How Much Should a Startup Spend on Marketing? Budget Ranges by Stage

1. Blog
2. Measurement & Budget

By Saroj Jha · June 8, 2026 · 9 min read

My working marketing budget bands for startups, from pre-seed to Series B, and how to pressure-test the number with LTV:CAC and payback.

7.8%10–30%

• Large-company benchmark (7.8%)
• My working range for startups (10–30%)

share of revenue spent on marketing

Benchmark vs reality

7.8%

of revenue — the large-company benchmark

The benchmark everyone quotes is 7.8%.

In my working bands, most startups put somewhere between 10% and 30% of their base into marketing, but the base itself changes with stage, and that matters more than the percentage. Before product-market fit I’d budget against cash burn (30–60% of it). At Seed I’d budget against the round you raised (10–20% of it, spread across its runway). At Series A I’d budget against ARR (20–30%), and at Series B against revenue (12–20%). These are my operating ranges, not a published benchmark. Whatever number you land on, check it against LTV:CAC and CAC payback. Those two numbers tell you whether the spend is an investment or a leak.

The rest of this post explains where those bands come from, why the benchmark you keep finding doesn’t fit a startup, and how to test your own number.

Why doesn’t the “7–9% of revenue” benchmark apply to you?

The figures founders usually find are accurate. They just describe different companies.

• Large companies: Gartner’s 2026 CMO Spend Survey^[1] puts marketing budgets at 7.8% of company revenue, up slightly from 7.7% in the 2025 edition^[3]. The 2026 survey covered 401 marketing leaders in North America, the UK and Europe, and the vast majority worked at companies with more than $1 billion in revenue. Treat it as a large-company reference point, not a startup benchmark.
• US companies of all sizes: The CMO Survey’s 2026 report^[4] (Duke Fuqua, Deloitte and the AMA; 308 marketing leaders at for-profit US companies, surveyed in January 2026) puts marketing at 9.0% of company revenue on average, down from 9.4% the year before. That is a mean: the median answer was 5%^[5].
• Private B2B SaaS: SaaS Capital’s 2026 spending benchmarks^[2], drawn from more than 1,000 private B2B SaaS companies, put the median marketing spend at 8% of ARR, unchanged from 2025. The same survey found equity-backed companies spend 100% more on marketing than bootstrapped ones: roughly double.

Almost none of that applies to a startup building demand from zero, and the reason is the denominator. A pre-revenue company has no revenue to take a percentage of. A Series A company growing 3× has a revenue figure that will be out of date in two quarters. A bootstrapped founder and a venture-backed one at the same stage are solving different equations. That SaaS Capital gap reflects a different time horizon, not a lack of discipline: the venture-backed company is using investor capital to buy growth sooner.

So the useful question isn’t “what percent?” It’s “what does my stage require, and can my unit economics support it?” In my view, stage shapes the right budget more than industry or company type does. I start from a stage range and then pressure-test it against the numbers.

Does a marketing dollar really work harder early on?

There’s some evidence that it does. A meta-analysis in the Journal of Marketing Research ^[6] pooled 751 short-term advertising elasticities from 56 studies published between 1960 and 2008. It found an average short-term elasticity of about 0.12, which means advertising moves sales, but modestly. It also found that elasticity is higher in the early stage of a product’s life cycle than in the mature stage.

One caveat: those estimates come from direct-to-consumer brand advertising, not B2B startups. I read the finding as directional support for a higher early budget, not as a number to plan with.

What should a startup budget for marketing at each stage?

The table below shows AAJ’s working bands, which are my view and not a published benchmark. They cover marketing only, not sales and marketing combined. For comparison, SaaS Capital’s 2026 medians are 15% of ARR for sales and 8% for marketing, about 23% together; medians don’t add exactly, so treat the sum as approximate^[2].

Read the denominator column carefully. Early on you’re spending against cash burn or the round you raised. Later you’re spending against ARR or revenue. Mixing those up makes any benchmark comparison meaningless.

My early-stage bands sit well above SaaS Capital’s 8% median^[2] because a young company’s ARR is small, so the same dollars show up as a much larger percentage. My mature band sits below Gartner’s 7.8%^[1] because Gartner counts the whole marketing budget at mostly $1B+ companies across industries.

Do this

Find your stage in the table, apply the range to your own base, and write down the dollar figure and what it buys. Then test that figure against LTV:CAC and payback, below, before you commit it.

Free tool

Put your own numbers through these bands

The free Marketing Budget Calculator asks for your stage and base, then gives a budget and a channel split.

How does this work at pre-seed and before product-market fit?

A percentage of revenue means nothing here because there’s barely any revenue. You’re spending against cash. Almost everything should be founder-led: you write the content yourself, run small paid tests to find a channel that converts, and show up where your first customers already spend time. If you’re spending six figures on marketing before product-market fit, in my view you’re usually buying noise.

What does the Seed budget pay for?

This is channel-validation money. You’ve raised, you have runway, and the job is to find one or two acquisition channels that work repeatably before you pour fuel on them. Expect to fund early paid media, a baseline content engine, and basic CRM and analytics. Every dollar should be a data point, not a bet.

Running case: Example Co. Fictional

Example Co. raised a $2M seed round planned to last 18 months. At 15% of the round, marketing gets $300,000 over the runway. That comes to about $200,000 a year, or roughly $16,700 a month.

What changes at Series A?

This is the most important budget transition in a startup’s life. You’re moving from founder-led, scrappy acquisition to a repeatable demand engine. That means your first dedicated marketing hire or a serious agency relationship. It also means the demand-gen infrastructure (attribution, lifecycle, paid at scale) that makes growth predictable. When Example Co. reaches $3M ARR at Series A, 25% of ARR is $750,000 a year, about $62,500 a month. Companies that stay on seed-era budgets at Series A starve the exact engine investors just paid them to build.

