How much should you spend on Facebook ads?

Every 'spend $10/day' rule you've read ignores the only thing that matters: what a customer is worth to you. A $10/day budget is generous for a business selling a $2,000 service and hopelessly thin for one selling a $20 product on volume. The right budget is derived from your economics, not copied from a blog.

So instead of asking 'how much should I spend?', ask three questions: what can I afford to pay for a customer, how many customers do I want this month, and what's the minimum budget the platform needs to actually work? Answer those and the number falls out.

Start with your average order value or deal size and your gross margin. If you sell a service for a set fee, your margin is what's left after delivery costs. If customers buy repeatedly, use lifetime value (LTV) — the total gross profit from an average customer over their lifetime — not just the first purchase.

This number is your ceiling. You can afford to spend up to your margin (or a healthy fraction of LTV) to acquire a customer and still be in the game. A business that only counts the first sale will always be out-bid by a competitor who understands their repeat value.

Your maximum CPA is the most you can pay for a customer and still hit your profit goal. A simple version: gross profit per customer minus the profit you want to keep equals what you can spend to acquire them. That's your break-even-to-target CPA range.

From CPA you can derive a break-even ROAS (return on ad spend): roughly your selling price divided by your gross profit per sale. Knowing both numbers turns ad reporting from 'is this good?' into 'is this above or below my line?'.

Meta's delivery system needs a minimum volume of conversion events per ad set per week to exit the 'learning phase' and stabilise. If your budget is too small to generate that volume, the campaign never learns, costs stay high and results look random — and you wrongly conclude 'ads don't work for me'.

Practically, your daily budget per ad set should be at least a small multiple of your target CPA, so you can collect several conversions a week. If your target CPA is high and your budget can't support even a few conversions weekly, optimise for a cheaper upstream event (a lead instead of a purchase) until you have the volume — or concentrate budget into fewer ad sets rather than spreading it thin.

Begin with a budget you can afford to treat as a test — enough to fund the learning phase for one or two focused campaigns, no more. The goal of the first few weeks isn't profit; it's data: which audience, creative and offer actually convert.

Once a campaign is reliably beating your target CPA, scale it — but gently. Large sudden budget jumps reset the learning phase and can spike costs. Increasing a winning ad set's budget in modest steps every few days, and duplicating winners into fresh audiences, tends to hold efficiency better than doubling overnight.

A common, sane split for a growing account: put most of your budget behind proven winners (scaling), reserve a slice for testing new creative and audiences, and keep a small always-on retargeting campaign for warm traffic. Retargeting is usually your cheapest conversions — but it's limited by how much cold traffic you're bringing in, so don't starve the top of the funnel to feed it.

Review weekly. Move money toward what's beating your CPA line and away from what isn't. Budgets aren't set-and-forget; they're steering.

• •Spreading a small budget across many ad sets, so none get enough data to learn.
• •Judging campaigns on clicks or CPL instead of cost per sale and ROAS.
• •Scaling too fast and resetting the learning phase, then blaming the platform.
• •Ignoring repeat purchase value, so you under-invest versus competitors who don't.
• •Killing a campaign before it's had enough conversions to prove anything.