Week 2 - a guarantee, honest copy and a launch date.

I'm building Paperjoy, a service where a robot writes real-pen handwritten postcards and mails them anywhere in Poland. Last week was about changing plans. This week was about getting everything around the robot ready for launch.

Last week's list.

Quick check on what I promised last week:

• Drafts. Done. Agents can now create a draft, preview it, update it and only then submit it, so nothing hits the plotter without a human saying yes.
• Ship samples to leads. Done. I took the whole batch to the post office on Thursday.
• Second machine. Mostly built, I had to re-order a linear rail block that has been binding.
• Another SEO page. Two, actually. More below.
• Rebuild onboarding. In progress. A new 4-step onboarding and an empty-state dashboard are in a PR, still being polished.
• Launch assets. Started: a first brand book and the groundwork for a press kit.

Not bad.

Is the post office reliable enough?

This was the question I spent the most time on. Poczta Polska doesn't publish a loss rate for ordinary mail, so I dug into what is public. In UKE's 2025 tests, 86% of economy letters arrived within 5 working days and 98% within 10. The real problem isn't lost mail, it's late mail.

Registered mail would cost 7.80 zł per letter, which would eat the whole margin. It also has to be signed for, and nobody wants a thank-you card that feels like a letter from the tax office.

So I'm staying on economy postage and adding a 30-day delivery guarantee instead: if a card sent to a correct address hasn't arrived within 30 days, you get your credits back. It's in the terms now and has its own page at /gwarancja. With 98% arriving within 10 days, it costs me almost nothing and hopefully makes trying Paperjoy a lot less scary.

I also scoped priority postage, but parked it until after launch. Keeping the launch scope small.

Prices.

New per-card prices went live: Start 10.90 zł, Pro 9.00 zł (after a second small tweak on Thursday) and Business 7.50 zł per card, net, postage included. Monthly fees didn't change. The main reason: Business finally makes money on every card.

SEO.

The Christmas cards for businesses page is finally live (it had been a 404 for way too long), and so is the thank-you cards page. Both now show carousels of real cards that I photographed this week instead of mockups. The season ramps up in November, so the timing matters.

Monday's SEO audit was good news. Rankings are starting to move: "kartki z podziękowaniem za zakupy" (thank-you cards for purchases) went from #26 to #18 on desktop, and "odręczne kartki dla firm" (handwritten cards for businesses) entered the top 40 at #29. The Christmas page is only a few days old, so it doesn't rank yet. Also, there was a first visit from ChatGPT!

Small things that took a lot of time.

• Envelopes follow Poczta Polska's addressing rules. The city and postcode line is always in capitals.
• Polish typography. Both handwriting fonts now have proper „Polish quotes" and an ellipsis.
• Better plotter pacing. The G-code now has explicit feed rates. Pen down, up have better timings now,both speed and quality are better now.
• Faster first loads after deploys, and the warm-up requests no longer spam my monitoring with fake "slow route" alerts.

Hardware: a fountain pen lifter.

As an experiment, I spent Sunday working on a fountain pen lifter model for the plotters. More on that once it's further along.

Launch date.

Paperjoy goes public on Monday, 2 November. 1 November is a public holiday in Poland, so everything launch-related moves one day. With two plotters at about 6 minutes per card and envelope, I can write up to about 120 cards a day.

Next week.

• Finish onboarding and merge it.
• Press kit done by Friday.
• Start building backlinks.
• Get everything ready for launch day.
• Keep going with the fountain pen lifter.

One week to go. See you next week.