#general-chat
Thread
Hello, just started advertising my first brand on Meta, pet space.
On the second day, we had fantastic results. 15% conversion, 10 purchases, 5% CTR, and 4.5 ROAS despite high CPMS. The campaign ended because I stupidly set the end date incorrectly.
When I restarted the ads, the performance tanked across the board, lower CPMS but far lower CTR% and 0% conversion for two days straight now.
I'm only spending 50-60 dollars a day. Should I just let this run? New to the Meta game.
Any thoughts here?
in short - yes you'll need to let it run. Set a budget you're happy to spend daily for the next 14-30 days. one campaign, one ad set, and drop a new ad in every 3-4 days.
What is your ad volume on that budget?
@Fergal Dinan is right, you'll have to let it run so it can find the riight audience and optimise over time.
Focus on ROI, ignore CPMs (for the most part, it's just a distracting metric). Higher CPMs can mean higher-value audience (hence the higher price), which is why your CVR % is higher.
Focus on testing new ads/messaging angles consistently, and refreshing them. This way, you give the algorithm lots of options (ads, angles) to optimise for sales.
Good luck!
@Fergal Dinan I have like 3 ads. The original ad that worked gets the majority of the spend and the new ads make performance worse.
@Stefan Bozic thanks!
Do you test the new angles in new ad sets? Or the winning ad set? I turned off all the ad sets with angles that performed worse than the winning angle.
@Alex Martino for the stage you are at, I would take this approach.
1 campaign, 1 ad set
• Launch about 6 ads (eg: 3 videos, 2 images, 1 carousel) and make sure each individual asset is strong on it's own.
• Set your budget and every week, drop in 2-3 new ads.
• The goal is to keep your budget consistent and reduce your CPA/increase your ROAS
• As your CPA reduces or holds consistent, increase your budget
• Don't worry so much about turning ads off at this stage, you want to launch new ads that gain spend.
• Every week the goal is to improve something.
• Often not doing anything is also the best approach. Make sure you're not making changes daily. Allow 3-4 days before any new addition/change
• Make sure you're keeping an eye on the business metrics as well (nCPA, NC-ROAS). I look at Meta spend as a dial and as long as the business metrics are where I need them to be, turn up or down the dial (spend)
• If you want to turn an ad off that's getting spend, make a hypothesis ("if I turn off this ad I believe the spend will distribute to the other ads and reduce the account CPA). Make the call, assess it in 3-4 days and turn back on if incorrect.
You don't need loads of ads. Too many with small spend creates no data validation.
Check out the images attached. These are both 30 day look-back windows. Both clients have 5 or less ads running.
I've tried the mass creation, but my number one recommendation is be more systematic, find creators who are a great fit for the brand and produce good content.
In 1-2 months, it's much better to look back and take calculated steps every week.
@Fergal Dinan really helpful, thanks!
Anything change in terms of advice here if you just have statics? We don't have enough assets to create videos. Though I suppose I could create a video off concepts.