Most "daily budget" guides hand you a flat number — spend fifty dollars, spend a hundred — and stop there. That number is useless without two things: the cost of the result you're buying, and the margin on the order you're trying to win.
This page gives you the formula, the platform rules behind it, and a worked example on real operating numbers. If you already run Meta ads with months of spend history, this is how to set a budget that both delivers and stays profitable.
How much should you spend on Meta ads per day?
The platform floor is tiny. You can run an ad for as little as a dollar a day on impression-based campaigns, according to Meta's documented minimums reported by Vizup — but that's a technical minimum, not a working budget.
The working minimum is tied to what each conversion costs you. For any campaign using a cost-per-result goal, Meta's guidance is that your daily budget should be at least five times your target cost per result. If you'd happily pay ten dollars per purchase, fifty dollars a day per ad set is your floor.
Everything above that floor is a profit decision, not a delivery decision. The platform will spend whatever you give it — your job is to stop at the point where the next dollar stops paying for itself. That line lives in your margin, and we'll get to the math.
The real floor: your budget has to clear the learning phase
Every new or significantly edited ad set enters a learning phase while Meta figures out who to show it to. Delivery is less stable and cost per result is more volatile until it exits.
An ad set exits learning after it gathers about 50 optimization events in a roughly seven-day window, per Meta's Business Help Center as cited by Vizup. If you optimize for purchases and can't generate ~50 purchases a week at your current budget, the ad set can get stuck in "Learning Limited" — unstable and usually more expensive.
That penalty is measurable. Meta's own analysis found ad sets under the fifty-event threshold ran about 28% more expensive per purchase — a directional first-party figure, not a guarantee for your account, but it shows the cost of underfunding an ad set is real.
The 5× cost-per-result rule, in plain math
The five-times rule is just the fifty-event math reversed into a daily number. Here's the arithmetic.
If your cost per purchase is $20, fifty purchases cost $1,000, and $1,000 ÷ 7 days = roughly $143 per day per ad set. If your cost per purchase is $40, that's $2,000 a week, or about $286 a day. Vizup walks the same formula — (target CPA × 50) ÷ 7 — and lands near $214 a day at a $30 cost per purchase.
The uncomfortable takeaway: if you set a $50 daily budget on an ad set whose purchases cost $40, the math can't produce fifty events in a week, so the ad set may never stabilize. You're asking the system to learn without funding the lesson.
What happens when you have too many ad sets
Fifty events is per ad set, not per campaign. Split one budget across five ad sets and each one needs its own fifty events — so fragmenting your account multiplies the learning tax and starves every ad set of data.
Consolidating ad sets is often the cheapest fix for a store stuck in learning. Fewer ad sets means more events each, which is the whole logic behind Meta steering accounts toward broad, consolidated campaigns. For how long this phase actually lasts and how to read it, see our breakdown of the Facebook ads learning phase and how long it takes.
How Meta actually spends your daily budget
A daily budget is an average, not a hard cap. Meta treats it as a target across the week and shifts spend to the days with the best opportunities.
On a strong day, Meta may spend up to 75% above your daily budget, while total weekly spend won't exceed your daily budget times seven. So a $100 daily budget can hit $175 on a Tuesday and pull back on a slow Sunday, but the week still lands near $700.
This matters for how you read results. Judging an ad set by a single day's spend and cost per result is noise — you're looking at one slice of a weekly average. Evaluate on a seven-day rolling window instead.
Daily budget vs. lifetime budget
A daily budget keeps spend roughly level every day and runs until you pause it — the right default for always-on prospecting and for ad sets you're trying to push through the learning phase on a steady cadence.
A lifetime budget spreads a fixed total across a start and end date and lets Meta front-load or taper within that window. It fits time-boxed promotions — a holiday sale, a product launch — where you care about the total, not the daily pace.
For an operating store running evergreen acquisition, daily budgets are usually the simpler read. You'll change the number as you scale; you won't change the mechanism.
The number the recommendations skip: break-even ROAS
Here's what the generic guides never tell you: a daily budget that delivers can still lose money. Delivery and profit are two different gates, and the second one is pure arithmetic.
Break-even ROAS — the return on ad spend where revenue exactly covers the cost of goods plus the ad spend — is simply 1 ÷ your contribution margin. Contribution margin is the share of revenue left after product cost, shipping, and transaction fees, before ad spend.
