On Meta's platforms, the creative is the targeting. The algorithm decides who sees your ad largely based on how people respond to it, so the fastest way to improve results is to give it better creative to work with. The challenge on a small budget is learning quickly without burning cash. This is the framework we use.
Start with angles, not ads
An angle is the reason someone would care. For a meal-kit offer, angles might be time saved on weeknights, less food waste, trying new cuisines or cheaper than takeaway. Before designing anything, we write five to eight angles and pick the three we believe in most. Testing angles first tells you what message resonates. Testing button colours first tells you almost nothing.
A simple account structure
- One campaign per offer, optimised for the conversion event that matters (a lead or purchase, not a page view).
- Broad targeting within the offer's allowed countries and age range. Let creative do the work.
- Three to five ads per ad set, each representing a different angle or format.
- A modest daily budget you can sustain for at least a week. Constant edits reset learning.
Formats worth testing
| Format | Why it works |
|---|---|
| Short vertical video (6–15s) | Native to Reels and Stories; strong hook in the first second |
| Static image with a clear claim | Cheap to produce, quick to iterate |
| Carousel comparison | Great for “which option suits you?” stories |
| User-style footage | Feels personal; must be genuine and properly licensed |
How to read the results
Early metrics are signals, not verdicts. We look at them in order:
- Thumb-stop rate (3-second views ÷ impressions) tells you if the hook works.
- Click-through rate tells you if the message earns interest.
- Landing page conversion rate tells you if the page delivers on the promise.
- Cost per result vs payout tells you whether to keep going.
If the hook fails, change the first second. If clicks are high but conversions low, the problem is usually the landing page or a mismatch between ad and page, not the audience.
Set kill rules before you launch
Decide when you will stop before you start. Emotion is expensive.
A common rule of thumb: if an ad has spent about twice the payout without a single conversion, pause it. If it has converted but at a cost well above payout after a fair sample, iterate on the winner's angle or move on. Write the rule down and stick to it.
Stay inside the lines
Meta's advertising standards are strict, especially around health, personal attributes and before-and-after claims. Ads must not imply that the viewer has a condition (“Struggling with your weight?”) and must not make claims the advertiser cannot support. We review every ad against the platform rules and the advertiser's own approved claims before it goes live. A rejected ad costs a day. A disabled account costs far more.
Scaling what works
When an ad set is profitable for several consecutive days, increase budget gradually, roughly 20% at a time, so the delivery system can adjust. Meanwhile, keep a steady trickle of new creative in testing. Every winner fades eventually. A pipeline of fresh angles is the only sustainable edge.