If you've never run a paid ad campaign before, the first question isn't which platform to use or what to say in the ad. It's what number to put in the budget field. Type in too little and you'll never find out if the campaign works. Type in too much and you're gambling money you might need for payroll next month.
There is a workable range. For businesses under $5 million in revenue, a commonly cited range is 5% to 9% of gross revenue for total marketing spend, based on SBA-attributed small business guidance and The CMO Survey. Paid ads are typically one line inside that total, not the whole budget. But percentage-of-revenue math only gets you a starting number. The more useful question is whether that number is big enough to produce meaningful data once it reaches an ad platform, a separate calculation most budgeting advice skips.
What the benchmarks say
The percentage most often quoted comes from The CMO Survey, which put marketing spend at a mean of 8.96% and a median of 5% of revenue in January 2026, on 154 valid responses. A small number of businesses spending heavily on marketing pulls the average up, while most businesses spend closer to the median. If you're a smaller operation, the median is a more realistic comparison point than the mean.
These figures are not specific to paid ads. They cover the full marketing budget: your website, photography, email, social media management, and other marketing work, with paid ads as one piece inside that total. If you're spending 6% of revenue on marketing overall, the paid ads portion of that is usually smaller still, often somewhere between a third and half of the total, depending on how much you're already investing in your Google Business Profile, SEO, or content that generates traffic without an ongoing per-click cost.
Why a percentage alone can leave you underfunded
Percentage-of-revenue math doesn't account for the volume an ad platform needs before it can produce useful data. If your monthly budget only produces a handful of clicks, you don't have a failed campaign, you have a sample size too small to draw any conclusion from. A campaign that gets 15 clicks a month and zero conversions isn't evidence the ad doesn't work. It's evidence you haven't given it enough traffic to know either way.
That underfunding is why so many small businesses try paid ads once, see no results, and conclude it doesn't work for them. The budget was sized to fit inside a comfortable percentage of revenue, but it was never large enough to reach the volume where the platform's own bidding systems have enough data to start improving performance. A campaign needs a certain number of conversions in a rolling window before automated bidding tools have anything to learn from, and a budget spread too thin never gets there.
A more practical way to size the number
Instead of starting from a percentage, start from what a new customer is worth to your business and work backward.
- Figure out what one new customer is worth. Not just the first purchase, what they're worth if they stick around or refer others.
- Decide how many new customers you want paid ads to bring in during a test period. A 60 to 90 day window is usually enough to see a real pattern.
- Estimate what it costs to get a click in your industry and location, and what share of clicks realistically turn into a lead or sale.
- Multiply it out. If you want 10 new customers, expect 1 in 20 clicks to convert, and clicks run $3 each, that's 200 clicks needed, or roughly $600 for the test window, not counting anything spent refining the campaign along the way.
That number might land above or below what a straight percentage-of-revenue calculation would suggest. The percentage is a sanity check on affordability. The test-budget math is what tells you whether the number you're about to spend is large enough to produce a usable result.
Treat the first budget as a test, not a commitment
The first month or two of any paid ad campaign should be treated as a controlled test, not a locked-in monthly expense. Set the number, let it run long enough to gather real data, then look at what it cost to get a customer versus what that customer is worth. Adjust from there. Some budgets that look reasonable on paper need to go up to reach the volume where the numbers make sense, while some smaller, sharper campaigns outperform bigger, unfocused ones. You won't know which until you've run it and looked at the numbers, which is a separate skill from picking the starting budget.
Key Takeaways
- Revenue-based benchmarks put total marketing spend for a business under $5 million in revenue between 5% and 9% of gross revenue, with paid ads as one line inside that total, not the whole amount.
- The CMO Survey found a mean of 8.96% and a median of 5% of revenue spent on marketing across firms of all sizes in January 2026, and the median is the more realistic comparison for a smaller business.
- A percentage-of-revenue number only checks affordability. It doesn't check whether the budget is large enough to produce enough clicks and conversions for an ad platform's bidding systems to learn from.
- A more reliable way to size a first budget is to work backward from what a new customer is worth, how many new customers you want in a 60 to 90 day test, and what a click typically costs and converts at in your market.
- The first paid ad budget should be treated as a test period with a defined end date, not a permanent monthly commitment, so it can be adjusted based on actual cost-per-customer once the test produces data.
Mindstate Strategy's paid ads service handles paid ad budget sizing and the data decisions that follow for clients who want a number they can defend. If you're trying to figure out where to start, get in touch and we'll walk through what a realistic test budget looks like for your specific business.
