How much should a new Shopify store spend on Facebook ads?
A budget that is too small never leaves the learning phase and teaches you nothing. Here is how to work out a starting number from your own margin and price.
The honest answer is that there is no universal number — but there is a way to calculate yours, and it is not guesswork.
The trap most new stores fall into is spending too little. A $5/day budget feels prudent. It is actually the most expensive option available to you, because it buys you nothing you can learn from.
Why tiny budgets waste money
Meta’s delivery system needs roughly 50 optimisation events per ad set per week to leave the learning phase and deliver predictably. If your event is a purchase, that is 50 purchases a week.
At a $30 cost per acquisition, 50 purchases cost $1,500 — about $215 a day for a single ad set. Most new stores cannot start there, and they should not try.
What matters is understanding what happens when you do not reach that threshold: your ad set stays in learning, delivery stays erratic, your costs stay higher than they would otherwise be, and — worst of all — the data you collect cannot tell you whether the campaign works. You spend $150 over a month, get four sales, and still have no idea whether the product, the creative or the audience was the problem.
Calculate your floor from your own numbers
Work backwards from the cheapest signal you can afford to optimise for.
Step 1 — estimate your CPA. If you have no data, a usable starting assumption for a typical consumer product is somewhere between $20 and $50 per purchase. Take the pessimistic end.
Step 2 — pick a cheaper optimisation event. You almost certainly cannot afford 50 purchases a week at the start. You can often afford 50 add-to-carts, which typically cost roughly a fifth to a tenth of a purchase. Optimising for add-to-cart gets you out of the learning phase at a budget you can actually sustain, and you switch to purchase optimisation once volume supports it.
Step 3 — do the arithmetic.
weekly budget = 50 × cost per chosen event
daily budget = weekly budget / 7At a $4 add-to-cart cost: 50 × $4 = $200 a week, about $29 a day. That is a realistic floor for one ad set that will actually learn something.
Step 4 — check it against your margin. If $29 a day is more than you can lose for three weeks straight, your budget is not the problem — your funnel is not ready for paid traffic yet. That is worth knowing before you spend anything.
Structure matters more than size
Given a fixed budget, how you split it decides how much you learn.
Two ad sets at $15/day each will both stay in learning and neither will produce a usable answer. One ad set at $30/day will. Concentration beats spread when your budget is small — you are buying a decision, not coverage.
A workable starting structure for a store with under $1,000 a month to spend:
- One campaign, conversion objective.
- One or two ad sets — a broad interest cluster and, if you have any customer data, a lookalike.
- Three to five ads per ad set, different creative angles rather than different crops of the same image.
Add ad sets when you can afford to keep each one above its learning threshold, not before.
Give it a proper testing window
Plan for three weeks minimum before you judge anything.
The first week is learning: the numbers are not predictive and you should mostly leave things alone. The second week gives you your first real read. The third tells you whether week two was a pattern or an accident.
Budget the whole window up front and treat it as the cost of finding out — because that is exactly what it is. Stopping a campaign after five days because the ROAS looks bad is the single most expensive habit in small-budget advertising: you pay the full price of the learning phase and then throw away the result.
What to spend once it works
Once you have a campaign clearing your break-even ROAS reliably, the budget question changes shape. It stops being “what can I afford to risk” and becomes “how much can I buy at this price before the price rises”.
That is a much better problem, and it is answered by scaling gradually — around 20% every three or four days — and watching for the point where CPA starts climbing. That point is your current ceiling, and it moves as you add creative and audiences.
AdPilot’s simulator projects impressions, reach, CTR, conversions and expected ROAS for a given budget, market and duration — so you can see roughly what a $20/day and a $50/day campaign each buy you before committing either.