Skip to content
The Log

Month 67

month-67.png

Production is a different animal than preview — I knew that going in, in the abstract. What I didn't understand is that you can't plan your way past it; you have to iterate through it, hit the thing that only shows up once real infrastructure is on the line, and fix it. Every week this month turned up another piece — a contract to sign, a credential to split out, a small tweak that preview had quietly been covering for. None of it was hard on its own; it was the accumulation, learning the gap one piece at a time instead of all at once. And I'm fairly sure I've over-provisioned the resources for what's running on them right now — better that direction than the other, but it's not free, and I know it.

The Numbers
month 67 · august 2026
Hours
174
this month
Issues closed
231
this month
Blockers
10
launch blockers
Days to launch
full launch · date open
Spent
€822
this month
Revenue
€0
monthly
Issues closed 231
Blockers resolved
64/74

What Moved

All the boring, load-bearing pieces landed: pricing and VAT are consistent everywhere a price shows up now; billing can actually take a payment and show what's on file; error reporting stopped silently dropping events, and this time I built a canary so it can't do that quietly again. Contracts got signed. Credentials that preview and production had been sharing got split apart, one at a time, as I found each place they still overlapped. None of it is a feature anyone will notice. All of it is the difference between a demo and something you can point real traffic at.

I'd be lying if I said I've found everything. Production turns up its own list of problems, at its own pace, and I'm sure there's more of it waiting.

What Didn't

The over-provisioning isn't the part I'm not proud of — it's a trade I made on purpose, and I'd make it again. Paying for headroom I might not need yet is cheaper than the alternative, and it buys the kind of scalability I can't retrofit once something's under load. What sits heavier is what I've been asking of the people who've actually been testing this so far — not a beta signup, but their time and attention, week after week, answering my questions and running through flows so I could find the next thing to fix. That's a real cost to them, not a hypothetical one, and I'm not sure I've been paying it back yet.

Would I Pay For This Today?

Yes. Every time I push a PR on this website and a working preview pops up seconds later, running on infrastructure I built — that's the answer, right there. It's not hypothetical anymore; it's the thing happening every day I work. It fills me with a kind of pride that's hard to fake, and it's the closest I've gotten to feeling like the platform pays for itself, whether or not a single euro has landed yet.

Next Month

Keep chasing feedback — not waiting for it to land in my inbox, going and asking for it. Close out what's left of the blocker list; ten of them are still open, and I want that at zero before I write the next one of these. And keep sanding down the rough edges in the day-to-day experience — the stuff that doesn't break anything, just wears on you a little every time you touch it — until running strackt actually feels as smooth as it needs to.

It works, it scales, now let's see how it breaks.