about
Twenty blog posts. Twenty different answers.
That is the week before a launch. Post at 12:01am. No, post Tuesday. Build an audience first. No, that is a waste of a year. Charge more. Charge nothing. Everybody sounds certain and almost all of it is one person’s good week, written up afterwards as though it were a rule.
I got tired of guessing which one to believe. So I went and read what a few hundred builders actually wrote about their own launches, and counted how many of them arrived at the same thing without ever talking to each other.
What I actually did
- 01
Read the writing, not the tweets
Builders announce launches on social and explain them on their own sites. So the corpus is long-form posts, retrospectives and teardowns, read end to end rather than skimmed for quotes.
- 02
Pulled out one tactic at a time
Every discrete move became its own claim, carrying whatever number it produced. No summaries, no paraphrased vibes. A seven-step Product Hunt walkthrough becomes seven things you can act on.
- 03
Only trusted what repeats
Claims describing the same underlying move get grouped, and the count of separate builders who reached it becomes the signal. One person saying it is an anecdote. Five arriving at it independently is something else.
- 04
Kept the arguments in
Where builders flatly contradict each other, both sides stay on the page with their evidence. Averaging them into one tidy answer would throw away the most useful thing in the whole corpus.
Who is doing this
I’m Stephen. I work in marketing in the UK and I’ve spent about nine years teaching myself to build software in the evenings. So I sit in an odd spot: marketing pays the bills, but I ship the things too.
That gap is the whole reason this exists. I have run campaigns that looked clever and moved nothing, and shipped small things that went nowhere for reasons nobody warned me about. After a while you stop wanting opinions and start wanting to know who actually tried it. Most of what I build in my own time is open-source tooling, and it comes from the same itch: take a thing people do by hand and badly, and make it checkable.
One rule I will not bend. Nothing here is quoted. Every line is written fresh and checked against the source, and every tactic links back to the post it came from so you can go and read the person who did it. If that means the site is smaller than it could be, fine. It is their work. I am just the one who counted.
Go and find something that worked.
Start where you are. Most of what decides a launch happens before launch day, so that is usually the honest answer to where to begin.