The biggest shift after 3.5 years:
building is maybe 20% of the game.
The harder parts are:
- distribution
- validation
- consistency
- talking to users
- finding painful enough problems
What’s interesting is that uncertainty never disappears.
At first the uncertainty was: “Can I even build something?”
Now it’s:
- “Am I solving the right problem?”
- “Is this niche strong enough?”
- “Should I double down or move on?”
I’ve definitely shifted toward a distribution/validation-first mindset.
But I still don’t fully know what the correct approach is.
Should one:
- build in domains they genuinely care about?
- or enter boring underserved niches where real money/problems exist?
For now, I’m trying to learn both:
how to spot real pain and how to distribute solutions consistently enough for compounding to happen.
3.5 years of chasing “indie hacking sovereignty” outside my 9-5 job taught me something uncomfortable:
Most of my early projects failed long before launch -because distribution was either ignored or treated as an afterthought.
A quick retrospective 🧵
7) PrivacyStack
A privacy-focused boilerplate for building small client-side tools with optional monetization.
The idea made a lot of sense when I started:
help developers quickly launch tiny SaaS products without databases or backend complexity.
Then AI coding tools exploded.
After starting to use Claude Code heavily, I realized something uncomfortable:
the value of boilerplates changed dramatically almost overnight.
First time I truly experienced how quickly market timing can invalidate a product direction.
8) RemoteTaxCalc
Probably the first project where I can clearly see SEO working “as expected.”
The idea: remote worker tax calculators + digital nomad visa content.
This time I approached things very differently:
- SEO from day one
- programmatic content
- internal linking
- distribution awareness early
- continuous publishing
And for the first time: Google started responding relatively quickly.
Still tiny, but:
- impressions grow steadily
- pages index properly
- traffic compounds slowly
- I finally understand why people say SEO is a long game
Now the problems are different:
- backlinks
- authority
- content maintenance
- understanding what users actually care about
Czech Republic tax calculator is live 🇨🇿, 25th country on RemoteTaxCalc.
The Živnostenský list (freelancer trade license) is one of the most popular paths for remote workers in Prague - but it creates full Czech tax residency.
What's inside:
- 15%/23% income tax brackets
- 60% flat-rate expense deduction for IT/trade freelancers
- CSSZ + VZP breakdowns for both employees and contractors
- Side-by-side comparisons vs Poland, Hungary, and Estonia
At $100K, a Czech contractor keeps ~$86K after the 60% expense deduction brings the effective rate to just 13.7%.
Employees? 26.4% effective rate - still competitive for Central Europe.
Try it: https://t.co/8kQpLcaRpQ
"What gross salary do I need to take home $5,000/month in Portugal?"
Added a new free tool that answers this for 23 countries. No signup.
https://t.co/YxJzXrgVd2
Actually, this was the third sale 🙂
They’re still irregular and feel almost random. Maybe the SEO improvements helped especially after applying some ideas from skils you've recommended, but there’s still been no active marketing.
Interesting feeling when something starts moving a bit on its own.
Turns out Gumroad requires $100 in sales before enabling post-purchase email workflows - a bit of a limitation😢.
A recent purchase on Batchr (while my focus is on RemoteTaxCalc) creates an unexpected situation: new user, new context, and unclear priority.
It’s an uncomfortable reminder that real feedback only comes once someone actually pays.
Still figuring out how to balance focus with these side signals from users.
Added a new free tool to RemoteTaxCalc:
Foreign income → PLN conversion using official NBP exchange rates for Polish tax residents.
Flow:
- select USD/EUR/GBP
- enter amount + payment date
- app automatically finds the correct NBP rate date (including weekends/holidays)
- returns PLN value + official NBP table reference
Built around the official NBP API instead of manual table checking/scraping.
Small feature, but removes a repetitive accounting annoyance for remote workers and freelancers in Poland.
Left my other project, https://t.co/WoeOLnE2WU, to “boil” for a while after shipping a few improvements (freemium, better SEO, pSEO).
Recently noticed improved GSC metrics and got another sale.
Interesting feeling:
- on one side, it’s a chance to talk directly with a customer and learn what they actually need
- on the other, every paying user creates a bit more maintenance responsibility
Also noticing how mentally expensive context-switching between two projects can be - especially when you’re still unsure which one is truly worth doubling down on.