Why I'm writing in public: trust before sales
I write in public because it is the only marketing that works before anyone has heard of you. Attention costs money I don’t have, and trust resists every pitch. What works is being useful in the open, over and over, until the people who need what you do already know your name.
The logic is boringly simple. You share your work, people see how you think, some of them start to trust you, a few follow along, and when they need what you do they already know who to ask. You cannot pitch your way into trust. You earn it by being useful before anyone owes you anything.
Why it works when nothing else does
- It is proof, not a claim. A post that walks through an analysis is evidence you can do the analysis. A landing page saying “expert in machine learning” is evidence of nothing.
- It compounds while you sleep. Every post is a permanent answer to a question someone will eventually search for. Ads stop the day you stop paying. A post that answers a real question keeps working.
- It filters. The people who find you through a post about RFM segmentation want RFM segmentation. That beats any lead a cold email ever produced.
- It forces you to understand the thing. A method you only half followed produces an explanation that reads like one. Writing is the debugger for your own understanding.
The idea is old. Nadia Eghbal’s Working in Public makes the case for open-source maintainers. Basecamp’s communication guide makes it inside a company: “Writing solidifies, chat dissolves.” The pattern holds either way. Work done where people can see it can be checked, borrowed and built on. That is worth more than the same work done in private.
The rule I’m holding myself to
Publish the thing that was hard to figure out, not the thing that makes me look good. Those are different posts, and only one is worth anyone’s time. Concretely:
- Show the data or the code. If a claim rests on a number, the number gets a source.
- Write up the failures too. The model that did not work is more useful than the one that did, and much rarer to find written down.
- No paywalls on the actual insight. If the useful part sits behind a subscribe wall, it was marketing pretending to be writing.
What that looks like here
So I am going to write down what I learn doing data science and machine learning. Methods I find useful, papers I liked, things that surprised me, mistakes I made so you don’t have to. No paywall, no gatekeeping, no “subscribe to unlock the real insight.” Just the actual work, in public.
The obvious objection is that giving the work away costs you the sale. My experience says the opposite. Anyone who could do it themselves from a blog post was never going to hire you. Anyone who was going to hire someone now has a reason to pick you. You keep everything you had, and gain the one thing money cannot buy.
Some of it will be useful to you. Some of it will just be me thinking out loud. Either way it is here, and it is free. If it resonates, say hi. That is how communities start.