πŸ”΅ A/B Split-Testing Results πŸš€
A/B split-testing the original LTO funnel vs a new layout.
So far, so good. (see image below)
3.8% conversion rate and up from the previous one.
Testing this new page for 2 of my LTO's, both for organic and paid ads. (yes, I have different funnels for different traffic)
A lot of low ticket offers use a typical "AC funnel" outline, pioneered by Alen Sultanic.
I wanted to create my own 100% original version so there's no issues when I share my templates with people in my community.
As always, I create a bunch of versions and spend thousands to test them before I share them.
I'll drop the link to the new one if you want to check it out.
It follows a more traditional "e-com" funnel, but for digital products.
In the end, the "sequence of sections" matter less than the psychological elements of a high-converting sales page.
Can you make your offer feel new and different than what they've already tried?
Can you get them faster results by offering a quicker shortcut?
Can you offer them more convenience by removing most of the effort from their plate?
Those are the keys to a great offer.
Then for the sales page, you want 90% of these pillars:
- Curiosity (keep attention long enough)
- Trust (believability)
- Credibility (authority)
- Social proof (result for others, happy customers)
- Transformation (desired outcome)
- Convenience (shortcut)
- Future pacing (imagine what can be)
- Risk reversal (stakes)
- Urgency (why now)
Most importantly...
Answer the questions in their mind, even the subconscious ones.
"How's this different than what I do now?"
"Does this work in my situation?"
And so on.
Fun times.
Happy Friday,
G
14
4 comments
Gusten Sun
7
πŸ”΅ A/B Split-Testing Results πŸš€
Freedom Fastlaner
skool.com/fastlaner
Build an online business you actually love and create freedom by design.
Leaderboard (30-day)
Powered by