1D vs 2D Barcodes — Why QR and DataMatrix Are Taking Over, and GS1 Sunrise 2027

TL;DR

1D barcodes encode data in parallel bars along a single axis — fast to scan, limited to ~20–25 characters. 2D matrix codes use both axes, storing hundreds of characters with built-in Reed-Solomon error correction. QR Code (ISO/IEC 18004, royalty-free since Denso Wave's release) dominates consumer and retail use. DataMatrix (ISO/IEC 16022) is the mandated carrier for pharmaceutical serialization in Korea under MFDS regulation. GS1 Sunrise 2027 requires retail POS systems worldwide to process 2D formats alongside EAN/UPC by 2027.

1D vs 2D — Data Density and Damage Recovery

The fundamental difference between 1D and 2D barcodes is the axis along which information is encoded.

A 1D barcode encodes data in the width and spacing of parallel vertical bars scanned along a single horizontal axis. The maximum practical payload for common 1D symbologies — Code 128, EAN-13, ITF-14 — sits between 20 and 48 characters, governed by the physical width of the label and the printing resolution available. Increasing payload length means increasing label width linearly. This constraint is not a defect; for a retail shelf edge or a shipping carton, 13 digits (EAN-13) is sufficient to carry a GTIN that resolves to a complete product record in a database lookup.

2D codes divide the encoding task across both horizontal and vertical axes. A QR Code version 40 symbol (177×177 modules) can hold up to 7,089 numeric characters or 4,296 alphanumeric characters within a single printed square. DataMatrix in its largest configuration (144×144 modules) reaches 3,116 numeric or 2,335 alphanumeric characters. In practice, most production use cases operate at a fraction of maximum capacity: a GS1 Application Identifier string for pharmaceutical serialization — GTIN plus expiry plus batch plus serial — typically runs 40–60 characters, well within a 16×16 or 24×24 DataMatrix symbol that prints legibly at 5 mm across.

Error Correction: Why 2D Codes Are Resilient

The practical advantage of 2D codes over 1D in industrial environments is error correction, not raw data capacity.

QR Code implements Reed-Solomon error correction at four levels:

Level	Recovery Capacity	Common Use
L (Low)	Up to 7% of codewords	High-density printing, clean environments
M (Medium)	Up to 15% of codewords	General retail, marketing
Q (Quartile)	Up to 25% of codewords	Light industrial, outdoor labels
H (High)	Up to 30% of codewords	Heavy industrial, pharma, cold chain

DataMatrix implements its own Reed-Solomon scheme with recovery capacity roughly comparable to QR Level M across most standard sizes. The exact recovery capability depends on the version (ECC 200 is the current standard) and symbol size.

A 1D barcode has no inherent error correction — a single bar printed too wide or too narrow, or a scratch across the label, can cause a misread or a no-read. Most 1D symbologies use a single checksum digit to detect errors, but a failed checksum produces a rejection, not a recovery. This makes surface condition and print quality critical for 1D in ways that are less severe for 2D codes with high error correction levels.

In Korean cold chain and outdoor logistics contexts — where label condensation, label adhesion failure, and physical abrasion are real operational factors — this distinction drives meaningful differences in scan reliability between 1D and 2D deployments.

Physical Constraints

The minimum reliable print size for a 1D barcode on consumer packaging using standard EAN-13 is approximately 37.29 mm wide × 25.93 mm tall at 100% magnification (GS1 specification). 2D codes have no such minimum footprint constraint from the symbology standard itself — the constraint is imposed by the scanner's optics and resolution. A 10×10 mm DataMatrix symbol, printed at adequate resolution, is decodable by a smartphone camera or an industrial area-scan imager at close range.

This footprint advantage matters directly for Korean pharmaceutical blister packs and ampoules, where available label area is often less than 2 cm×, and for electronics components with serialized labels applied to individual parts on an assembly line.

Why QR and DataMatrix Are Gaining Traction

QR Code: Open Standard, Global Adoption

QR Code was invented in 1994 by Masahiro Hara at Denso Wave, a Toyota supplier, to track automotive components during manufacturing. The design objective was a code that could be scanned at high speed from any direction — the three finder patterns at three corners of the symbol enable orientation detection without requiring the scanner to align with the label.

The turning point for QR adoption was Denso Wave's decision to publish the QR Code standard as ISO/IEC 18004 and to release it royalty-free. Any manufacturer, app developer, or printer can generate and read QR codes without licensing fees. This contrasts with some earlier 2D symbologies that carried proprietary restrictions. The open licensing decision removed the economic barrier that had slowed adoption of competing 2D formats.

