"Spend 7–10% of revenue on marketing" is the standard answer and it is close to useless, because it ignores both your margin and whether you could serve the customers it would produce.
Here is the arithmetic we actually use.
Start with capacity, not revenue
How many additional customers could you serve next month without hiring?
That is the ceiling. Advertising beyond it produces inquiries you cannot service, which costs money, staff time and reviews.
Say you could take 15 more patients.
Then your acquisition cost
From your own numbers, not a benchmark — the clinic version is here and the home services version here.
Say $180 per patient.
15 × $180 = $2,700 a month. That is the budget, and it has nothing to do with a percentage of anything.
Budget is capacity multiplied by acquisition cost. Revenue percentage is a rule of thumb for people who have not measured either.
The floor nobody mentions
There is a minimum below which Meta campaigns do not work, and it is not about fairness — it is about the learning phase.
A campaign needs a meaningful number of conversions per week before delivery optimization stabilizes. Below roughly $1,500–2,000 a month for most local service businesses, you are spreading spend too thin for the algorithm to learn, and results are erratic in a way that looks like bad management and is actually insufficient data.
If your capacity-based budget comes out below that floor, the honest options are to spend the floor or not to run Meta ads.
We put our own gate at $3,000 a month all in for exactly this reason, and we turn away budgets below it. It is not a preference — it is that we would be taking money for a campaign that cannot work yet, and being blamed for it later.
When to increase
When cost per customer is comfortably below your ceiling and your capacity has room. Then you are buying customers at a discount and more spend is straightforwardly good.
Increase in steps of 20–30%, not doubles. A large jump resets the learning phase and performance dips before it recovers.
When not to
When you cannot serve what you have. Fix capacity first.
When response time is poor. Adding budget to a business that answers leads the next morning buys more leads to waste. It is the largest lever in the operation.
When cost per customer is above your ceiling. Something is broken — offer, page, follow-up, targeting. More budget makes it worse faster.
If your budget was set as a percentage or as whatever felt affordable, working it out from capacity and acquisition cost usually moves it — and not always upward.