Search this question and you'll find the same recycled answer everywhere: "spend 5–10% of revenue on marketing." It's not wrong, exactly. It's just useless — a rule written for averages, applied to businesses that aren't average.
A better budget comes from two numbers you can work out this afternoon: what a customer is worth to you, and what you can afford to pay to acquire one.
Step 1: Know what a customer is worth
Not the first sale — the relationship. A salon customer spending $60 a visit, eight visits a year, for three years, is worth $1,440, not $60. This number (lifetime value, LTV) is the ceiling on everything else.
Step 2: Decide what you'll pay for one
A common healthy target: acquire customers at one-third or less of their lifetime value. Our salon can rationally pay up to ~$480 to win a customer — which suddenly makes a $30 cost-per-lead with a 20% close rate ($150 per customer) look like a bargain rather than an expense.
Step 3: Budget = target customers × allowed cost
Want 10 new customers a month at $150 acquisition cost? Your working media budget is $1,500/month plus the cost of whoever runs it. That number came from youreconomics — not a percentage rule invented for someone else's business.
The two budgets people forget
- The learning budget. A new channel needs 2–3 months of data before it can be judged. If you can only fund one month, don't start — you'll quit at the worst possible moment and conclude the channel "doesn't work."
- The infrastructure budget. Tracking, landing pages, follow-up automation. Unsexy, one-time-ish, and the reason identical ad spends produce wildly different results for different companies.
When percentage rules ARE useful
As a sanity check, after the math: growth-stage businesses commonly land between 7–15% of revenue; mature ones 5–8%; aggressive launches north of 20%. If your unit-economics budget lands wildly outside those bands, re-check the assumptions — usually the close rate is optimistic.
The real answer
Spend the amount your unit economics justify, protect a learning period long enough to get truth, and put infrastructure before volume. If you don't know your numbers well enough to run this math, that is the diagnosis — and it's the first thing a growth audit produces: your LTV, your real acquisition costs by channel, and the budget they justify.