What changes at Series B and beyond?

At Series B, growth and efficiency have to coexist for the first time. The budget funds full-funnel coverage, brand, events and a real team, but you’re expected to show payback. This is where sloppy attribution starts to cost real money: misallocating 20% of a $5M budget is a $1M mistake. Past Series B, the absolute budget keeps climbing. The percentage of revenue can even tick up as you fund new products and markets, which surprises founders who expect the number to only fall as they scale.

How do you pressure-test your number?

A stage range tells you whether you’re roughly in the right neighborhood. Your unit economics tell you whether the specific number is right for you. The academic literature treats marketing the same way. A review of 285 articles at the marketing–finance interface^[7] in the International Journal of Research in Marketing covers the research that links marketing actions to financial outcomes such as firm value. In practice, two metrics do most of the judging.

LTV:CAC ratio

The common reference point is 3:1, meaning three dollars of lifetime gross profit for every dollar of fully loaded acquisition cost. David Skok’s SaaS Metrics 2.0^[8] says the best SaaS businesses run above 3, sometimes as high as 7 or 8. Bessemer’s Scaling to $100 Million^[9] recommends investing in acquisition once the ratio is 3x or better. Treat 3:1 as a widely used rule of thumb, not a law. If your spend is below your band but LTV:CAC is healthy and climbing, you’re probably underinvesting. If your spend is above your band and the ratio is under 3:1, you’re buying unprofitable growth.

CAC payback period

How many months does it take for a customer’s gross profit to pay back what you spent to acquire them? Skok treats 12 months as the line^[8] and says profitability looks anemic beyond it. Bessemer sets the bar by customer segment: under 12 months for SMB, under 18 for mid-market and under 24 for enterprise^[9]. At Seed, I’d weight payback most heavily, because it’s measured in the same currency as your runway. At Series A and B, I’d hold it to the Bessemer band for your segment before adding budget.

How should you split the budget across channels? Once you have a number, a common heuristic is 70/20/10: 70% to proven channels, 20% to emerging channels showing early promise, and 10% to genuine experiments. It’s a rule of thumb, not a researched ratio. I use it because it keeps the core engine funded while making sure you’re building the next channel before the current one saturates.

Quick gut check

• Your spend sits inside your stage’s range.
• Your LTV:CAC clears 3:1.
• Payback is within the band for your segment.

If all three hold, you’re well calibrated: focus on execution, not the budget line. The number only needs surgery when those signals disagree.

How should you size a social media budget?

Set the total marketing budget first, using the stage bands above or the Marketing Budget Calculator. Social is a share of that number, not a separate one. I don’t give a percentage for social, because it depends on whether social is a main channel for your buyers or a side one. The useful work is splitting social into its three costs and funding one platform properly before adding a second.

• People time

  Writing, posting, replying. This is the largest cost of organic social, even though no invoice arrives. Price it at the hours it takes.

• Creative

  Design, video and editing, in-house or freelance.

• Paid

  Boosts and ads. Size this line with the Paid Media Budget Allocator, not by gut feel.

Start with one platform, the one where your buyers already look (see Which marketing channels should a startup start with?). Fund it for a six to eight week test with a stop condition written before you start: the cost per conversation, demo or signup you’ll accept. Add a second platform only after the first has a cost per outcome you can live with. These test lengths and the one-platform rule are my rule of thumb, not a benchmark.

Example Co. is fictional; the hourly rate, creative cost and stop condition are illustrative inputs, not benchmarks. The total is the sum of the three lines: $2,000 + $600 + $800.

Running case: Example Co. Fictional

On Example Co.’s $750,000 Series A budget, a quarter is $187,500, and the 10% experiments share in Figure 5 is $18,750 of it. The $3,400 test is under 2% of the quarter and about 18% of that experiments share. Example Co. treats it as one of its experiments, not as a new permanent line. If the test meets the stop condition, social moves to the 20% emerging bucket next quarter.

Experiments, 10% of the quarter

$3,400 the social test $18,750 experiments share

Do this

Write the stop condition before the first post. Track people time as a cost, even if it is the founder’s time. For what to post and what to measure once it’s funded, see Social Media Strategy for Startups.

What mistakes waste startup marketing budgets?

A few patterns come up again and again when a budget isn’t working.

• Copying an enterprise number. If you benchmark a seed-stage company against a 7–8% large-company average, you’ll be badly underfunded at exactly the moment you need to find a channel.
• A budget that’s all payroll. If 60% of your “marketing budget” is headcount and 15% is actual programs, you have a salary line, not a growth engine. My working split for a growth-stage team is roughly 40–50% people, 20–25% technology and 30–35% programs. That’s my rule of thumb, not a benchmark.
• Cutting reflexively in a downturn. The research here is more conditional than the folklore suggests. A Journal of Marketing study by Srinivasan, Lilien and Sridhar^[10] studied US public companies from 1969 to 2008, across seven recessions. It found that whether spending more on advertising and R&D in a recession pays off in profits and stock returns depends on the firm’s market share, financial leverage and product-market profile. My takeaway for a startup: don’t cut across the board. Cut what can’t show payback and protect what can.
• Deciding on last-touch attribution. If the final click is the only thing you can measure, you’ll overfund bottom-funnel conversion and starve the brand and content work that fills the funnel. Incrementality testing is the cleaner way to settle which channels actually cause pipeline.

For the fuller data set by stage, company type and channel, see AAJ’s Marketing Budget Benchmarks 2026 report.

If your question is how to get more from the budget you already have, read Is your marketing budget working hard enough?

If you want me to look at your own funnel, positioning and channels, my Growth Audit ends with three prioritized fixes and a 30-day action plan.