There is no going rate for marketing a small business — there is a going rate for attention in your category, and your budget is whatever it takes to buy enough of that attention to cover the weeks your calendar would otherwise be thin.
That sounds like a dodge. It isn't. Every budget figure you've read in an article was reverse-engineered from someone else's auction, someone else's close rate and someone else's margin, then flattened into a percentage that reads like advice. The useful version of this question is narrower and answerable in an afternoon: what does one interested stranger cost where you operate, how many of them do you need before one of them pays you, and how much is that payment worth. Those three numbers give you a budget with a reason attached. A percentage of revenue gives you a number you'll abandon the first slow quarter.
Start at the click, not at the retainer
Marketing pricing has two layers and most owners only get shown the top one. The top layer is labor — the retainer, the hourly rate, the per-project fee. The bottom layer is media: what it costs to put your name in front of someone who is actively looking. The bottom layer is set by an auction you don't control, and it is the same for you as it is for the competitor who outbids you. You can negotiate labor. You cannot negotiate the auction.
Where the money actually goes
When a proposal arrives as a single monthly number, ask for it split. Every real marketing spend divides into the same four categories, and each one gets priced differently and proved differently. A vendor who can't split their number is either bundling media into their fee or hasn't thought about it, and both are worth knowing before you sign.
| What you're paying for | How it gets priced | What proves it worked |
|---|---|---|
| Media | An auction, per click, set by your competitors | Cost per booked job — not cost per click |
| Creative and copy | Per asset, or folded into a monthly fee | Which version people actually respond to |
| Management | Retainer, or a share of what you spend | The changes made, and what each one moved |
| Measurement | Set up once, then maintained | You can name where your last five calls came from |
The question has a season
This isn't a question people ask idly. Across seven distinct ways of phrasing it, roughly 340 searches a month come from owners trying to price this decision. The volume isn't flat, either: it peaked at 110 searches in March and bottomed out at 50 in July. That shape tells you something about who's asking. Budget questions cluster when the year is being planned and when the pipeline looks uncertain — not when the phone is ringing. If you're reading this in your busy season, you're early, and early is the only time this decision gets made calmly.
Your ceiling is set by what a customer is worth
Work it backwards instead of forwards. Take the revenue from your average job, subtract what it costs you to deliver it, and you have the margin on one new customer. Now estimate how many genuine inquiries you turn into work — most owners know this within a reasonable range even without a CRM. Divide that margin by your close rate and you have the absolute most you can pay for one inquiry before the sale stops being worth having. Everything below that ceiling is a business decision about how fast you want to grow. Everything above it is a leak, no matter how good the reporting looks. If you can't complete that calculation, the first thing to buy isn't advertising — it's the ability to see which calls came from where.
Do these in order, starting this week
First, write down your ceiling per inquiry, even as a rough figure, and put it somewhere you'll see it. Second, make sure inquiries are tracked to a source — calls, forms, whatever your customers actually use — so next quarter's version of this question has evidence behind it instead of a hunch. Third, pick one channel and fund it properly for long enough to read a result; a budget split across four channels usually produces four inconclusive experiments. Only then does the size of the number matter.
Hiring help is a reasonable option at exactly one point in that sequence: when you know your ceiling, and the work of staying under it — auction management, tracking that survives a website change, reading results honestly — costs you more of your own billable hours than it costs to hand over. That's a math problem, not a leap of faith. If you're not there yet, the cheapest useful month you can spend is the one where you finally find out what a customer is worth to you.