How Much Should I Spend on Advertising to Get Results?

Almost every advertising budget starts life as a guess. Someone picks a number that feels survivable — $2000 a month, $5000, $10k— and the campaign gets built backwards from it. Three months later the honest question surfaces: was that number ever right?

It probably wasn't, because the number was never the decision. The decision is what you need the money to do. A budget that ignores your margin, your average order value and your channel's minimum efficient spend isn't a budget. It's a deposit on a lesson.

This guide gives you the three legitimate ways to set an advertising budget, the arithmetic behind each one, and the floor below which a channel physically cannot produce results — no matter how good the creative is.

Key Takeaways

• Benchmarks set the ceiling, not the number. The CMO Survey's January 2026 edition puts marketing at 9.0% of revenue, while Gartner's 2026 CMO Spend Survey reports 7.8% — the gap exists because they survey different company sizes.
• The only budget method that survives contact with reality is bottom-up: start from the customers you need, work back through conversion rate and cost per click.
• Every channel has a minimum efficient spend. Meta's optimisation requires roughly 50 conversions per ad set per week to exit the learning phase — below that you are paying for volatility, not performance.
• A 3:1 lifetime-value-to-CAC ratio is the widely used solvency line. Below it you are buying revenue at a loss you haven't measured yet.
• Budget and time are interchangeable. Halving your spend does not halve your results — it usually delays the learning that makes results possible at all.

What's the short answer on how much to spend?

If you need a number today: most small and mid-sized businesses should commit 7–12% of gross revenue to marketing, with roughly 50–70% of that going to paid advertising. On $1M in revenue that means $70,000–$120,000 in total marketing, of which $35,000–$85,000 is media spend — call it $3,000–$7,000 a month.

That range is a starting position for a planning conversation, not an answer. It ignores your margin, your sales cycle, whether you are defending a position or trying to take one, and whether your category's click costs are $1.63 or $9.87. Use it to sanity-check a number someone has already proposed. Do not use it to set one.

The rest of this guide replaces that shortcut with math you can defend.

Where do the percentage-of-revenue benchmarks actually come from?

Two surveys anchor almost every "marketing budget percentage" article on the internet, and they disagree — which is useful once you know why.

Gartner's 2026 CMO Spend Survey reports marketing budgets at 7.8% of company revenue. The CMO Survey, in its January 2026 edition, reports 9.0%. The difference isn't a measurement error. Gartner surveys large enterprises, most above $1 billion in revenue. The CMO Survey includes a much broader base of smaller US firms. Smaller companies spend a higher share of revenue on marketing because they are buying awareness that larger companies already own.

That pattern continues downward. Analyses of sub-£10M businesses put marketing spend nearer 16.8% of revenue, and the US Small Business Administration has long guided firms under $5M in revenue toward 7–8% of gross revenue — but only when net margins are healthy, in the 10–12% range or better.

The B2B/B2C split matters more than most planners allow for. B2B companies typically run 6–7% of revenue; B2C companies run 9–12%. A B2B firm copying a DTC brand's budget ratio is over-spending against a longer, relationship-led sales cycle.

Here is the structural problem with all of it. A percentage of revenue is a backward-looking constraint. It tells you what you can afford based on what you already sold. It says nothing about what it costs to acquire the next customer in your specific market, which is the only question a media budget actually answers.

Treat these figures as a guardrail. If your bottom-up calculation lands at 3% of revenue, you are probably under-investing and will stagnate. If it lands at 40%, you are either a venture-funded land-grab or you have a margin problem that advertising will amplify rather than fix.

What are the three ways to set an advertising budget?

There are only three defensible methods. Most businesses use the weakest one.

1. Top-down: percentage of revenue

You take last year's revenue, multiply by a percentage, and divide by twelve.

Where it works: mature businesses with stable, predictable revenue and an established channel mix. It is a budgeting method — it answers "what can we afford?"

Where it fails: growth businesses. The method is circular. Low revenue produces a low budget, which produces low growth, which produces a low budget next year. It also allocates identical spend to a month with a product launch and a month with nothing happening.

2. Bottom-up: goal-backwards from unit economics

You start with the number of customers you need, then work backwards through your conversion rates and click costs to the media spend required to produce them.

Where it works: everywhere you have — or can estimate — a conversion rate and an order value. This is the method serious media buyers use, and it is the one worked through in full below.

Where it fails: genuinely new products with no conversion data at all. Even then you can borrow category benchmarks and treat the first 60 days as a paid research project with an explicit learning budget.

3. Competitive parity and share of voice

You estimate what competitors spend and match or exceed it, on the logic that share of voice tends to lead share of market.

Where it works: defending a category position, or entering one where a small number of players dominate the auction. Useful as a cross-check on the other two.