A 50% margin means you break even at 2.0x ROAS (1 ÷ 0.50). A 40% margin needs 2.5x. A thin 30% margin needs 3.33x before you've made a cent — which is why paid acquisition gets hard fast on low-margin products. Your "right" budget is the one that keeps your marginal ROAS above that break-even line, not just your average.
A worked example on real numbers
Say you run a store doing 340 orders a month at a $31 average order value, spending $2,800 a month on Meta — about $93 a day. Your product cost, shipping, and fees eat 55% of revenue, so your contribution margin is 45% and your break-even ROAS is 1 ÷ 0.45 = about 2.22x.
Your cost per purchase is $2,800 ÷ 340 = roughly $8.24. Five times that is about $41 a day per ad set as a delivery floor, so your $93 budget comfortably clears the learning phase on a single consolidated ad set — good.
Now you scale. You push the daily budget from $93 to $150 and revenue climbs, but the last $57 of daily spend only brought in $90 of new revenue. That increment ran at 90 ÷ 57 = 1.58x — below your 2.22x break-even. Your headline ROAS still looks green while your last dollars are losing money. That's the trap the daily-budget guides never mention, and it's covered in depth in our guide to profitable ad scaling.
How to raise your budget without resetting learning
Large budget jumps count as a significant edit and can throw an ad set back into learning. The practitioner convention is to raise a daily budget by no more than about 20% every couple of days — a safe default, not a Meta law, that keeps edits small enough to avoid a reset while delivery re-stabilizes.
The cadence isn't the real ceiling, though — your marginal ROAS is. You can follow the 20% rule perfectly and still scale into losses because each increment buys a less-responsive slice of audience. Watch the return on the new spend, not just whether you obeyed the step size.
If you want the official thresholds rather than folklore, start with the Meta ads minimum budget rules from the help center. When budgets hold but results slide, the cause is often creative fatigue — our rundown of the leading software for ad fatigue detection covers the warning signs.
Where Victor fits
Victor is the AI employee inside PodVector AI. He connects to your Meta Ads account and your Shopify store, computes the true per-order profit on every sale — product cost from Printify, Printful, or Gelato, plus shipping and fees — and watches each ad set against your real break-even ROAS, not a vanity number.
When an ad set's marginal return falls below break-even, Victor pauses it. When a winner has headroom, he raises the budget in steps that respect the learning phase. He changes budgets and bids in Meta Ads directly, so the recommendation and the action are the same motion.
That's the difference between a daily-budget blog and an operator who runs on your numbers. Start with PodVector AI and let Victor set budgets off your real margins.
FAQs
What is the minimum daily budget for Meta ads?
The technical floor is about a dollar a day on impression campaigns, per Meta's documented minimums, but that won't deliver meaningfully. The working minimum is roughly five times your target cost per result per ad set, which is what Meta needs to gather enough data. Below that, ad sets tend to stall in the learning phase.
How much should I spend per day on Meta ads for my store?
Enough per ad set to generate about fifty optimization events a week — take your cost per purchase, multiply by fifty, and divide by seven. Above that delivery floor, the ceiling is wherever your marginal ROAS crosses your break-even ROAS. Those two numbers set your range; a copied dollar figure doesn't.
Why is my ad set stuck in "Learning Limited"?
Usually the budget is too low for the fifty-events-per-week threshold, or you've split spend across too many ad sets so none gather enough data. Meta's own analysis showed sub-threshold ad sets ran about 28% more expensive per purchase. Raise the budget, consolidate ad sets, or verify your pixel is actually reporting the conversions that happened.
Does increasing my daily budget restart the learning phase?
Large increases can, because they count as a significant edit. Small ones generally don't. The common safe default is raising by no more than about 20% every couple of days, which keeps changes below the reset threshold while delivery re-stabilizes.
Will Meta spend exactly my daily budget each day?
No. Meta averages your daily budget across the week and can spend up to 75% over on a high-opportunity day, capped at seven times the daily amount over the week. Read performance on a seven-day window, not day by day.
Does a good ROAS mean I should spend more?
Not on its own. Average ROAS says nothing about whether the next dollar is profitable — a strong average can hide a marginal return below break-even. Scale on the return of your newest spend, and stop when that marginal number drops under 1 ÷ your contribution margin.