If a Meta campaign has ever refused to launch for no apparent reason, this is very likely why. Meta enforces a minimum daily budget, it applies per ad set rather than per campaign, and it is roughly three times higher for conversion objectives than for impression ones. The error message does not mention any of that — it simply fails.
The rule
Per ad set, not per campaign
Three ad sets means three minimums. This is the part that catches people building layered audience tests.
Tiered by optimisation goal
Impression-based objectives (reach, brand awareness) sit at a base rate. Conversion-based objectives (purchase, lead, content view, add to cart) sit at roughly three times that base.
Denominated in the account currency
Not converted from dollars. Each currency has its own floor, and the three-times relationship holds within it.
A concrete worked example, from an AED ad account: the impression floor is about AED 3.69 a day, and the conversion floor about AED 11.07 a day. Three times, per ad set.
What it means in practice
The arithmetic compounds faster than people expect.
- 1One conversion ad set: one conversion floor per day. Roughly AED 330 a month on that account.
- 2Two audience segments to compare: two floors. Above AED 22 a day, roughly AED 660 a month, before you have bought a single result.
- 3Four product categories in four ad sets: four floors. Above AED 44 a day, roughly AED 1,350 a month.
💡 So "test it with a small budget and see" has a hard limit, and hitting it is not a sign you did something wrong. We have had a launch refused outright for being set at AED 10 a day on a conversion objective — nothing was created, no money moved, and the error did not mention the three-times rule.
Why the conversion tier is higher
Not arbitrary. A conversion objective asks Meta to find people likely to take a specific action, which requires enough events to learn from. Below a certain daily spend the ad set never accumulates enough conversions to exit the learning phase, so it delivers erratically and expensively.
Meta would rather refuse the campaign than take your money for something that structurally cannot work. That is defensible — the failure is that it does not explain itself.
The trap just above the floor
Worth knowing because it is subtler than outright refusal. A budget only slightly above the minimum is accepted and then delivers poorly — throttled, sporadic, expensive per result. It looks like a targeting or creative problem and it is neither.
A rough working rule: if you are within about 20% of the floor, you are under-funded rather than lean, and you will get a better result from one properly funded ad set than two barely funded ones.
What to do at a genuinely small budget
Run one ad set
Properly funded, well above the floor. Resist the urge to split. Two starved ad sets teach you nothing and deliver worse than one.
Consider a cheaper objective
A traffic objective sits in the lower tier. It is not as good as a conversion objective — but a funded traffic campaign beats a starved conversion one.
Look at search instead
Google Search is far more forgiving of small budgets, because intent does the work. A handful of clicks from people actively searching can produce a customer this week.
Wait and save
Unglamorous, and sometimes correct. Two months of budget spent properly beats four months spent below the threshold.
What a tool should do about this
The minimum is retrievable from the API for a given ad account — so a tool can and should check it before launching, rather than letting you type a number and fail in front of you.
Ask any vendor: does it validate my budget against the platform minimum for my account and objective before launch? A tool that accepts any number and reports a failure afterwards has left you to discover an undocumented rule by trial and error.
💡 Meta also accepts a validate-only flag on ad set creation, which checks a payload against the live API and returns real errors without creating anything or spending anything. A tool using it catches budget violations, objective conflicts and creative rule breaches before your campaign does.
One related thing, since it causes the same confusion
Once a campaign does launch, expect four to five hours before a new click or conversion ad set serves its first impression. Reach objectives start within minutes. Comparing the two is the most common reason people conclude a properly funded campaign is broken when it is simply waiting.
Budgets checked against the real minimum before launch
Adyft validates your budget against the platform floor for your account and objective, so campaigns do not fail for reasons nobody explains. 14-day free trial, no card.
Start free — no card needed