GS1 Sunrise 2027: the retailer's field guide to the 2D barcode transition
By the end of 2027, retail points of sale are expected to read 2D barcodes alongside the linear symbols they have scanned for fifty years. Most of what has been written about that sentence is wrong in one of two directions.
It is either described as a mandate with a switch-off date, which it is not, or dismissed as a packaging change, which badly understates it. This site is the execution layer: what the transition actually requires of a point of sale, and how to find out whether your lanes can do it.
What Sunrise 2027 actually is
The formal name is Ambition 2027. In the United States it is usually called Sunrise 2027. It is a voluntary, industry-led effort coordinated through GS1. It is not a regulation, not legislation, and not a mandate enforced by any authority. No date has been announced for switching off linear barcodes, and the GS1 implementation guideline states that linear barcodes will coexist with 2D barcodes for as long as there are uses for them.
The word sunrise is doing useful work. A sunrise is a capability coming online, not a switch being thrown. Reading it as a deadline produces two opposite mistakes. One is panic, which turns a staged program into a rush to replace scanners that only needed reconfiguring. The other is complacency, which ignores the dual-marked packs already arriving in stores, carrying both symbols, either of which a lane may read first.
Voluntary does not mean irrelevant. You still have to ring up the products your suppliers ship. Your customer cannot be asked to care whether a failed transaction came from optics, a disabled setting, an unsupported syntax, or the POS interface. The lane either sells the item or it does not.
"A sunrise is a capability coming online, not a switch being thrown."
Sunrise 2027, Chapter 1
The one sentence that sets the floor
A point of sale must identify, decode, beep once, and transmit one GTIN from any GS1-conformant retail barcode, linear or 2D.
That is the minimum, and it is deliberately narrow. It does not require your POS to store or act on batch, expiry, or serial data. It does require that a dual-marked pack produces one beep and one receipt line rather than two, because two is a customer-facing defect. And it requires that what arrives at the register is a usable product identity rather than a raw URL-shaped string.
A lane that beeps but hands the application nothing usable has not met the threshold. A retailer counting scanner replacements instead of completed transactions will believe it is further along than it is.
Five carriers, not one test
A conformant retail point of sale has to handle five carrier and syntax combinations:
-
EAN/UPC with plain syntax
-
GS1 DataBar with GS1 element string syntax
-
GS1 DataMatrix with GS1 element string syntax
-
Data Matrix carrying a GS1 Digital Link URI
-
QR Code carrying a GS1 Digital Link URI
Read that as five separate test cases. The distinction most often missed sits inside the Data Matrix symbol: GS1 DataMatrix and a Data Matrix carrying a Digital Link URI can be the same physical shape. What differs is the data inside.
Two symbols that look identical to a person can demand different interpretation from scanner and POS software, and that is what breaks projects after a demonstration that appeared to succeed. A checklist that says only "2D enabled" has not described the requirement.
How ready are you, actually
The book includes a ten-domain readiness assessment with fifty testable criteria. Four of the ten domains are hard gates, and they carry 55 of the 100 points, because together they decide whether a transaction completes at all:
-
Scanner hardware inventory and capability, 15 points
-
Scanner configuration and syntax handling, 15 points
-
POS application readiness, 15 points
-
Duplicate-scan and dual-marking handling, 10 points
Each gate must reach 80 percent on its own. The threshold never applies to the four combined, so a strong gate cannot carry a weak one, and a high overall score with a failed gate is not readiness. The remaining six domains cover data element policy, storage and governance, downstream systems, private label labeling, supplier master data, and associate training.
The evidence standard is the rule that keeps the exercise honest: a claim without a dated artifact does not score. And your own team scores it, not a vendor, because a vendor scoring your readiness is scoring its own delivery.
This assessment is the author's instrument. It is not a GS1 standard, not a certification, and no score it produces carries standing with GS1 or any GS1 member organization.
About the book
Sunrise 2027 is a 214-page working field guide for the person who has to deliver the change: the store systems director, the retail IT lead, the supply chain manager who has to brief an executive, defend a budget, and hold vendors accountable. It covers the requirement, the standards without the standards document, the ten-domain assessment, the specific ways these projects fail in practice, vendor questions written so they must be answered with a version number or a date, and a ninety-day plan.
Every technical claim is cited with its source and its publication date, because guidance in this field moves and a claim without a date is worthless to you. All 74 sources are listed and maintained on this site. Where the industry has no settled answer, the book says so plainly instead of filling the gap with confidence.
By Michael Talbot. Available August 2026 in paperback and Kindle.
This is an independent publication. It is not affiliated with, endorsed by, or sponsored by GS1 or any GS1 member organization. It recommends no vendor and sells no service.