I comment often enough when talking to clients or new hires that I have been running growth & marketing for the better part of 20 years and I still get it wrong 70% of the time.
The CrowdTamers AI launch is one of the 70%. Initial launch results are all pretty bad, but that’s not actually discouraging at all, as we learned a lot.
Let’s dive in!
I deliberately didn’t promote or hustle on social media with my audience to try and drive organic signups. I’m writing this up as a case study for my own approach for founders who don’t have an audience, so I would invalidate my own case study if I cheated. 🙂
All of the experiments I’ll be doing around launching CrowdTamers AI will be based on my Million Dollar Funnel, so check that article out first so you know what we’re doing here.
The Goals
First up it’s worth keeping in mind what we’re trying to accomplish with our ER-1 launch ads for CrowdTamers AI.
| Goal | Target | Result |
| Impression | 5000 | |
| Click-Through Rate %age | 2.5 | |
| Clicks | 126 | |
| Signups | 20 | |
| Spend | 300 |
Positioning Tests
We were testing 2 different ideas:
- Audience targeting – are we better off trying to sell CrowdTamers AI to founders or to marketers?
– My bet: Founders would do better - What copy + image will do best?
– My bet: The pug. Look at the cute face!
Quantitative Results
| Goal | Target | Result |
| Impression | 5000 | 26902 |
| Click Through Rate %age | 2.5 | 0.42 |
| Clicks | 126 | 112 |
| Signups | 20 | 4 |
| Spend | 300 | 224.7 |
Positioning Test results
Test 1: Which persona responds better?
It’s a little hard to say—performance was pretty bad across both personas, but the “Binoculars” ad was targeted to marketers, and the other two were targeted to founders. I have a new relatively weak hypothesis that founders are not actually the best market for CrowdTamers AI—as least, not given the current positioning.
Future Experiments
- Test CMO / Director of marketing persona more rigorously independently with new creative
- Test Founder persona with a different message entirely.
Test 2: Which copy / design performs better
Again, the signal here is quite weak, to the point where I don’t think I can make any conclusions. All copy performed much worse than the target, so we need to scrap it all and start again.
Future experiments
- Completely new creative needed.
- Test creative against one persona at a time.
Next steps
As I already wrote last week, I’ve gone through a complete re-write of my ads and now have 10 (!) new ads to try against my marketing persona. Let’s see if any of these perform notably better.
It can be easy to be discouraged when you see your launch ads perform so poorly out of the gate. It’s worth noting that I’ve been doing this for 20+ years, I called both of my tests wrong last week, and all that means is I’m excited to see which of my tests next week might do better.
