Friday confession: I spent more time this week thinking about distribution than building
Building 4 side projects while working full-time means every weekend hour matters. And this week? I barely touched code.
Here's what actually happened:
Monday-Thursday (day job): Zero product work, obviously.
Friday evening: Had this whole plan to ship a new feature for SomeYum. Instead, I found myself in a rabbit hole reading about how other indie hackers are getting their first users. Two hours gone.
The realization hit me hard. I've shipped features nobody knows about. I've polished UIs for apps with zero traffic. Building is the easy part for me. Distribution is where I'm completely lost.
So I'm shifting focus. Next week, before I write any code, I'm doing one distribution experiment per project:
- SomeYum: posting in 3 recipe subreddits
- TellMeMo: reaching out to 5 people who complain about meeting overload on Twitter
- VibeCheck: one cold email to a dev newsletter
Small experiments. Just to see what gets any traction at all.
Anyone else caught in the "build mode" trap? What finally got you to switch to distribution?
@ItsKondrat Good question! It's more comprehensive than just subreddit recommendations.
A typical brief includes:
Where your ICP actually hangs out — specific subreddits, forums, Discord servers, Twitter hashtags, with subscriber/activity counts so you know which are worth your time
Recent threads with buying intent — actual links to conversations where people are asking for exactly what you build, or complaining about the problem you solve. Not generic discussions, but threads where engagement would make sense.
Customer language — the exact phrases people use when describing their pain. "I hate scrolling through 10 recipe blogs" is gold. Way more useful than guessing at marketing copy.
Competitor gaps — what people complain about with existing solutions. Where you can differentiate.
For SomeYum, I'd look at meal prep communities, busy-professional cooking threads, and people frustrated with recipe blog SEO spam. Happy to put one together if you want to see the format firsthand.
The build trap is real — I spent 6 months building features for my AI SaaS before realizing I had no idea where my customers hung out.
Your experiment approach is smart. A few things that worked for me:
Narrow the ICP first. "Recipe enthusiasts" is too broad. Meal preppers, busy parents, and keto dieters hang out in completely different places.
Listen before you post. Lurk in subreddits and note the exact language people use. Then use THAT language in your posts.
One project at a time. 4 projects = 4x distraction. Go deep on one.
I've been doing AI-generated customer discovery briefs for pre-PMF founders — basically a report on where your ICP hangs out, what they complain about, and specific threads to engage with. Happy to put one together for SomeYum if you want a sample.
This is solid and the ICP thing hit home. You're right, "recipe enthusiasts" is way too fuzzy. When I think about it, the people actually using SomeYum are mostly busy folks who hate scrolling through 10 recipe blogs. That's a sharper angle.
The "one project at a time" advice is tough for me. I keep them separate intentionally - when I'm stuck on one, I switch to another and usually come back with fresh eyes. But you're probably right that it slows down the distribution work since I never go deep enough on any single channel.
Curious about the discovery brief - what does the output actually look like? Is it mostly subreddit recommendations or do you pull actual threads/posts to respond to?