Does building in public actually work?
Short answer
It works for a specific and narrow purpose: building an audience of other founders who are interested in the process of building. That audience is genuinely valuable for feedback, encouragement and early testers, and it is almost never your customer base. Most founders who build in public successfully end up with an audience that will cheer their launch but not buy the product, because the people watching are makers, not users.
The longer version
It works, but for a narrower purpose than most people who try it expect. Building in public reliably builds an audience of other builders — people who find the process of making things interesting. That audience is genuinely useful: they give feedback, they test early, they are encouraging when you need it, and they will show up on launch day.
What they will usually not do is buy the product, because they are makers rather than users. This is the trap. You get real numbers going up — followers, engagement, launch-day traffic — and none of it converts, and it takes a long time to work out that the audience you built is simply not the audience for the thing you built.
It works properly when your customers are also builders, which is why it looks so effective for developer tools and founder products. Outside that overlap, treat it as a way to find collaborators, feedback and moral support, and get your customers somewhere else.
The full argument, with the rest of the picture around it:
Every Way to Promote a Product, and Why Developers Get Each One Wrong