Postmint

Two weeks ago I launched Postmint, a tool that turns one sentence into a finished social graphic in your brand colours. Eight days later: about 118 unique visitors, zero signups that weren't me. Here's the whole autopsy, because the posts that helped me most were the ones with real numbers in them.

My first prototype generated marketing graphics straight from a text prompt. I made twenty images, looked at them, and would not have posted a single one. You know the look: half-empty canvas, a grey gradient doing nothing, a fake button painted on that leads nowhere, a number the model invented. Technically an image. Also obviously slop, and posting it damages a brand more than posting nothing does.

So I deleted it and spent months on the unglamorous part.

What the thing actually does

The model composes an HTML document. That document gets rendered in a real headless browser, and then two separate passes look at the result.

The browser proves geometry. It walks the rendered DOM and reports five text faults it can measure rather than guess: text extending past the canvas, text clipped by an ancestor that hides overflow, a box too small for its own text, text painted over by an opaque shape, text overlapping other text. Facts with element references attached.

A vision pass then looks at the actual screenshot and judges it against the brief and the reference style. Contrast, empty composition, everything geometry can't see.

Both feed a revise step, and the loop runs a couple of rounds before you ever see the result. If a render or a model call fails mid-loop, it falls back to the best version so far rather than erroring out.

The launch, honestly

• About 118 unique visitors over 8 days. Half DE, half US.

• Roughly 2 of them reached the register page. 1.7%.

• Zero external signups.

• 80% logged as "Direct", because I tagged exactly nothing.

• My Show HN got flagged in 25 minutes. Baity title, AI-fatigued crowd, cold account, and a link straight to a signup wall. Entirely deserved.

Two bugs did most of the damage.

The landing page was the first. Visitors from social bounced in about five seconds. The opening screen explained the product where it should have shown what the product makes, and nobody scrolled far enough to find out.

Then the one that stings. My analytics never recorded a single signup, because the goal wasn't firing at all: it got lost across a redirect. So the number I actually cared about, signup → first generation, was structurally unmeasurable for the entire launch. I spent a week reading a dashboard that could not have shown me a signup even if there had been one.

That second bug is why I'd push this on anyone about to launch. Register a throwaway account in an incognito window and confirm each funnel event actually fires. Five minutes. I skipped it and burned my only launch spike.

Since then I've fixed the goal, cut a release, and watched an incognito signup fire the events end to end. The funnel finally records. I rebuilt the landing's first screen around output rather than explanation. I killed the blank canvas too: a brand-new user used to land on an empty brief box and get asked to invent the one sentence they don't yet know the shape of, so it's prefilled with an example now and the first render is a button press. Every channel link is tagged, so this time I'll know where people came from.

Still zero external paying customers. The gallery of what it makes is here if you want to poke at it, no signup needed to look: https://postmint.de/ih

The question I actually want answered: Postmint is German-first, the buyers I have in mind are DACH small businesses and shops, and yet the two most engaged audiences in my launch data were IndieHackers and Product Hunt. English, technical, 40+ minute sessions. Is that real interest worth chasing, or builders admiring the machinery who will never buy? I can't tell, and it changes what I build next.

Earlier this year I had a working prototype that turned a text prompt into a marketing graphic. I made about 20, looked at them, and realized I wouldn't put a single one on my own account.

You know the look. A half-empty layout, a gray gradient in the background, a button drawn on that leads nowhere, and usually a number the model just invented. It screamed "a machine made this." So I deleted it and spent the next few weeks rebuilding from scratch.

My bar the whole time was uncomfortable and simple: would I actually post this on my own account? For most results the answer was no for a long time.

What changed the output:

• Every design starts from a hand-made template instead of a blank canvas, so it looks designed, not random.

• It reads your brand kit (colors, font, logo) and applies it automatically.

• It checks its own text for contrast and legibility and fixes what fails, instead of shipping unreadable copy.

• You refine by chatting ("make the headline bigger", "shorter") instead of opening design tools.

It's live at postmint.de. One sentence in, a finished post out, in Instagram / Story / Carousel / LinkedIn sizes. Free tier is 10 posts a month, no card.

I'd genuinely value feedback from this crowd on three things:

1. Run one generation. Does the result clear the "would you post this" bar, or where does it still look off?

2. Did anything in the flow make you hesitate before you even generated? (My worst number is people who sign up and never try it.)

3. The landing page and positioning: does it land, or what's confusing?

Tear it apart. I'd rather hear it here than not.