wage

This is where the calculation becomes personal rather than accounting. If the hours would have been overtime at a known rate, or a second shift you could actually have taken, the alternative value is that rate and the comparison is hard-edged. If they come from an evening you would have spent doing nothing chargeable, the honest figure is close to zero and almost any activity clears — which is not self-delusion, it is the correct answer to the question actually being asked.

The one value that is never zero is rest. Hours taken from sleep are borrowed rather than free, and the repayment shows up in the main job, which is usually where the money actually comes from. An activity that clears the threshold only by cannibalising the thing that pays for everything else has not cleared it.

The regime threshold is a different question

There is a second threshold in this subject and it is not the same one. Every country sets a level of activity above which an occasional sale becomes a declared business, with registration, contributions and often value added tax attached. That threshold is a real amount, it is set by an instrument, and crossing it changes the arithmetic above by changing the deduction rate. It is covered on this site in its own article, because it answers a legal question rather than an economic one — and conflating the two is why people describe an activity as "not worth declaring" when they mean it is not worth doing, or the reverse.

Frequently asked questions

So there is no turnover figure at all?

There is one, but it is an output rather than a test. Once you have fixed the hours you are willing to give and the rate your hour must beat, the required turnover follows — and it is usually larger than people expect, which is the useful part. Eight hours a week against a 25-euro alternative comes to about 19,800 euros a year. What you cannot do is start from a turnover figure and conclude anything, because the same turnover over twice the hours is half the deal.

What should I use as the value of my hour?

The best available use of that specific hour, not your salary divided by working hours. A Saturday morning is not an hour of your job and cannot be sold to your employer, so your hourly wage is the wrong figure unless overtime is genuinely on offer. Ask instead what you would actually have done and what it was worth: a second contract you turned down has a price, a rested Monday has a price you can estimate from how much your main work suffers, and an evening in front of a screen has a price close to nothing.

Does the answer change once the activity grows?

The rate condition does not, but three of its inputs do. Direct costs usually fall as a share of revenue with volume, which lowers the required rate. Contributions often rise as a share, because regimes designed for small activity give way to ordinary ones — which raises it. And the alternative value of the hours rises, because past a certain commitment the hours stop coming from spare evenings and start coming from the main job. Those three move in different directions, which is why an activity can pass the test at a small scale, fail it in the middle, and pass it again as a full occupation.

How do I count the hours honestly?

Count every hour the activity causes, not the hours spent producing. The producing hours are the ones people count and they are usually the minority: quoting, invoicing, chasing payment, buying supplies, packing, answering questions from people who never buy, the tax return at the end of the year. Recording a single normal month honestly is enough — most people find the total is between a third and twice again what they assumed, and the direction is almost never downward.

The activity fails the test but I enjoy it. Is that wrong?

No, and the arithmetic is not an argument against it. What the calculation gives you is the price of the enjoyment, stated per hour, which is a far better thing to know than a vague sense that it ought to pay for itself. An activity that returns 12 euros an hour against an alternative worth 25 costs you 13 euros an hour to do — and plenty of things are worth that. The failure mode this article exists to prevent is the opposite one: believing it pays, scaling it on that belief, and finding out at the point where the hours have started coming from somewhere that mattered.