“Build in public” became gospel so fast that nobody stopped to ask which parts of the kitchen were supposed to stay closed. One day it was a quirky habit of a few indie hackers tweeting their revenue charts; the next it was doctrine, repeated at conferences and in pitch decks as though secrecy were a character flaw and a public dashboard were proof of virtue. Somewhere along the way, sharing the work quietly mutated into performing the work loudly.
The instinct is understandable, and at its best it is genuinely powerful. Founders now broadcast their monthly recurring revenue, their churn, their failed launches and their unfinished roadmaps, and audiences reward the candour with attention, trust and sometimes customers. The logic is sound: transparency lowers the cost of belief. When a founder admits a mistake before a critic can point it out, they look honest rather than caught. The research backs the upside — most leaders, it turns out, under-share rather than over-share, missing easy chances to build warmth and trust. The whole playbook of founder-led growth runs on this fuel: post your bets publicly, and let yourself be publicly wrong.
But a café is not a single room, and this is where the metaphor earns its keep. The front of house is meant to be seen — the chalkboard menu, the hiss of the machine, the barista’s easy patter, the warm theatre of a place that wants you to stay. That is the part you put on display, and you should. Then there is the back room, where the actual recipe gets refined: the supplier nobody else has found, the roast curve dialled in over a hundred ruined batches, the margin maths that make the whole thing survive. No serious café livestreams its supplier contracts, and none invites the customers in to argue about the roast curve mid-batch. The performance out front is real hospitality; the work out back is real advantage. Both are honest. They simply belong to different rooms, and confusing the two is how you end up with a beautiful room and no recipe worth protecting.
Here is the part the gospel skips. Transparency builds trust and an audience, but indiscriminate openness quietly bills you for it. Strategy is the first thing to leak. Post the roadmap and your sharpest competitor reads it for free, then ships your best idea while you are still narrating the plan. Noise is the second cost. Open the decision to the timeline and you invite a thousand opinions on a choice that needed exactly one, and the founder who answers every reply has stopped doing the work to talk about doing the work. And there is a subtler tax: openness can curdle into theatre. Harvard Business Review has warned for years that transparency can backfire — that being watched changes behaviour, nudging people to optimise for how the disclosure looks rather than what it does. A vulnerability post engineered for engagement is not vulnerability. It is content wearing vulnerability’s apron. The difference between confessing a failure and performing one is whether you actually learned anything before you hit send.
None of which makes openness the enemy. The most quietly devastating case for transparency is that a16z, by its own telling, weaponised it in an industry built on secrecy — turning visible thinking into a moat while everyone else hoarded theirs. But read that story closely and the lesson inverts the cliché. They published their thesis, not their term sheets. They made their ideas loud and kept their edge deliberate. What looked like radical openness was a curated front of house with a very disciplined back room. That is the whole trick, and almost nobody copies the disciplined half.
So the question was never whether to be transparent. It is which room you are standing in when you decide to start talking. Show the room, share the journey, let people watch the machine hiss and the orders come in — that is how trust gets made, and pretending otherwise is its own kind of dishonesty. But brew the real recipe in private, where the heat is, where the failures are honest because no one is clapping. Not everything worth doing is worth narrating, and the founders who understand that are usually the ones too busy refining the roast to livestream it.