Consumer smartphone cameras integrated native QR decoding — without requiring a separate app — beginning with iOS 11 (2017) and Android equivalents shortly after. This created a scanning infrastructure of several billion devices worldwide, which in turn drove the QR Code's adoption in payment, marketing, loyalty, and authentication use cases that go far beyond its industrial origins. GS1 Digital Link, the standard that encodes a product's GTIN into a QR Code URL, is the mechanism by which QR Code is entering the retail supply chain as a primary barcode carrier under the Sunrise 2027 initiative.

DataMatrix: Precision Engineering for Regulated Environments

DataMatrix was developed independently from QR Code, with an early commercial version published by RVSI Acuity CiMatrix in the late 1980s. The AIM (Association for Automatic Identification and Mobility) adopted it as a standard, which later became ISO/IEC 16022. Unlike QR Code, DataMatrix carries no broad consumer recognition — it is a working code for industrial and regulated environments.

Two properties make DataMatrix the engineering choice for pharmaceutical and components serialization. First, its encoding density at small symbol sizes is superior: a 10×10 module DataMatrix can encode 6 numeric digits, sufficient for a short serial number or a product-specific identifier, in a symbol small enough to print on a 2 mm component. QR Code's minimum symbol version (21×21 modules) encodes slightly less data per unit area than DataMatrix at comparable symbol sizes. Second, the regulatory community — particularly the health care supply chain — converged on DataMatrix before QR reached widespread awareness, and regulatory mandates rarely change their technical specifications once established.

The GS1 Application Identifier (AI) structure is symbology-independent — the same AI payload (GTIN, expiry date, batch, serial) can be encoded in any compliant 2D carrier — but in practice, pharmaceutical serialization regulations in Korea, Europe, and the United States have standardized on DataMatrix as the physical carrier. Switching the carrier symbology would require re-validation of all printing, verification, and scanning equipment already qualified under existing regulatory filings.

Royalty and Cost Comparison

For a new deployment today, the IP cost comparison is straightforward: both QR Code (ISO/IEC 18004) and DataMatrix ECC 200 (ISO/IEC 16022) are fully open standards with no royalty obligations. A system integrator or in-house development team can generate and decode either format without licensing costs at the symbology level.

Cost differences between QR and DataMatrix deployments arise from implementation context, not IP:

• Label printing qualification: Pharmaceutical DataMatrix labels must meet ISO/IEC 15415 print quality grades that require calibrated verifier equipment. QR for consumer applications rarely demands this level of verification.
• Scanner hardware: Fixed laser scanners cannot read 2D codes; 2D codes require area-scan imagers or camera-based readers. For operations migrating from 1D fixed scanners to 2D, hardware replacement is the dominant cost line.
• System integration: DataMatrix in pharma carries mandatory GS1 AI parsing requirements. Any backend system that receives scanned data must parse the parenthetical AI structure correctly — (01), (17), (10), (21) — which may require middleware development if the existing system passes raw barcode strings without parsing.

For the Scandit SDK specifically, enabling or disabling a symbology is a single API call — see the iOS code sample above and the Scandit Docs — Configure Barcode Symbologies (iOS) for the full SymbologySettings surface. The SDK handles both QR and DataMatrix natively; the integration cost is in the surrounding workflow, not the scanning layer.

GS1 Sunrise 2027 — Korea Market Response

The Global Mandate

GS1 Sunrise 2027 is a coordinated industry commitment, not a single law, by which retailers worldwide agree to update their point-of-sale systems to read and process 2D barcodes — primarily QR Code carrying GS1 Digital Link URLs — by 2027. The initiative addresses a structural limitation of the current EAN/UPC regime: a barcode printed on a consumer product today can encode only the GTIN, which provides no space for batch number, expiry date, country of origin, or other attributes the supply chain increasingly needs at the retail scan point.

A GS1 Digital Link QR Code on the same label encodes all of that within a standard URL structure:

https://id.gs1.org/01/{gtin}/10/{batch}/17/{expiry}

The retailer's point-of-sale system resolves this URL either against a local database or against the GS1-compliant web resolver, returning product information without any change to the labeling data model. The 2D code is backward-compatible in the sense that the GTIN is still present — but legacy scanners that decode only Code 128 or EAN-13 and ignore QR codes will be unable to process the richer payload.

