How Much Should You Spend on Facebook Ads? (Work Out Your First Campaign Budget Before You Burn Cash)
There's no universal 'right' daily budget, there's a right way to think about it. I walk through test budgets, the learning phase, and the percent-of-revenue approach step by step, with illustrative numbers you swap for your own.
Short answer: There's no single number that works for everyone's Facebook ad budget, but there is a logic that works for everyone. Start by setting aside a test budget, think of it as your "tuition," money spent not to generate sales right away but to find out which targeting and creative actually work. As a rule, your daily budget needs to be at least a few times your target cost per conversion, otherwise the system runs out of budget before it ever learns anything. The second approach is percent of revenue: you set aside a fixed percentage of your monthly revenue for ads and hold that number steady. For a business just starting out, beginning with a modest daily number and scaling up gradually as results come in is a lot safer than torching a big budget on day one. Below, I'll walk through that logic step by step with illustrative numbers.
The problem isn't "how much," it's "how much for what"
Be skeptical of anyone who answers "how much should I budget for Facebook ads per month" with a flat number. The right number depends on your product's price, your profit margin, the conversion you're targeting, and your industry. Someone selling a thousand-dollar product and someone selling a hundred-dollar product can't run the same budget logic.
The real question isn't "how much," it's "what am I trying to learn with this budget, and what does that learning cost." The money you spend in week one isn't there to generate sales, it's there so the system can learn who actually buys your product. Try to set a budget without understanding that, and you're firing without a target.
The question on the surface, and the one underneath it
The surface question: "How many dollars should I spend per day?" The real question: "How many attempts does the system need to land one result (a sale, a form fill, a click), and what does that whole batch of attempts cost?"
Answer the surface question with some random number ("fifty bucks a day should do it") and that number might mean nothing at all in your specific industry. In some categories a conversion costs a few dollars, in others it runs a lot higher. Solve the real question instead and the budget falls out naturally: "if one conversion costs X, I need to spend at least a few X for the system to learn," and you build from there.
The learning-phase reality: how Facebook actually spends your money
Meta's ad system doesn't deliver its best results the moment you launch a new ad set. First it shows the ad to a mix of people, learns who reacts, and only then starts showing it more precisely to the people likely to convert. This is called the "learning phase," and the money spent during it isn't there to generate profit directly, it's there to train the system.
Set a budget without accepting that reality, and you'll kill the campaign on day three because "no sales came in," when the system hadn't even finished learning yet. I did this with my own early campaigns, got impatient and shut things down too early, and only later realized the real mistake wasn't a small budget, it was not respecting the learning window.
The generally accepted logic is this: an ad set needs to hit a certain number of conversions per week to graduate out of the learning phase (Meta's own documentation can shift on the exact number, check the platform for the current figure). Set your budget with that number in mind, large enough to realistically reach it in a week, or the system never learns and keeps resetting to zero.
How to work out a test budget
Here's the logic to follow when sizing a test budget:
-
Decide what you consider an acceptable cost for the conversion you're targeting (a sale, a form fill, a message). Say your product sells for $300 and your margin is healthy, so you're fine with a sale costing you $60 in ad spend (this is a purely illustrative example, run the math on your own margin).
-
Set your daily budget to several times that cost. While the system is learning, some of those attempts will be wasted, that's normal. Start with a daily budget below your acceptable cost and the system can't even register a single conversion before the money's gone, which means it learns nothing.
-
Leave it alone for at least a week, longer if you can. Changing the budget or toggling the ad on and off every day resets the learning phase, and you're back to square one.
-
Look at the real numbers at the end of the week, then decide. Keep going, change the creative, change the targeting, or accept that this product just doesn't work on Facebook, the decision gets clear once you have real data.
The percent-of-revenue approach
The second, longer-term approach is this: you set aside a fixed percentage of your monthly revenue for marketing, and that number scales up automatically as the business grows. Small and mid-sized businesses generally keep marketing spend within a certain range of revenue, and that range shifts depending on the industry and growth targets, so quoting one exact percentage would be misleading.
The advantage of this approach: your budget isn't arbitrary, it's tied to what the business can actually support. The downside: it doesn't work for a brand-new business with no revenue yet, because there's no revenue to take a percentage of. In that case, start with the test-budget logic and shift to percent-of-revenue once the first sales start coming in.
