How much should you spend on Meta ads?

Start from your margin, not your ambition

If your average order value is ₦40,000 and your gross margin is 45%, you can afford roughly ₦18,000 per sale before you're working for free. That ceiling is your reality check: it tells you whether your offer, not your budget, is the bottleneck.

Budget for learning, then for scaling

Give the algorithm enough spend to exit the learning phase — typically around 50 conversions per ad set per week. If your cost per purchase is ₦5,000, that's roughly ₦250,000 weekly per ad set, which is why most small brands should run one or two ad sets rather than ten.

• Testing: 20–30% of budget on new creative angles.
• Scaling: 60–70% behind proven winners.
• Retargeting: 10–20% on warm audiences.

The three-week read rule

Don't judge a campaign before three weeks or roughly 50 conversions, whichever comes first. Before that you're measuring noise. After that, you're measuring creative, audience and offer — in that order of impact.

Scale in steps, not leaps

Increase budgets by 20–30% every few days rather than doubling overnight. Sharp jumps reset the learning phase and spike your cost per result. If you need volume faster, add a new ad set on a different audience or placement instead.

When to spend less

If your cost per result is above your margin ceiling after three weeks of honest testing, fix the offer or the landing page before adding budget. More money into a broken funnel just buys a faster loss.