One thesis: the 1D UPC on your pack is the only asset you own that generates data exclusively for other people. Every scan happens on a retailer's lane, posts to a retailer's transaction log, and reaches you — if it reaches you at all — as syndicated data resold back weeks later, aggregated past the point of usefulness. The mark's owner is the one party the mark carries no signal for.
This post is the engineering-side account of that asymmetry, written for the brand seat that owns the artwork and has never had a first-party line to the consumer of a specific unit. The executive-register version — the budget argument, the retailer-timeline argument — lives on the family's business door: Your barcode was a checkout token; the same square can be a line to your customer. This is its dark twin: the same claim, with the payloads shown.
The information budget of twelve digits
What a UPC-A physically carries is twelve digits: your GS1 company prefix, an item reference, a check digit. Run one through the resolver:
$ barcoding resolve 036000291452 --json
{ "scheme": "gs1.gtin", "carrier": "upc-a",
"canonical": "00036000291452", "check": "mod10 pass" }That is the whole payload. No lot. No expiry. No unit. It answers exactly one question — which product class is this? — and it answers it to a fixed scanner at a lane you do not operate. The lane transaction it triggers (item, price, basket, timestamp, store) is created on the retailer's hardware and belongs to the retailer's data estate. What flows back to you is what the syndication and panel industry resells: class-grain, weeks-latent, and identical to what your competitors buy about you.
Now the same printed square, changed only in what it encodes — a QR carrying a GS1 Digital Link URI with the AIs a brand may choose to add:
$ barcoding resolve 'https://id.example-brand.com/01/00036000291452/10/L23085/21/00417' --json
{ "scheme": "gs1.digital-link", "carrier": "qr",
"canonical": "00036000291452", "check": "mod10 pass",
"ai": { "10": "L23085", "21": "00417" },
"resolves": "https — any phone camera" }Class, batch, unit — and an address. Twelve digits became a record with a serial number and a destination you own. Same square of ink, two entirely different economies.
The mechanics, stated within the claims discipline
Because this territory is full of vendor copy that fails a GS1-literate reader's check, the mechanics in one honest paragraph: only a QR carrying a Digital Link URI resolves in a consumer's phone camera — GS1 DataMatrix clears the POS lane but native camera apps do not reliably decode it. An element-string payload is not a URI and resolves to nothing in a browser. And Sunrise 2027 puts nothing on your pack: the programme's baseline is that retail POS can extract a GTIN from a 2D symbol by end of 2027. Which symbology you print, which payload form, and which AIs travel — those are your decisions as the GTIN owner. The decision is landing on your desk inside an artwork cycle either way; the only question is whether it is made as a compliance chore or as an encoding decision about who your mark can talk to.
That is why this is a budget unlock and not a line item on the supply-chain ledger. A trade-compliance mark competes for compliance money. A mark that can carry a first-party address competes for brand and insights money — the budget that already pays for panel data precisely because no first-party signal exists.
The honesty block
Here is the sentence the category's marketing will not print: whether consumers scan at volume is a hypothesis, not a fact. The mechanism above is GS1's published design and you can verify every step of it with a phone and the resolve transcript on this page. What no one can show you is proof that shoppers in your category will point a camera at your pack in numbers that matter — that evidence does not exist yet, for anyone. The mark creates the option; the option's value is the bet.
What is checkable today, you can check without us: print a Digital Link QR for one of your own GTINs — generate renders it locally from your GTIN and the AIs you choose, no agreement or account required — scan it with your own phone, and watch twelve digits become a resolvable address. The write side is free because GS1's standards are royalty-free by policy; the grammar lives at /schemes/gtin.
Where the signal would have to land
If the option pays, every scan is a record that needs somewhere to live — a record layer that is yours, not another network's. That capability is what the access flow asks about: say what arrives at your door and what you print on it, and the provisioning order follows the list. It costs one email address, and it is the only thing on this page that asks for anything.
The asymmetry, restated once: today your barcode is scanned more times per day than any other artifact your brand produces, and you learn nothing from any of those scans. The 2D transition is the one moment the economics of that square get renegotiated. The retailer's lane is upgrading on GS1's clock either way. Whose mark it reads — a checkout token or a line to you — is decided at your artwork desk.
Keys open the network half of the verb set. Get an API key — say what arrives at your door, and the provisioning order follows the list.
We answer in writing. We take at most five conversations a month, only when you ask for one, and only after you already have the written read.