Chapter 04 / 6Budget and timeline expectations
Budget for learning first, then for scale. We don't quote a single starting budget, because the right figure depends on your product's price and what a customer costs you.
Budget for learning first, then for scale. We don't quote a single starting budget, because the right figure depends on your product's price and what a customer costs you.
Size the test from your cost per customer
A test only tells you something if each ad gets enough spend to show a result. A common rule of thumb is to let every ad spend a few times your target cost per acquisition before you judge it. That's a rule of thumb, and it isn't a benchmark. Work out your target cost per customer first, then multiply it by the number of ads you want to compare. The result is your test budget. Our ad budget calculator and break-even ROAS calculator do the arithmetic.
Test few angles properly

Spread a small budget over many ads and all you get is noise. Test one angle per cut and three to five openers on the same body, then judge on retention and cost per result. Our A/B testing guide explains how to keep the test clean.
Plan for a few weeks, not days
Delivery systems need volume and time to settle, and a creative that looks weak early can recover once they do. Plan for a few weeks of learning before you decide an approach doesn't work. Keep each platform's results separate, because Meta and Google optimize on different signals.
Keep reading: The creative challenge, The formats and specs that matter, all chapters of Reels and Shorts Ads Playbook: Acquire Customers With Short-Form Video, and the ad creative guide.