Ask ten businesses how they set their advertising budget and most will admit the honest answer: they picked a number that felt affordable, spent it, and hoped. That's not a budget — it's a shrug with a dollar sign. A real advertising budget is a decision tied to a goal, sized by a method you can defend, and split deliberately across the work that gets you there.
The takeaway up front: set the number against what you're trying to achieve, not against what you spent last year — then split it across channels and funnel stages on purpose, and always ring-fence a slice for testing. How much you spend matters far less than whether the amount is connected to a goal and pointed at the right work.
Sizing the budget: four methods and when each fits
There is no universal "right" percentage, so start by choosing how you'll set the figure. Four methods dominate, ordered here from least to most rigorous — and the more your revenue depends on advertising, the further down this list you should be.
- Affordable method. Spend whatever is left after other costs. It's the most common approach and the weakest, because it ties advertising to leftover cash rather than to opportunity — you under-invest exactly when growth is possible and overspend when you're flush. Its only real merit is protecting cash flow for a fragile business, and that's a survival tactic, not a growth plan.
- Percentage of revenue. Commit a fixed share of revenue — many established businesses land somewhere in the single digits to low double digits of revenue, higher for growth-stage or highly competitive categories. Its strength is simplicity and discipline: the budget scales with the business. Its weakness is that it's backward-looking — basing this year's spend on last year's sales caps growth and, worse, cuts advertising in a downturn precisely when you may need visibility most.
- Competitive parity. Set spend relative to what competitors appear to invest, using auction-price signals and share-of-voice estimates. Useful as a sanity check — if you're spending a tenth of what a category leader does, that explains a lot — but a poor primary method, because it assumes competitors know what they're doing and that their goals match yours. They rarely do.
- Objective-and-task. Start from the goal, work backward to the spend. Define the result you want, estimate the conversions needed to hit it, then use your funnel math — cost per click, conversion rate, cost per acquisition — to price it. This is the most rigorous method and the one to graduate toward, because it's the only one that starts from what you're trying to achieve. Its cost is that it needs data you may not have yet, which is why new advertisers often start with a percentage and switch to objective-and-task once they know their numbers.
The honest sequence for most businesses: use percentage of revenue to set a ceiling you can live with, use objective-and-task to check whether that ceiling can actually buy the goal, and use competitive parity to sanity-check both. If the goal needs more than the ceiling allows, the fix is a smaller goal or a longer timeline — not a hopeful budget.
Working backward from a goal
The objective-and-task method deserves a worked walkthrough, because it's where a budget stops being a guess. The chain is short: goal, to conversions, to spend.
Say the goal is 50 new customers in a quarter. If your landing page converts clicks to customers at 3%, you need roughly 1,650 clicks. If clicks in your channel cost around $2 each, that's about $3,300 in media — before creative, tools, and management. Now you have a number with a reason attached, and you can interrogate it: is the goal realistic at that conversion rate? Is $2 a click plausible in your category? Change any input and the budget changes with it.
The point isn't the arithmetic — it's that every assumption is now visible and testable. When results come in, you update the real conversion rate and cost per click and the budget self-corrects. A budget built this way improves every quarter; a budget picked from thin air just gets repeated. If you don't yet know your conversion rate or cost per click, that's your first project, and our advertising strategy guide covers setting up the goal and measurement that feed this math.
Splitting the budget across channels and stages
A total is only half the decision. The other half is allocation, and this is where most budgets quietly fail — money piles into whatever channel is easiest to buy rather than what the goal needs.
Split along two axes. First, by funnel stage: some spend builds awareness among people who don't know you, some captures demand from people already looking. A budget that's all bottom-funnel search harvests existing demand but never creates new demand, so growth stalls once you've captured everyone searching. A budget that's all top-funnel awareness generates interest it never converts. The right mix depends on your goal — a business that needs sales this quarter leans toward capture; one building a category leans toward creation.
Second, by channel, and here the rule from strategy holds: match the channel to the audience and the goal, and state the reason. Weight spend toward channels with the intent, reach, and measurability your goal requires, not toward the one you find most comfortable. Whatever the split, avoid spreading a small budget so thin across five channels that none gets enough to leave the learning phase. Concentration usually beats dilution when the budget is modest — better to win one channel than to be invisible on five.
Always reserve a slice for testing
Whatever the total, carve out a fixed share — a common working range is 10–20% — for deliberate experiments: new channels, new audiences, new creative angles. This isn't optional overhead; it's how the budget stays alive. Every channel decays, every audience saturates, and today's winner is tomorrow's tired ad. The testing reserve is what funds the next winner before the current one fades.
Treat it as a separate line so it doesn't get raided the moment the core campaign wants more. Its job is different: the core budget buys known results, the test budget buys information. Judge them by different standards — a test that "fails" by not scaling still earns its keep if it tells you where not to spend next quarter. Run tests small and read them honestly; scaling a fluke is how testing money gets wasted.
What to cut first when the budget tightens
Budgets shrink, and how you cut reveals whether you understood the budget in the first place. The instinct is to cut whatever is biggest or most expensive. The discipline is to cut by return and role instead.
Cut in roughly this order. First, anything you can't measure — if you can't tell whether spend produced a result, you can't defend it under pressure, so it goes first. Second, proven-negative spend: channels or campaigns where testing already showed weak or falling return. Third, trim the testing reserve before the proven core — painful, because it mortgages future growth, but a business under real pressure keeps the spend that pays now. Protect last the spend with clear, measured, positive return, even if it's large; cutting your best-performing channel to save money is how a budget cut becomes a revenue cut. And before cutting at all, confirm a decline is a genuine return problem and not simply worn-out creative — the fix for that is new creative, not less budget. Deciding what's actually working means measuring real lift, which our marketing analytics guide covers in depth.
FAQ
What percentage of revenue should I spend on advertising?
There's no single correct figure — it depends on your margins, growth stage, and how competitive your category is, so established businesses often sit in the single digits to low double digits of revenue while growth-stage or crowded categories spend more. Treat percentage of revenue as a ceiling you can afford, then check with objective-and-task math whether that ceiling can actually buy your goal.
How do I set an advertising budget with no historical data?
Start with a percentage-of-revenue figure you can afford to lose, then spend it deliberately on a small test rather than a full launch. The first budget's real job is to produce data — your conversion rate and cost per click — so that the next budget can be built by working backward from a goal. Keep the test small enough that learning is cheap.
How should I divide my advertising budget across channels?
Split by funnel stage first (some spend to create demand, some to capture it) and by channel second, weighting toward the channels whose intent, reach, and measurability fit your goal. Avoid spreading a small budget so thin that no channel escapes the learning phase — concentration usually beats dilution when money is tight.
How much of my budget should go to testing?
A common working range is 10–20%, kept as a separate line so the core campaign can't quietly absorb it. The core budget buys known results; the testing reserve buys information about your next winner. Judge them differently — a test that rules a channel out has still earned its cost.
What should I cut first if I have to reduce ad spend?
Cut unmeasurable spend first, then proven-negative campaigns, then trim the testing reserve — and protect your measured, positive-return spend last, even if it's your largest line. Before cutting anything, confirm the decline is a real return problem and not just fatigued creative, since that calls for new creative rather than a smaller budget.
Next step
Pick the sizing method that matches the data you actually have: percentage of revenue if you're starting out, objective-and-task once you know your numbers. Set the figure against a specific goal, split it across funnel stages and channels on purpose, and ring-fence 10–20% for testing before anything else claims it. Then let real results move the money — scaling what earns, cutting what can't be measured. Want a budget built backward from your goals and allocated with reasons? Talk to the team at advertisingagencywebsite.com.
