For the VP of operations and supply chain at a co-manufactured mid-market CPG — the one seat that holds everything, where co-man mandates and retailer requests both land, and where a questionnaire with a date is a familiar species of Tuesday.
There is a document making its way toward your inbox. It is a spreadsheet or a portal form, it is from one of your retailers, it has "2D" or "Sunrise" in the subject line, and it has a response date. If it has already arrived, this brief is your answer key. If it has not, you have the rare thing your seat almost never gets: advance notice — because this particular questionnaire is structurally predictable, question by question, from public information.
Why the questionnaire is inevitable — and derivable
GS1 Sunrise 2027 places its expectation on the lane: retail point-of-sale reads and processes 2D barcodes and extracts the GTIN by end of December 2027. It is GS1's programme, an industry commitment rather than a law, and it obliges you — the supplier — to precisely nothing.
But a retailer cannot certify a lane against marks it has never seen. Every capability the lane must prove pushes a question uphill to the parties who control what arrives at it: the lane's obligations flow uphill as suppliers' questionnaires. The retailer must test decode → so it asks what symbology you print. It must parse both payload syntaxes → so it asks which form your marks carry. It must handle marks with additional data → so it asks whether you encode lot and expiry. It must schedule certification against real packs → so it asks for your timeline. That derivation is the whole trick: you can reconstruct the questionnaire from the lane's own to-do list, before any retailer writes it. The grocery side's obligations are public; here they are, flowed uphill.
The model questionnaire, with the capability that answers each line
| The question they will ask | Why the lane needs it | The answer that closes the line |
|---|---|---|
| "Which 2D symbology do your consumer units carry, and when?" | Decode certification per symbology | A named choice — QR for consumer-facing packs, and the choice is yours as GTIN owner, not the retailer's |
| "Do your marks carry element strings or a GS1 Digital Link URI?" | The POS parser handles both syntaxes; certification wants to know which is coming | The URI form, on your domain, if the consumer touchpoint matters to you at all — the two payloads and what each can do |
| "Will you encode lot and expiry (AIs 10 and 17)?" | Lanes and receiving get new behavior only if the data arrives | Your explicit choice, stated as one — lot/expiry are optional AIs, the brand's call, never Sunrise's requirement |
| "Will the 1D UPC remain during transition?" | Dual-marking keeps uncertified lanes ringing | Yes, dual-marked, until the retailer confirms 2D-only readiness in writing |
| "What is your transition timeline by SKU family?" | Certification scheduling | Your artwork calendar, read out loud — which is why the artwork cycle is the decision window |
| "Who is your 2D technical contact?" | Someone must answer follow-ups | You — or the capability partner answering from your record |
Six lines. Nothing in them is exotic, and that is the point: every answer is a decision, not a project. A brand that has made the decisions answers in an afternoon. A brand that has not, forwards the questionnaire to its co-manufacturer — who executes marks but does not choose them — and gets back the correct and useless reply that it is the brand's call.
Answering from capability vs. answering per retailer
Now the part that is specifically about your seat. You will not get one questionnaire; you will get one per retailer, on different dates, in different formats, asking these same six things with different column headers. The per-retailer strategy — scramble, answer, file, repeat — costs a working week per retailer and produces answers that drift out of sync with each other, which a future questionnaire will catch.
The one-seat survival strategy is to answer from capability, once: a standing one-page 2D position — symbology, URI form, AI choices, dual-mark policy, timeline by SKU family, technical contact — written before the first questionnaire lands, from which every retailer's form is a transcription exercise. The six decisions above are that page; this brief is most of its first draft. The same page answers your co-manufacturer's spec-sheet questions and your next private-label conversation, because they are all downstream of the same six decisions.
This is your own procurement rule — buy outcomes, refuse projects — applied to 2D readiness. You have no engineering function to run a "2D programme," and you do not need one: the readiness the questionnaires probe is a set of decisions plus a record behind them. The decisions cost meetings. The record — where what your marks make possible actually lands, as conformant EPCIS 2.0 events with an attested who on every capture — is an outcome you buy, not a system you staff; the stateless doors of the engine behind it are live at epcis.dev, and every open gap on our side is a named entry in the open P0 ledger, where we tell you the truth about it.
What to have in writing before the first one lands
- The one-page 2D position — the six answers, dated, versioned, owned by your seat.
- The artwork-cycle map — which SKU families refresh when, because every timeline answer is really this map read aloud.
- The co-man addendum — one paragraph to each co-manufacturer stating the marks your specs will require, so their line answer is ready when a retailer's follow-up needs it.
- The record decision — where scans and events land, decided once, so "who is your technical contact" has an answer that survives an audit.
The questionnaire is predictable; being ready for it is a decision-making exercise, not a project. The written read for your situation — your retailer mix, your artwork calendar, your co-man arrangements, and which of the six decisions is actually urgent for you — starts at get started: your address first under the one-message promise, then questions that branch on your answers, ending in a read that locks when it ends.