For operators running Scandit SDK integrations at retail POS, the technical implication is that symbology configuration must include Symbology.qr in the active set. Any integration that hard-codes only ean13UPCA or code128 as active symbologies will require reconfiguration before Sunrise 2027 labels enter the store's incoming goods stream. Reviewing your active symbology list now — before suppliers begin transitioning label stock — avoids an emergency configuration change at a time when production volumes are high.

Korean Pharma Serialization — MFDS Mandate

Korea's pharmaceutical serialization requirement operates on a separate regulatory track from GS1 Sunrise 2027, but both converge on the same conclusion: 2D barcodes are mandatory.

Korea's Ministry of Food and Drug Safety (식품의약품안전처, MFDS) requires unit-level serialization of prescription pharmaceuticals under the Drug Serialization Management System (의약품 일련번호 관리). The mandated format is GS1 DataMatrix encoding at minimum the following Application Identifiers:

• (01) — GTIN (14 digits)
• (17) — Expiry date (YYMMDD)
• (10) — Batch/lot number
• (21) — Serial number (unique per unit)

This is not a recommendation; it is a condition of market authorization. Pharmaceutical manufacturers and importers operating in Korea must print, verify, and report this data to the MFDS serialization database. The scanning and verification systems at distribution, pharmacy, and hospital dispensing points must decode GS1 DataMatrix and parse the AI structure correctly.

For Korean pharmaceutical clients deploying Scandit SDK scanning at dispensing or verification points, the filtering requirements are strict: the active symbology set must include DataMatrix but can typically exclude QR Code (secondary packaging rarely carries QR in this regulated context), and the barcode filter should constrain accepted codes to the expected AI pattern to avoid misreads from logistics labels that may also carry DataMatrix codes with different payload structures.

GS1 Korea maintains a certification program for scanning equipment and software used in the MFDS serialization chain; Data Connect engineers can assist with the testing protocol as part of a Scandit SDK integration engagement.

Food Traceability — Korean Regulatory Trajectory

Korea's food traceability framework, administered by the Ministry of Agriculture, Food and Rural Affairs (농림축산식품부, MAFRA) and the Ministry of Food and Drug Safety, is expanding the scope of products requiring full supply chain traceability records.

While the current traceability mandate applies to a defined list of product categories — processed food, imported items under specific HS codes, and certain agricultural products — the regulatory trajectory points toward broader 2D barcode adoption on food packaging, driven by the same data enrichment logic underlying GS1 Sunrise 2027. Major Korean food producers preparing for major retail chain shelf compliance are already conducting qualification trials with GS1 DataMatrix and GS1 Digital Link QR codes on primary packaging.

The practical planning horizon for food producers is: pilot 2D barcode integration on new product lines or label redesigns now, so that by 2026 the printing, verification, and reporting infrastructure is production-qualified ahead of the Sunrise 2027 retail scanner cutover.

Retail POS Migration Timeline — Korean Context

Korean general retail is coordinating its POS scanner upgrade program through GS1 Korea's Sunrise 2027 working group. The typical migration path involves:

1. Hardware audit (2024–2025): Identify fixed omnidirectional laser scanners (unable to read 2D) and schedule replacement with area-scan imagers or 2D camera-based units.
2. Software update (2025–2026): Update POS middleware to parse GS1 Digital Link URLs and map the extracted GTIN and AI fields to the item master database.
3. Supplier qualification (2026): Test incoming goods with 2D primary labels against the updated scanning and database system.
4. Full operation (2027): Retail scanners process 2D-primary labels without operator intervention.

For Scandit SDK deployments within this migration — handheld assisted scanning at receiving docks, inventory counting, or mobile POS applications — the Sunrise 2027 readiness check is a symbology configuration audit, not a hardware replacement. The SDK's camera-based scanning is already 2D-capable; the configuration action is confirming that QR (and DataMatrix for pharma-adjacent operations) is in the active symbology set. See GS1 — 2D Barcodes & GS1 Standards for the authoritative technical specifications.

DataMatrix vs QR — Decision Guide

The question of which 2D format to deploy is not always a free choice — in regulated industries, the standard is specified. Where choice exists, the decision reduces to a small number of criteria:

When DataMatrix Is the Right Answer

• Pharmaceutical serialization in Korea: MFDS regulation mandates GS1 DataMatrix. No alternatives are compliant.
• Tiny label area: DataMatrix achieves higher data density at small symbol sizes. Labels under 5 mm × 5 mm where DataMatrix can encode a viable payload but QR's minimum symbol version cannot.
• Direct part marking: Laser-etched or dot-peen DataMatrix on metal components for aerospace, automotive, and electronics assembly. The industrial standards for direct part marking (ISO/IEC 29158, previously AIM DPM) reference DataMatrix as the primary symbology.
• Supply chain where GS1 DataMatrix is already the supplier standard: If your trading partners are already printing DataMatrix, introducing QR creates a dual-symbology scanning requirement with no corresponding benefit.

