The Facebook Ads Budget Calculator computes the minimum budget a Meta campaign needs to work as Meta's delivery system intends: it starts from your target CPA and the conversion volume required to exit the learning phase (50 conversions per ad set per week by default), multiplies out to a weekly and monthly budget, and then checks that budget against profitability, comparing your target CPA with the maximum CPA your revenue and margin can support.
How It Works
Entering a CPM also yields the impressions that budget buys. The verdict has three states: a target CPA at or below 80% of the maximum is comfortably profitable; between 80% and the maximum it is profitable but tight; above the maximum the plan loses money on every conversion even if it performs exactly to target.
Parameters
| Input | Required | Meaning | |---|---|---| | Target CPA | Yes | What you aim to pay per conversion | | Ad sets | No | Number of ad sets (default 1) | | Conversions per week per set | No | Default 50, Meta's learning benchmark | | Revenue per conversion | No | Unlocks the profitability check | | Gross margin % | No | With revenue, sets the maximum profitable CPA | | CPM | No | Unlocks the impression estimate |
Worked example
Target CPA 2,000 a week, the account 17,200**, buying a modeled 430 conversions a month. Add revenue per conversion of 60: the 20 of headroom per sale**, and the verdict is comfortably profitable. At a 58 and the verdict flips to a warning despite nothing else changing: at 97% of the maximum, normal week-to-week CPA variance will push individual weeks underwater, which is exactly the planning insight the 80% comfort band encodes.
Why the budget floor is about learning, not spend
Meta's optimization needs roughly 50 conversion events per ad set per week to exit the learning phase and stabilize delivery; below that, the system explores erratically and CPAs swing for reasons unrelated to your creative. That requirement converts directly into a budget floor through your CPA, which is why the calculator is built around it, and why its first planning lesson is usually subtractive: if the floor for three ad sets exceeds your budget, the fix is fewer ad sets (consolidate audiences and creative variants into one set that can reach 50 conversions), not starving all three. It also explains a common beginner paradox: doubling the budget of an under-funded campaign can lower its CPA, because the extra spend buys the algorithm out of learning rather than just buying more impressions.
Setting the inputs honestly
The target CPA should come from your economics, not your hopes: the maximum profitable CPA the calculator derives (revenue x margin) is the ceiling, a working target sits at 70 to 80% of it to leave room for variance, returns, and the platform fees the margin may not include. Conversions per week can be lowered for high-ticket offers where 50 weekly conversions is an unrealistic volume for anyone; the calculator accepts it, but understand the trade you are making explicit: fewer optimization events means slower, noisier learning, and lead-quality signals (CRM-fed offline conversions) matter more, not less. The 4.3 multiplier is the average weeks per month, so the monthly figure is a planning average, not a calendar-month promise.
Limitations
A planning model on your assumptions: it does not forecast your actual CPA (auction competition, seasonality, creative quality, and audience saturation move it), does not include creative production, agency, software, or tax costs, and treats the learning requirement as a fixed 50 where Meta's actual thresholds vary by optimization event and have changed over time. Profitability verdicts are only as honest as the margin entered; use contribution margin after fulfilment and returns for a verdict you can bank on.
Frequently Asked Questions
How much should a small business spend on Facebook ads to start?
Enough for one ad set to reach the learning volume at your realistic CPA, or an explicit test budget you accept may not exit learning. The calculator turns your CPA into that floor: at a 1,250 a week; below that, consolidate everything into a single ad set and judge results over longer windows.
Why does the calculator multiply by 4.3 for the monthly budget?
Because a month averages 4.3 weeks (52 weeks over 12 months), and Meta's learning requirement is weekly. Budgeting a flat 4 weeks understates the monthly need by about 7% and recreates the under-funding the tool exists to prevent.
My target CPA is above my maximum profitable CPA. What actually fixes that?
Only economics, not settings: raise price or average order value, improve margin (cheaper fulfilment, fewer returns), increase conversion value through bundles, or count verified repeat-purchase value in the revenue figure. The calculator's warning is the plan telling you the business model, not the media plan, is the constraint.