/Advertising
Question
As a beginner, should I test more creatives or more audiences first?
Posted by •11/20/2025
I have a limited ad budget and I don't know where to put my testing focus.
Some people say:
- “Creatives are 80% of results, test as many as possible”
Others say:
- “Audience targeting is everything, test a bunch of interests/lookalikes”
My situation:
- Budget: about $20–$30/day
- Niche: pet grooming
- I have:
- 3 video creatives
- 2 image creatives
- A bunch of possible interests to target
Questions:
1. For a small budget, is it smarter to:
- Keep audience broad and test many creatives?
- Or keep a few creatives and test many audiences?
2. What does your **simple** testing structure look like when you launch a new product?
Some people say:
- “Creatives are 80% of results, test as many as possible”
Others say:
- “Audience targeting is everything, test a bunch of interests/lookalikes”
My situation:
- Budget: about $20–$30/day
- Niche: pet grooming
- I have:
- 3 video creatives
- 2 image creatives
- A bunch of possible interests to target
Questions:
1. For a small budget, is it smarter to:
- Keep audience broad and test many creatives?
- Or keep a few creatives and test many audiences?
2. What does your **simple** testing structure look like when you launch a new product?
4 Replies
•11/20/2025
Makes sense. I’ll keep audiences simple and put my energy into testing different hooks and creatives instead of drowning in a list of 30 interests. Thanks for the clear explanation.
•11/20/2025
I wasted a lot of money testing 15 different interests with the same boring ad.
When I finally made 3 completely different hooks (problems, angles, opening lines), I got my first profitable day with just BROAD + one pet interest.
When I finally made 3 completely different hooks (problems, angles, opening lines), I got my first profitable day with just BROAD + one pet interest.
•11/20/2025
A simple rule I use:
- If CTR < 1% almost everywhere → creative problem
- If CTR is good but ROAS is bad → landing page / offer problem
- If CTR & CVR are good but scaling breaks things → then play with audiences and structure
Most beginners skip straight to “advanced audience strategies” when they’ve never truly tested 3–5 different hooks in their creatives.
- If CTR < 1% almost everywhere → creative problem
- If CTR is good but ROAS is bad → landing page / offer problem
- If CTR & CVR are good but scaling breaks things → then play with audiences and structure
Most beginners skip straight to “advanced audience strategies” when they’ve never truly tested 3–5 different hooks in their creatives.
•11/20/2025
Short answer: **test creatives first**.
Why:
- Meta’s algorithm is very good at finding people who might buy **if your ad is good**.
- A bad creative in 20 audiences is still a bad creative.
- A strong creative in 1–2 broad audiences often beats “interest stacking” chaos.
With your budget, I’d do:
- 3–4 ad sets:
- 2 broad (no interests, just country + age + gender)
- 2 simple interests (e.g. “Dogs”, “Dog grooming”)
- Inside each ad set:
- Same 3–4 creatives
Let Meta tell you:
- Which creative gets best CTR / cheapest clicks
- Then you can kill losers and duplicate winners into new setups later.
Once you have a clearly winning creative, you can:
- Try new audiences (countries, age brackets, interests)
- But don’t get lost in 20 tiny ad sets with weak ads.
Why:
- Meta’s algorithm is very good at finding people who might buy **if your ad is good**.
- A bad creative in 20 audiences is still a bad creative.
- A strong creative in 1–2 broad audiences often beats “interest stacking” chaos.
With your budget, I’d do:
- 3–4 ad sets:
- 2 broad (no interests, just country + age + gender)
- 2 simple interests (e.g. “Dogs”, “Dog grooming”)
- Inside each ad set:
- Same 3–4 creatives
Let Meta tell you:
- Which creative gets best CTR / cheapest clicks
- Then you can kill losers and duplicate winners into new setups later.
Once you have a clearly winning creative, you can:
- Try new audiences (countries, age brackets, interests)
- But don’t get lost in 20 tiny ad sets with weak ads.