How Much Should a Small Business Spend on Advertising?

As much as you can spend while the next customer still costs less than that customer is worth to you, and not a cent more. That is the whole rule. It is not a percentage of revenue, because revenue does not tell you what a customer is worth — margin does, and two businesses with identical revenue can have completely different budgets once margin enters the picture.

Every page answering this question gives you a percentage: five to twenty percent of revenue, adjusted for your stage or industry. That number is popular because it is easy to calculate and useless for the same reason — it never once asks what a customer is worth to you, which is the only number that actually decides whether a given budget makes sense.

How much should a small business spend on advertising?

The budget that makes sense is the one where you keep finding customers at or under your break‑even cost, and you stop the moment you cannot. That is a moving target, not a fixed number you set in January and leave alone, and it has nothing to do with what percentage of last year’s revenue feels reasonable.

Getting to that number takes two things: knowing what a customer is allowed to cost you (your ceiling), and knowing how many customers exist at or under that cost (your capacity). The ceiling is arithmetic you can do from your own numbers. I have already laid it out in full in what a customer should cost you, with a worked example line by line. This post assumes you have that number, or will get it, and answers the question that post does not: how big should the whole budget be.

Why doesn’t the 5% to 20% of revenue rule work?

Because revenue says nothing about what is left after you make the thing you sell. Take two businesses doing $50,000 a month in revenue. The 10% rule gives both of them a $5,000 advertising budget. One of them runs 70% gross margin and has $35,000 left to work with after cost of goods; the other runs 30% margin and has $15,000. A $5,000 budget is a reasonable fraction of the first business’s margin and more than a third of the second’s. The rule cannot tell the two apart, because it was never built from margin in the first place.

The rule has a second problem, quieter than the first: it sets the number once a year or once a quarter, as if the right budget were a constant. It is not. The right budget moves with how many customers are available to buy at your price this month, which is a market condition, not a fraction of your own income statement.

What actually sets the ceiling on what you can spend?

Your break‑even cost per customer, which comes from your margin, not your revenue. If you have not calculated it, the short version from the full derivation is this: take your contribution margin per order — revenue minus cost of goods, shipping, payment processing and a returns allowance — and that is roughly what a new customer can cost you before the first order stops paying for itself. Everything about how big your total budget should be is downstream of that one number.

Once you have the ceiling, the budget question becomes: how many customers can you reach this month at or under that cost? That number is not fixed either. It depends on how competitive your keywords and audiences are right now, which changes week to week and is the entire reason platforms run on auctions rather than fixed rate cards.

Is the budget unlimited if my margin can cover it?

No, and this is the part the margin framing leaves out if you stop at the ceiling alone. Ad auctions get more expensive as you buy more of them, because you exhaust the cheapest available customers first and the next ones cost more to reach. A budget that clears your ceiling comfortably at $2,000 a month can fail it entirely at $10,000, on the exact same account, because the marginal customer at $10,000 of spend is not the same customer you were getting at $2,000.

Google is explicit that ads run as an auction where cost depends on competition for the same keywords and audiences, which is why choosing your bid and budget is framed as an ongoing decision rather than a number you set once. The same logic sits underneath Target CPA bidding, which lets you set the average cost you are willing to pay per conversion and have the platform find as much volume as it can without going over it — which is, in effect, automating the exact ceiling‑and‑capacity question this post is about.

How do I know when I’ve hit my real spending limit?

Track your cost per new customer on the most recent slice of spend, not the account average, and watch it against your ceiling as you add budget. The account average can sit comfortably under your ceiling for months while the newest dollars you are spending are already over it, because the average blends in the cheap customers you acquired earlier.

The test that ends the argument: raise spend by a fixed amount, usually 15 to 20%, and watch what happens to the trailing‑7‑day cost per new customer over the following week or two. If it stays under your ceiling, you had room. If it jumps past it, you found your capacity for this month, and the right move is to hold there rather than keep pushing on the logic that a bigger number always works harder.

What if I don’t have enough sales data to calculate this yet?

Start with what you can afford to lose entirely, not with a formula, because a formula built on data you do not have yet is a guess wearing a number’s clothing. A new business has no margin history reliable enough to set a precise ceiling, and no cohort old enough to show a repeat rate.

The honest approach: pick a test budget bounded by cash you can afford to spend without a return, run it for long enough to get a real cost‑per‑customer reading — usually several weeks, not several days — and use that reading to build the margin ceiling the moment you have enough orders to calculate cost of goods and returns with confidence. The test budget is not the ongoing budget. It exists to produce the number that sets the ongoing one.

Does the platform change the answer?

The ceiling does not change by platform — a customer is worth the same to your business regardless of where you found them. What changes is how fast you hit your capacity on each one, because the competitive pool and the auction dynamics differ by channel. I cover the practical difference in Google Ads versus Meta Ads for ecommerce, and the short version is that the right split is a number you measure, not one you decide in advance.

What does not change across channels: the discipline of watching the newest dollars, not the account average, and stopping at capacity rather than at a percentage someone else published. That discipline is most of what I actually do for the accounts I run, which is the part of ecommerce strategy that looks the least like strategy and matters the most.

Questions people ask