Where it fails: as a primary method. You cannot see competitors' margins, their lifetime value, or whether their spend is working. Matching a competitor who is quietly losing money is an expensive way to copy someone's mistake.

The practical answer: set the budget bottom-up, sanity-check it against the percentage-of-revenue guardrail, and use competitive intelligence to decide whether the auction you are entering is realistically winnable at that number.

How do you calculate an advertising budget from unit economics?

This is the calculation that replaces the guess. It runs in five steps, and you need four numbers to start: your average order value, your gross margin, your website conversion rate, and your channel's cost per click.

Work through it with a real example — a DTC apparel brand, because the category has well-documented benchmarks.

Step 1: Establish what a customer is worth

Fashion and apparel brands typically run 50–60% gross margins. DTC Shopify stores show a median AOV in the $85–95 range, though category averages vary widely — accessories run $30–150, luxury fashion $200–1,000+.

Take a $90 AOV at 55% gross margin. Each first order contributes $49.50 in gross profit.

Now the number most brands skip: returns. Apparel return rates run 20–40%, with US clothing averaging around 26%. A 30% return rate means the effective contribution per order attempt is $49.50 × 0.70 = $34.65.

Step 2: Set your maximum allowable CAC

The widely used solvency line is a 3:1 LTV-to-CAC ratio. If a customer's lifetime gross profit is three times what you paid to acquire them, the business services its overhead and still funds growth.

Assume this brand's customers buy 2.2 times over their first 24 months. Lifetime gross profit is $34.65 × 2.2 = $76.23. At a 3:1 ratio, maximum allowable CAC is $25.41.

That number is going to feel uncomfortably low, and it should — this is exactly where most fashion brands discover their problem. Reported fashion CAC benchmarks span $31.80 to $187, depending heavily on sub-category and execution, with $30–80 the common band. A brand with a $25 allowable CAC operating in a category where acquisition realistically costs $45 does not have an advertising problem. It has a pricing, retention or margin problem that no media buyer can spend its way out of.

This is the single most valuable output of the exercise. Run it before you spend, not after.

Step 3: Work backwards to required traffic

Say the brand fixes its economics — raises AOV to $120 through bundling, improves repeat rate to 2.8 orders — and lands at an allowable CAC of $43.

At a 2.2% site conversion rate, one customer requires 1 ÷ 0.022 = 45 visitors.

Allowable cost per visitor is therefore $43 ÷ 45 = $0.96.

Step 4: Check that number against real auction prices

Now the reality test. Meta ads across industries average around $0.70 CPC for traffic campaigns, with the cheapest clicks in shopping, collectibles and gifts near $0.34. Google Ads search CPC averaged $2.96 in Q1 2026, up from $2.64 a year earlier, with cross-industry figures reported as high as $5.42 and a spread from $1.63 in arts and entertainment to $9.87 in legal services.

At $0.96 allowable cost per visitor, this brand can operate on Meta. It cannot profitably run broad Google search at $2.96 — not on first-order economics. That is a channel-selection conclusion falling directly out of the budget math, which is exactly how it should work.

Step 5: Multiply up to the monthly budget

Target: 200 new customers per month.

• 200 customers × 45 visitors = 9,000 visitors
• 9,000 visitors × $0.70 CPC = $6,300 per month
• Implied CAC: $6,300 ÷ 200 = $31.50 — inside the $43 allowable

You now have a budget with a reason attached. If someone asks why it's $6,300, the answer isn't "it felt right." It's "200 customers at a 2.2% conversion rate and a $0.70 click, with $11.50 per customer of headroom against our allowable CAC."

What is the minimum budget a channel needs to work at all?

This is the constraint that ruins more small budgets than bad creative ever has, and almost nobody accounts for it.

Modern ad platforms are machine-learning systems. They need a minimum volume of conversion events to model who converts. Below that threshold the algorithm is guessing, and your results swing wildly week to week for reasons that have nothing to do with your ads.

Meta's advertiser documentation puts the threshold at roughly 50 optimisation events per ad set per week to exit the learning phase. That gives you a hard formula for your floor:

Minimum viable monthly budget = target CPA × 50 × 4.3 weeks

At a $31.50 CPA, that is 31.50 × 50 × 4.3 = $6,772 per month — per ad set. A brand running $1,500 a month across four ad sets is generating around 12 conversions per week spread across four learning phases, none of which will ever stabilise.

When your calculated budget lands below the channel floor, you have four honest options:

1. Consolidate. Run one ad set instead of four. Fewer, better-fed campaigns beat many starved ones.
2. Optimise for a cheaper event. Optimise toward add-to-cart or lead rather than purchase — more events at lower cost — then move up the funnel once volume supports it.
3. Change channel. Search captures existing demand and works at lower volume because intent does the targeting work the algorithm would otherwise need data for.
4. Wait and accumulate. Three months of budget spent in one month often outperforms the same money spread across three.