The daily minimum: why a budget that's too low never works
Running ads on a daily budget that's absurdly small (say, a few dollars a day) is usually just money down the drain. Here's why: the platform's system needs a certain number of impressions and conversions to learn from, a budget that low never reaches that number, and you end up stuck in an endless learning loop with zero progress to show for it.
If your total budget is genuinely limited, don't stretch it thin over a long period, compress it into a short, focused test instead. Instead of spending a tiny daily amount for a full week, run a daily number big enough to actually teach the system for three days, then stop and evaluate what you learned. Spreading a small budget over a long stretch of time only makes the small budget even more meaningless.
A real example: a made-up ecommerce brand
Say you started a brand selling handmade leather wallets. The product sells for $400, and your margin runs around $150 per unit (these are purely illustrative numbers, calculate your real margin from your actual costs).
You've decided $40 in ad spend per sale is acceptable, because the remaining profit still works for the business. You set aside $150 a day as a test budget, enough for the system to see a handful of conversions and start learning over the course of a week. At the end of week one, you check the numbers: cost per sale came in at $55, a bit above your target but still in an acceptable range. You swap the creative and test again in week two, and this time it drops to $35. Now you start scaling the budget gradually, because you've found a pattern that actually works.
If cost per sale had come in at $200 in week one, blowing straight through your margin, that would've been the signal to question your targeting, your creative, or your pricing, not to reach for a bigger budget. Finding the actual problem comes before scaling spend.
Before you scale the budget: verify
Check three things before you push the budget higher.
Is conversion tracking actually working? If Meta Pixel or the Conversions API isn't set up correctly, the system can't see which click turned into a sale, and no amount of budget fixes that, the system is just flying blind. Before scaling, make a test purchase yourself and confirm tracking records it correctly.
Are your first-week results actually stable? One or two good days can be a fluke. Don't make a big budget decision without at least a week of data, two weeks is even better.
Is the product actually turning a profit? Once you factor in ad spend, are you still profitable? Some sellers get excited that "a sale came in" without noticing that once you subtract the ad cost, they actually lost money on it. Recalculate your real margin before every scaling decision.
Take this with you
- There's no universal right budget number, there's a right way to think about it: find the acceptable cost of one conversion, set your daily budget to several times that.
- Respect the learning phase, don't make snap decisions in the first few days.
- A daily budget that's too low never works, the system runs out of money before it learns anything. If your total budget is small, run a short, intense test instead of stretching it thin.
- Verify tracking is accurate and your real margin holds up before you scale the budget.
- I'm not going to sell you "spend this exact number, sales are guaranteed," most first test budgets get spent without a clean win, that loss is the cost of learning.
- Scale gradually as results come in, don't jump straight to a big budget.
Frequently asked questions
How much should someone running their first ad campaign start with?
There's no fixed number, but the logic is this: figure out what cost per conversion is acceptable for your own product, then set your daily budget to several times that cost so the system can see at least a handful of conversions during the learning phase. Starting with a number that's too low means the system never gets to learn anything, and the budget just evaporates.
How long does the learning phase last?
Usually around the first week, but the exact length depends on how many weekly conversions the ad set gets, so it varies a lot. Stopping and restarting the ad, or changing the budget, during this window resets the learning phase. The safest approach is to leave it alone and do a full review at the end of the week.
Is changing the budget every day harmful?
Yes, frequent, large budget swings (especially increases or decreases over twenty percent) can reset the learning phase. If you want to scale the budget, do it in small, gradual steps and give each increase a few days to settle before making another one.
If I don't have revenue yet (I'm just starting out), how do I set a budget?
The percent-of-revenue approach doesn't work yet because there's no revenue to take a percentage of. Start with the test-budget logic instead: take your total available budget, split it into a daily number that's still large enough for the system to learn from, and run a short, focused test window. Once real sales start coming in, you can shift to revenue-based budgeting.
What should I do if my ad cost keeps coming in above my target?
Don't reach for a bigger budget first, find the actual problem. The creative might not be grabbing attention, the targeting might be reaching the wrong audience, or the product page might not be converting visitors who land on it. Raising the budget without fixing the underlying issue just burns cash faster. Diagnose the problem on a small budget first, then scale.
