Originally published at moday.me. Building MODAY in public.

I built a store that can sell to the whole world. I just can't see who I'm selling to.

Five days out, and the thing I'm most afraid of is plain silence

Five days until MODAY opens on 5/18.

I've been writing fairly upbeat posts so far. The stack came together in three days because I handed everything to Claude. The distribution pipeline auto-fans-out across nine languages. The plumbing works.

All of that is true. On the engineering side, I have actual traction.

What scares me right now is something more boring than any of that.

Doors open. Nobody comes in. Zero orders. The plain version of the fear.

"We don't know if it'll sell yet" — that part is fine, that's how every launch starts. The scarier part is the next layer: I don't know what would make it sell, either.

I can't picture marketing-that-works for an international audience

MODAY was built global from day one. Nine languages, multiple currencies, Gelato printing locally in each region, Shopify Markets handling the routing. As a system for selling, it's open to the world.

What I don't have is a clear picture of how to bring people into that system.

More precisely: I can't picture international marketing that actually works. The "actually works" is the load-bearing word. There's no shortage of forms to try. Post on social. Set up SEO. Send press notes. Write on Reddit. Launch on Product Hunt. Spin up TikTok shorts. I know the playbook items.

What I can't see is the version that pays back the initial costs. That's the honest answer.

Paid is off the table for now

The fastest path would be paid: Meta, Google, TikTok ads. Especially for an international audience, dropping budget into paid is the quickest way to run initial validation loops.

But I don't have the budget. It's a one-person shop, and there is no ad spend. Simple fact.

This isn't an argument against paid — it's a sequencing thing. Once I see the rough shape of who's buying and why, paid is obviously a good amplifier. Until then I have to find traction without it.

Short-form video is the one thing I can half-imagine

The one channel I can sort-of picture is short-form video on social.

The product is visual enough that "oh, nice" lands in a single frame. Someone wearing a "MONDAY: System Booting..." tee, sitting in front of a camera Monday morning with coffee, one cut. For the right cultural pocket, I think that frame can travel.

The video mock-ups I can build with AI — Fashn.ai for the model composite, Kling or HeyGen for the motion. I've already sketched that stack.

But it's not running yet. The stack exists in my head; the implementation is behind.

I won't really know what works until I start running these videos and watching what actually catches. The plumbing is something AI can build for me. The hard part is what to show, and to whom.

The real problem — I can't see the customer

Strip away the marketing-channel anxiety and the truth underneath is simpler.

I can't see the customer.

I wrote earlier that "day-of-the-week tees should land with engineers and geeks." That's still a hypothesis. Who actually buys it, where they live, what mood they're in when they decide to buy — none of that has resolved into a real image in my head.

The picture I want to be able to draw, but can't yet: a 28-year-old in Berlin writing Rust, scrolling Reddit on a Friday night, and a "FRIDAY: Build Successful ✓" tee comes across his feed, and he laughs, and he adds it to cart. That specific scene, with a real face attached.

I have something blurry. But the face isn't there. What language is the page in for him. What device. What mood. The resolution is too low.

Without that resolution, I can't decide what to film. I can't decide which of the nine languages to push first. Before any acquisition tactic comes the soil that the tactics grow in, and I don't have the soil yet.

This isn't something AI can answer for me. AI can produce endless generic "marketing to engineers" advice. But the actual question — who, specifically, is MODAY's customer — only resolves once I'm in the field watching it happen. AI doesn't have that observation. I have to.

So: I won't see the customer until the doors are open.

I don't have a way to sell. I have a way to eliminate ways of not-selling.

5/18 isn't moving. The door opens that day.

Probably nobody comes in at first. Anyone who does might not buy. But once the door is open, signal starts arriving for the first time.

Which language page got viewed. Which day-of-the-week tee got added to cart. Which video frame made someone pause for half a second.

I don't have a "way to sell" right now. What I will have, once we open, is the slow process of crossing out, one at a time, the things that don't sell.

That's where the real brand probably starts.

The fear is real. I'm opening with the fear still attached.

More soon.

— Yoskee

moday.me