If you're testing different creatives or products with different custom audiences, then yes, it’s a good idea to separate them into different ad sets or even campaigns, depending on how distinct the audiences are and how clean you want your data to be.
1) If all ad sets are using the same custom audience (e.g., “site visitors last 30 days”):
It’s fine to keep them in the same campaign, but use audience exclusions to prevent internal competition.
Example: Ad Set 1 = Product A → Exclude Product B/C audiences
Ad Set 2 = Product B → Exclude Product A/C, and so on.
2) If each ad set has a different custom audience (like purchasers of specific products, or page engagers for each service):
Then you can either:
Keep them as separate ad sets in the same campaign, OR
Split into separate campaigns for better budget control and easier performance tracking.
3) For A/B testing ad creatives:
Run the A/B test within one ad set, or use Meta’s built-in A/B test tool.
Avoid testing the same creative in multiple ad sets with the same audience, as they’ll compete and inflate costs.
Use Meta’s Breakdown + Custom Reporting to track performance per audience, ad creative, and placement—this will help you decide whether to keep testing in one campaign or break things out further.