Ok Iāve been getting feedback that people like hearing about ads, so Iāll share what I learned today.
I was mentally fatigued bc Iāve launched maybe 8-9 creatives the past 3 weeks, with at least 2 variations, and a maximum of 4 variations. So total maybe 30 ads in 3 weeks. I learned a lot about fonts, colours, format, made a million bazillion headlines. My head was just hurting so today I had to do something different.
I focused today on mainly campaign structure, and using a lookalike audience to run a sales campaign. The problem is that whenever you run a sales campaign, it automatically categorizes your audience as āadvantage+ā (their AI audience tool) and you have to manually turn it off. But itās hidden deep in settings, and ChatGPT was not updated on meta ads. I watched like 7 YouTube videos, I looked through Twitter, Reddit, and probably 3-4 Facebook documentation articles on lookalike audiences. The problem is that semantic search on their docs are very bad, so the information was hidden deep. Eventually I was able to turn it on manual and do what I wanted.
The second thing I learned is that Iāve been killing ads too fast. I wanted my creatives to be so good that when I turn the campaign on they just print money. That obviously is not happening and I blew a bunch of money learning. But thatās what you gotta do to get there. So Iām letting my new campaigns run a bit longer just to see, even if they arenāt profitable in the first 24 hours. My product is fairly expensive so I have to play with a higher adspend to start.
The last thing I learned is that for meta to actually give me good estimates, I need to do at least 50 sales thru ads. Iāve done a few but no where near that amount. So I changed my campaign such that I focus on add to carts, so my campaign can warm up faster and meta can have more data sooner.
Even though I didnāt solve my problem of my brain hurting from thinking and searching Iām glad I learned today š 3 replies
0 recast
31 reactions
0 reply
0 recast
0 reaction