When QR Code Is the Right Answer

• Consumer-facing scan triggers: Any application where an end user is expected to scan with a smartphone camera. QR Code's consumer recognition and native OS support make it the default for marketing, loyalty, payment, and authentication use cases.
• GS1 Digital Link retail labeling: The Sunrise 2027 initiative is built around QR Code as the GS1 Digital Link carrier for general merchandise. Food and non-pharma consumer goods producers transitioning to 2D under Sunrise 2027 should use QR.
• High-volume printing where label verification is not mandated: QR Code's error correction at Level H provides robust recovery for labels printed under variable conditions without requiring the ISO/IEC 15415 verification process that pharmaceutical DataMatrix labels must pass.
• Composite codes with EAN/UPC: QR Code's GS1 Digital Link structure provides a clean upgrade path from EAN-13 because the GTIN is embedded in the URL. QR replaces — rather than coexists alongside — the EAN-13 for Sunrise 2027 compliant labels.

Comparison Summary

Criterion	DataMatrix	QR Code
Open standard	ISO/IEC 16022 (AIM)	ISO/IEC 18004 (Denso Wave)
Royalty	None	None
Max data capacity	~3,100 numeric chars	~7,000 numeric chars
Minimum practical symbol	~2 mm × 2 mm	~10 mm × 10 mm (version 1)
Error correction	Reed-Solomon (~15–25%)	Reed-Solomon (7–30%, levels L/M/Q/H)
Korean pharma mandate	Required (MFDS)	Not applicable
GS1 Sunrise 2027 primary	Not primary carrier	Primary carrier (GS1 Digital Link)
Consumer smartphone scan	Not widely recognized	Native OS support (iOS 11+, Android)
Direct part marking	ISO/IEC 29158 standard	Uncommon

Is 1D Really Going Away?

No, and it is worth being precise about what is and is not changing.

EAN-13, Code 128, ITF-14, and Code 39 will continue to operate in supply chains at scale for the foreseeable future. The economics and physics of high-throughput fixed scanning at conveyor speeds — where a laser scanner reads 1,000 codes per minute across a belt — do not favor replacement with area-scan camera systems in every case. The installed base of industrial scanning equipment is measured in millions of units globally; depreciation cycles alone mean a full replacement cannot happen on a five-year timeline.

What is changing is the assumption that 1D coverage is sufficient for a new integration. A software integration designed in 2025 that enables only Code 128 or EAN-13 symbologies will encounter operational gaps before 2028 — when 2D-primary labels from Sunrise 2027-compliant suppliers begin appearing in meaningful volumes. The cost of a symbology configuration update is low; the cost of an emergency POS or receiving system update under supply chain pressure is high. Planning for both families now is the lower-risk path.

In Korean logistics specifically, the coexistence of 1D and 2D is already the norm in several operational contexts. Major Korean logistics operators and e-commerce fulfillment providers use Code 128 waybills at high volumes while simultaneously processing GS1 DataMatrix pharmaceutical cartons and QR-coded returns labels. The scanner hardware and software in those environments must handle the full range — and it does, with Scandit SDK configurations that enable multiple symbology families in parallel.

The practical timeline for 1D's operational sunset in Korean retail — if such a transition ever completes — is beyond 2035. For industrial and logistics use cases, it is longer. The business case for replacing functional 1D infrastructure with 2D before the infrastructure reaches end of life is limited unless the new labels carry data that 1D cannot encode (serialization, expiry, batch) — which is precisely the scenario that GS1 Sunrise 2027 and MFDS serialization are creating.

Korean Market Case Studies

A Major Korean Logistics Carrier — Waybill and Parcel Scanning

A major Korean logistics carrier processes tens of millions of parcels annually through its national distribution network. The primary tracking code on waybills is a Code 128 barcode encoding a shipment identifier — a well-established 1D deployment optimized for fixed laser scanners at sort gates.

The complexity arises at the intersection of parcel logistics and pharmaceutical distribution, where this carrier also handles cold-chain pharmaceutical shipments subject to MFDS serialization. A single sortation line may process both consumer parcels with Code 128 waybills and pharmaceutical cartons carrying GS1 DataMatrix unit serialization labels. The scanning system must distinguish between these without operator intervention — which requires a symbology filtering configuration that reads both Code 128 and DataMatrix, uses the payload structure to determine which label type is present, and routes the data to the appropriate downstream system. This is a concrete example of the mixed-symbology filtering scenarios described in Data Connect's filtering guide.

