What variables do YOU test for on Meta Ads?

When you run ads for a company that does NOT have a custom audience or a clear data file, you need to redefine some priorities to test the right variables and optimize the fastest.
 
When running ads for a company without a large/accurate custom audience what variable would you firstly test via A/B testing or two concurrent campaigns?


All responses appreciated!
I will take data from both variables and compare them, my friend.
 
Test running ads at different times or in different geographic areas, over a period long enough to collect enough data and account for fluctuations, usually at least a week
 
Without a standard customer file, start by A/B testing ad elements (image, title, text, CTA) as they most impact campaign effectiveness.
 
Test the creative (image/video vs. copy) first. It usually impacts performance the most. Also try broad vs. interest-based targeting to see what FB’s algorithm favors. Lmk if it works
 
When custom audiences are limited, start by A/B testing:
  1. Creative (image vs. video)
  2. Ad copy (short vs. long)
  3. Audience interests or lookalikes
  4. Placements (auto vs. manual)
  5. Headline variations
Start with 1–2 variables at a time to keep results clear and actionable.
 
Start testing with these on Meta Ads:-
  1. Audience: broad vs. interest-based
  2. Creatives: different images/videos
  3. Copy: short vs. long, benefit vs. emotional
  4. CTA buttons: “Learn More” vs. “Shop Now”
  5. Placements: automatic vs. manual
  6. Budget: low vs. medium spend
Test, analyze, then focus on winners.
 
When running test ads, I often pay attention to factors such as which content attracts viewers, which time has more interactions, whether it attracts the right audience, and whether it differentiates between different time periods to adjust and allocate budget.
 
When running an advertising campaign, I always pay attention to the CTR factor in running ads to adjust other factors and be able to promptly edit the campaign when necessary.
 
in my opinion i will test different campaigns, and i will focus on ad copy and ad placement.
 
it makes a big difference even with the same subject. I leave the subject until last when I don't have a clear baseline. a long enough period of time, usually at least a week.
 
Totally agree with you! Testing creatives is definitely the best starting point, especially when you don’t have a precise audience. And MGID is a great tool for such cases — native ads really deliver solid results without complicated targeting. Great advice for anyone who wants to launch quickly and see the first results!
Absolutely agree! MGID really works well when you need to quickly test hypotheses without complex analytics or deep audience targeting. If you have strong creatives, the platform delivers stable traffic and a good cost per click
 
To can optimize the campaign i run them for a week and half and i change the ads or run multiple in paralel to see what ad perform the best then i just removed the rest and i will leave just the best performing 2 ads
 
When running ads without a large custom audience, you should prioritize Audience testing first. A/B testing different audience groups will help you discover who is interested in your product/service. From there, you can optimize other elements like ad copy and images.
 
When the audience size is still small or not yet refined, I usually start by testing creatives (images/videos + headlines) first, followed by audiences. Even small differences in creatives can show performance gaps early on. From there, I move to placements and bidding strategies. A/B testing is useful, but keep budgets tight to avoid wasting spend.
 
Start by testing creatives (image/video vs. copy) and audiences (interests, lookalikes, broad). These two factors often have the biggest impact when no strong custom data is available.
 
Back
Top