What does not work is running a sub-threshold budget indefinitely and concluding the channel doesn't work for your business. You never tested the channel. You tested the learning phase.

How long until advertising produces results?

Budget and time trade against each other, and the exchange rate is not linear.

A realistic timeline on a properly funded account looks like this:

• Weeks 1–2 — Learning. Volatile CPAs, algorithm exploring. Judge nothing. Any decision made here is noise-driven.
• Weeks 3–6 — Stabilisation. Ad sets exit learning, CPAs settle into a band. First real creative signal appears.
• Weeks 7–12 — Optimisation. Winning creative and audiences identified, budget reallocated. This is where efficiency gains actually land.
• Month 4+ — Compounding. Retargeting pools are populated, creative testing has a library behind it, and blended efficiency improves even when channel-level metrics look flat.

Halving the budget does not double the timeline — it can extend it indefinitely, because a starved account may never accumulate enough events to exit learning at all. This is the single most common cause of "we tried Facebook ads and they didn't work."

Plan for a minimum 90-day commitment at or above the channel floor. If you cannot fund 90 days at the floor, you cannot fund the channel. Pick a different one.

Where does the money actually get wasted?

Across audited accounts, budget leaks cluster into five patterns that recur at every spend level.

Spreading spend too thin

Five channels at $1,000 each will underperform one channel at $5,000, every time. Fragmented budgets keep every channel below its learning threshold simultaneously.

Confusing traffic with customers

Traffic campaigns optimise for the cheapest click, not the most likely buyer. A $0.34 CPC that converts at 0.2% is more expensive per customer than a $1.20 CPC converting at 3%. Optimise for the event that makes you money.

Ignoring the conversion side of the equation

If your site converts at 1.1% and the category median is 2.2%, doubling your conversion rate has exactly the same effect on CAC as halving your click cost — and it is usually cheaper and faster to achieve. Budget spent on the landing page is often better spent than budget added to the auction.

Killing campaigns during the learning phase

Turning off an ad set on day four because CPA looks bad destroys the accumulated learning and restarts the clock on the next one. You pay the learning tax repeatedly and never collect the return.

Budgeting for acquisition and forgetting retention

Your allowable CAC is a function of lifetime value. A brand that lifts repeat purchase rate from 1.4 to 2.4 orders has just increased its allowable CAC by more than 70% — which buys back auction competitiveness no bid strategy can match.

How should your budget change as you grow?

Budget structure should shift with company stage, not just scale linearly.

One threshold worth knowing: incrementality testing — the discipline of measuring what would happen if you switched a channel off — generally becomes worth its cost once monthly spend passes roughly £25,000–£50,000. Below that, blended marketing efficiency ratio is the more practical measurement tool.

Frequently Asked Questions

Is there a minimum budget worth starting with at all?

Yes, and it is set by your CPA rather than a universal figure. Multiply your target cost per acquisition by 50, then by 4.3, to get a monthly floor for a single Meta ad set. If that number is unaffordable, optimise for a cheaper conversion event or use search, where buyer intent substitutes for algorithmic learning volume.

Should I include agency or software fees in my advertising budget?

Track them separately but evaluate them together. Media spend buys impressions; management fees buy the decisions about those impressions. Keep the line items apart so you can see channel efficiency clearly, but always assess return on the combined figure — that is the number that actually leaves your account.

What percentage of my budget should go to testing?

A common structure allocates 70% to proven channels, 20% to scaling what is emerging, and 10% to genuine experiments. At small budgets, protect the 10% — it is the only thing preventing your entire account from depending on a single channel's pricing.

My competitor spends far more than me. Can I still compete?

On volume, no. On efficiency, frequently. A smaller advertiser with a sharper offer, a tighter audience and a better landing page can hold a lower CAC than a larger one buying broad reach. Compete on the specificity of the segment you serve, not the size of the auction bid.

How do I know if my budget is too low versus my ads just being bad?

Check conversion volume first. If your ad sets are producing fewer than 50 optimisation events per week, the budget is the binding constraint and creative quality cannot be assessed reliably. Above that threshold with poor results, the problem is the offer, the creative or the landing page.

The bottom line

An advertising budget is an output, not an input. It falls out of four numbers you already have or can estimate: what a customer is worth, what share of that you can afford to spend acquiring them, how many visitors it takes to produce one, and what those visitors cost in your auction.

Run that calculation before the next campaign brief. If the resulting number sits above your channel's learning floor and inside your percentage-of-revenue guardrail, you have a fundable plan. If it sits below the floor, you have learned something more valuable than any campaign would have taught you — and you have learned it before spending the money.