For mobile scanning deployments at carrier pickup points, the Scandit SDK configuration enables Code 128 for waybill scanning while keeping DataMatrix active for pharmaceutical exception handling. The active symbol count configuration for Code 128 is constrained to the expected waybill length range to suppress false reads from secondary packaging barcodes on the adjacent product.

A Korean Fulfillment Operator — Fulfillment and Returns

A Korean e-commerce fulfillment operator's centers operate at some of the highest scan densities in Korean logistics. A single camera frame in a high-throughput picking aisle may contain multiple product barcodes — EAN-13 on consumer items, Code 128 on logistics units, and increasingly QR codes on marketplace seller labels. The fulfillment scanning configuration requires symbology prioritization rather than exclusive selection: all three families must be active, but the result disambiguation must correctly identify which code is the intended target for the pick operation.

Returns processing at such operations presents an additional scenario. Returns labels are QR codes encoding a structured URL with order ID and return reason. At the returns intake station, the scanner must process the QR returns label while suppressing the product barcode on the item underneath. This is a post-decode filtering problem rather than a pre-scan symbology problem: both QR and EAN-13 must be active (the operator may need to scan the product barcode for condition assessment after processing the returns label), but the primary capture target must be identified by payload structure.

These operational requirements illustrate why the choice of barcode format cannot be made independently of the scanning environment design. Format selection, scanner configuration, and filtering logic are interrelated.

A Major Korean Retail Chain — Private Brand Traceability

A major Korean general merchandise retailer applies GS1-compliant barcodes to its private brand product line under its own GS1 company prefix. PB products have historically carried EAN-13 at point of sale and Code 128 on logistics units, with no serialization requirement for most SKUs.

The Sunrise 2027 migration creates a specific question for retailers with PB product lines: when does the EAN-13 on the consumer unit get supplemented or replaced with a QR Code carrying GS1 Digital Link? The answer depends on whether retail POS scanners are updated before the label revision cycle for each PB SKU. Major Korean retailers have indicated public participation in GS1 Korea's Sunrise 2027 qualification program, which implies POS scanner updates before 2027. For the PB label design team, the planning trigger is: any PB product with a label revision scheduled for 2026 or later should include the GS1 Digital Link QR Code alongside the EAN-13, so the label is Sunrise 2027 compliant from the point of printing.

For suppliers providing goods to retail chains under private label agreements, the same logic applies. The brand owner's responsibility under the GS1 Digital Link model is to register the GTIN in the GS1 registry and maintain the resolver record — the label itself carries only the URL structure.

MFDS Pharma Serialization — GS1 DataMatrix at Dispensing Points

The most technically precise Korean use case for GS1 DataMatrix is the pharmacy and hospital dispensing point under MFDS serialization. When a dispensing pharmacist scans a prescription drug package, the scanning system must:

1. Decode the GS1 DataMatrix symbol on the unit package
2. Parse the GS1 Application Identifier fields: GTIN (AI 01), expiry (AI 17), batch (AI 10), serial (AI 21)
3. Query the MFDS serialization database to verify that the serial number is authentic and not previously dispensed
4. Record the dispense event against the serial number

This workflow requires that the Scandit SDK integration passes the full raw DataMatrix payload — including AI delimiters — to the application layer without truncation. The backend parsing logic must handle the GS1 FNC1 character that separates variable-length AI fields. Any integration that strips non-alphanumeric characters from the scan result before passing it to the backend will corrupt the AI structure.

For integrators building MFDS-compliant dispensing systems with the Scandit SDK, Data Connect provides a proof-of-concept environment with a test DataMatrix set covering the full range of MFDS-required AI combinations. Contact Data Connect for access as part of a project engagement.

Last Updated

Last updated: 2026-05-01

For implementation support, a proof-of-concept review, or integration guidance for GS1 Sunrise 2027 readiness or MFDS pharma serialization in Korea, contact Data Connect.

// Enable QR and DataMatrix while disabling legacy 1D symbologies
// Consult Scandit Docs — Configure Barcode Symbologies (iOS) for the full API surface
let settings = BarcodeCaptureSettings()
settings.set(symbology: .qr, enabled: true)
settings.set(symbology: .dataMatrix, enabled: true)
settings.set(symbology: .ean13UPCA, enabled: false)
settings.set(symbology: .code128, enabled: false)