I've written dozens of landing pages for my own products over the past decade.
Spent thousands on ads. Hired the coaches. Followed all the frameworks.
Most still converted at 2%.
Then I realized something that completely changed how I think about landing pages.
The problem wasn't the copy, the design, or the offer.
The entire model was backwards.
I explain what I mean here:
https://t.co/SPgAHnqtmo
The funny thing about AI is realizing how easy it is to mimic our (so-called) intelligence.
Are we really that smart? Or are we just good at pattern recognition?
The hard thing about talking to people is realizing that everyone thinks he's right, even though that's rarely the case.
The second hardest is realizing you might not be right either.
I send most of my support questions on the weekend. That's when I have time to shop.
And if a company answers, on the weekend, they've got me as a client. That's how impressed I am.
AI makes support run on autopilot. Support agents can automate that for tou.
Just released - @openai's official Ruby SDK beta!! built by @stainless, who kindly hired me for a couple of days to test it. It's cool - auto generated from OpenAPI schema. strict typing. more options for Ruby AI devs = a very good thing https://t.co/6jVO9nVrLK #ruby#ai
The #1 lesson I've learned about running a business is lead generation keeps the lights on.
I made an AI generator that's helps me brainstorm lead magnet ideas.
Link in profile if you want it.
If you're thinking about how much ๐ฐ you'll save switching from something like Heroku to Kamal, think again.
Lost a whole day with stupid errors that lack documentation.
@justinamarsh Well, you only go through the setup phase once per project. So it depends on how many production-ready project you create in a relatively short amount of time.
I've deployed other projects with Kamal, but I forgot how to do it because I don't do it that often.
Value is subjective, so it's almost impossible to come up with good product ideas unless you're building products for yourself.
The alternative is to talk to people. Which is also the hardest.
The problem with MVPs (minimum viable products) is people tend to underestimate what the table stakes are.
Making useful stuff is about helping people, not about features and buzzwords.