For the head of brand at a co-manufactured CPG who has just been told "the 2D barcode is a co-man / retailer thing" and is about to let a decision that is structurally theirs get made by default, in someone else's plant, on someone else's clock.
Here is the confusion that costs mid-market brands the most, and it is an ownership confusion, not a technical one. Your product is packed in a facility you don't own, by a co-manufacturer whose systems you don't control, under a retailer mandate you didn't write. From inside that arrangement it is easy to conclude that the barcode — the Sunrise 2027 mark, the symbology, the data in it — is somebody else's call. It is not. It is yours, and the reason is clean: you own the GTIN, and you own the artwork.
The ownership boundary, stated once
A GTIN is assigned by the brand owner — the party that owns the product identity — from that party's own GS1 company prefix. If you licensed the prefix and assigned the number, you are the GTIN owner. A co-manufacturer packing your product to your spec does not own your GTIN; they operate a line. A retailer selling your product does not own your GTIN; they scan it. The GTIN is a claim of brand ownership, and in a co-pack arrangement that claim is almost always yours.
And the barcode — the printed symbol carrying that GTIN — lives on your artwork, which you own and approve. The co-man prints what your approved artwork tells them to print. They do not get to substitute a symbology because their line prefers it; they do not get to decide the data carried in your mark. The artwork is the instruction set, and the instruction set is yours.
Put those two facts together and the whole "is this mine to decide?" question dissolves: the identity is yours, the artwork carrying it is yours, therefore the symbology, the URI form, and which application identifiers travel are all your decisions. Made once, in your artwork cycle, not defaulted by a co-man and not dictated by a retailer.
The four decisions that are yours — and one that no artwork brief contains
As GTIN owner approving the artwork, four choices are on your desk whether you make them deliberately or by default:
- Symbology. A QR (phone-readable, can double as a consumer touchpoint) versus a GS1 DataMatrix (lane- and pharma-oriented, not reliably read by a phone camera). Decide this in a pure compliance frame and you will likely default to a mark no shopper can ever read.
- URI form. A GS1 Digital Link URI (web-resolvable — the identity written as a real address) versus GS1 element strings (a lane token that resolves to nothing in a browser). Only the URI form can ever be a touchpoint.
- Which application identifiers travel. GTIN is the Sunrise baseline. Lot and expiry are optional AIs you may add when the custody case or the freshness case justifies them — your call, not a mandate.
- Where the events behind the mark get recorded, and by whom. This is the one no artwork brief has ever contained, and the one that decides whether the mark is a dead square or a signal instrument — the difference between "we printed a compliant barcode" and "we can answer where this lot went and who observed it."
The first three are made inside the artwork cycle and then locked into a print run for the life of the pack. The fourth sits just outside the brief, which is exactly why it gets orphaned — no one on the artwork call owns it, so no one makes it.
Why "let the co-man handle it" is the expensive default
Because the co-man will handle it — the way a co-man handles anything that lands in their lap without an instruction: with their default, chosen to suit their line, not your customer or your custody case. That default is very often a lane-only mark, because from the plant's point of view the barcode's only job is to clear a scanner. You will have delegated a brand decision to a vendor optimizing for a different objective, and you will discover the result when it's printed on eighteen months of inventory.
The same boundary shows up on the custody side, and it is worth seeing because it's the same ownership fact viewed from the other end. When a lot goes wrong at a co-packer you don't own, "which stores, which shelves, on whose authority" becomes a phone-and-email reconstruction across a boundary where the operational data is contractually someone else's. You own the identity of the product; you do not, by default, own the record of what happened to it in a facility you don't run. The barcode decision and the custody-record decision are the same ownership question — what stays yours when someone else does the physical work — and the artwork cycle is the one cheap, recurring window where you get to answer it on purpose.
The window: your artwork cycle, and it's rarer than you think
Artwork changes are batched, scheduled, and expensive to reopen — which is precisely why the Sunrise mark arriving as "just a design change" is a trap. The next time your artwork opens, four decisions that will hold for years get made in the span of one approval call. Make them deliberately, as the GTIN owner you are, and the square you were going to print anyway becomes an asset you control. Let them default, and you have handed a brand decision to a factory and a retailer and printed the result a million times.
You don't need permission from your co-man to decide any of this. You need to decide it before the artwork closes. The full frame — what the mark can become, and the honest limit on what's built — is the parent pillar: Your barcode was a checkout token. The same square can be a line to your customer.
Put me on the seat list → /first-trace?segment=manufacturer
Pick "Manufacturer or brand — you pack it out and you own the GTIN." Your capture workspace is provisioned from this list, in order — one email when your seat is ready. The record layer's translate, validate and hash doors are live at epcis.dev. The named human sender and the verified sending address are a launch gate — P0-V4 in the open P0 ledger, stated plainly. The deeper door is the live get-started interview.