A customer buys a watch in a Geneva boutique, opens two newsletters and attends an event. Eighteen months later the watch is resold, then listed in a market where the brand has no dealer. The CRM records none of it. Product scan data does. Each tap on a product’s digital identity records the unit, the time, the approximate location and, with secure NFC, a verified authenticity result. A CRM describes people who identify themselves. Scans describe what happens to products.
A CRM is organised around a contact: an email address, a purchase, a consent, a campaign response. It knows the object only as a line on an order. Once the product leaves the point of sale, the brand loses sight of it.
Five drivers open that gap:
The counterfeit gap alone is large. The OECD and EUIPO estimate counterfeit trade at USD 467 billion in 2021, up to 2.3% of global trade, and USD 117 billion of EU imports, or 4.7% of the total (OECD/EUIPO, Mapping Global Trade in Fakes 2025). Clothing, footwear and leather goods account for 62% of seized counterfeits. Diverted genuine goods are not in those figures at all.
This is not a flaw in the CRM. It was built to describe people, and it does that well. The risk sits elsewhere, with the products.
Product scan data is the record created each time a person or a system reads the digital identity of one specific physical item. That identity is usually a secure NFC tag, a UHF RFID tag or a serialised printed code. It is built on a GTIN plus a unique serial number, typically expressed as a GS1 Digital Link.
Each scan is an event, not a contact. It carries a small, consistent set of facts.
| Field | Example | Used for |
|---|---|---|
| Unit | This specific bottle, bag or watch | Tying every event to one item’s history |
| Time | Date and time of the tap | Sequencing, first scans, scan frequency |
| Approximate location | Country or city | Seeing where products actually end up |
| Context | Phone type, language | Experience design, market insight |
| Authenticity verdict | Genuine, failed, replayed (secure NFC only) | Counterfeit and clone detection |
| Tamper status | Sealed or opened (tamper-evident tags only) | Refill and first-open evidence |
The distinction that matters: declared data tells a brand who says they own something. Produced data tells a brand what the something is doing.
CRM data is declared. It is rich, but limited to people who choose to identify themselves, and only as accurate as what they enter. Scan data is produced. It is narrower, but it covers every unit that is read, whoever holds it. With a secure carrier it is verified, not declared.
The chip class used for phone-tap authentication is NXP’s NTAG 424 DNA (ISO/IEC 14443-A, NFC Forum Type 4), with AES-128 keys and SUN (Secure Unique NFC) messaging. The process runs from the production line to the tap.
Strictly, the server does not send the chip a challenge. The chip signs a fresh counter on its own and the server verifies it. For the brand the effect is the same: every tap carries proof that only the genuine chip could produce.
A basic tag is a door with the key hanging beside it. Anyone can copy the key. A secure tag is a door that generates a new one-time code each time someone approaches, and only the door knows how to make the right one.
That is what turns a scan into evidence. A web address printed on a pack can be photographed and reused. A SUN message cannot.
The table below treats a serialised QR code as a printed code with a unique serial and no copy-detection feature.
| Question a brand asks | CRM record | Serialised QR scan | Secure NFC scan |
|---|---|---|---|
| Is this specific item genuine? | No | Weak: the code can be copied | Yes, cryptographically, on every tap |
| Where is this unit now? | Only if the owner declares it | Approximately, when scanned | Approximately, when tapped |
| Was it scanned outside its intended market? | No | Yes, if scanned | Yes, with an authenticity verdict |
| Has the seal been opened? | No | No | Yes, with a tamper-evident closure |
| Who holds it after resale? | No | Only if the buyer registers | Yes, through ownership transfer |
| Who is the buyer? | Yes, if registered | Only if they opt in | Only if they opt in |
The last row is the honest one. The CRM remains the system of record for identified people, and scans do not replace it. What scans add is a layer of facts about units that arrives without anyone filling in a form. Four signals stand out.
Authenticity signals. A failed or replayed verification is a precise alert: this tag did not come from the brand’s production, or it was copied. A cluster of failures in one region points to a counterfeit operation.
Geography of real use. Each tap can be compared with the market the unit was shipped to. Diversion then arrives as a lead with a serial number attached, traceable to a shipment. How grey tracking turns verification scans into a diversion map shows that analysis in detail.
Lifecycle events. Repair, resale, gifting and ownership changes each leave a verified event on the item’s record. A second-hand buyer’s verification becomes a data point no CRM would hold.
Silent engagement. A loyal customer who never fills in a survey still taps to check a refill or confirm authenticity. Scans show real use and replenishment patterns, not only stated preferences.
A QR code carries an identifier. A secure NFC chip carries an identifier and proves it is the original.
| Serialised printed QR | Copy-resistant print (ACM-secured barcode, Diamond Dust) | Secure NFC (NTAG 424 DNA class) | |
|---|---|---|---|
| Tracking and engagement | Yes | Yes | Yes |
| Proves authenticity | No: a photo scans the same | Yes: the print is hard to reproduce | Yes: new signed code on every tap |
| Tamper evidence | No | No | Yes, with TagTamper |
| Marginal cost | Near zero; the real cost is serialisation and platform | Print plus security feature | USD 0.45 to 0.65 per label at five-figure volumes (public catalogues, 2026) |
| Read by | Any phone camera | Phone camera, depending on feature | Any recent phone, no app |
For the Digital Product Passport, the ESPR (Regulation (EU) 2024/1781) is carrier-neutral. In practice a GS1 Digital Link QR is emerging as the default visible carrier. Many premium programmes put both on the same product, resolving to the same identity: NFC for trust, a visible carrier for compliance. A deeper comparison of NFC, QR, RFID and digital watermarks covers the trade-offs.
Scan data is not always the answer. Pretending otherwise leads to over-specified programmes.
The decision turns on what a wrong answer costs, not on technology preference. If a copied code would let a counterfeit pass as genuine, or a diverted pallet go unnoticed, the price of a chip is usually small by comparison.
| Category | Signals scans reveal | Typical carrier |
|---|---|---|
| Wine and spirits | Opened-before-tap status, diversion between distributors, refilled bottles | TagTamper NFC in closure or capsule |
| Luxury watches | Chain of custody across resale, authenticated service history | On-metal NFC, or tag in box or certificate card |
| Cosmetics and fragrance | Real usage and replenishment, fakes surfacing as failed verifications | NFC or secured codes on pack |
| Leather goods and fashion | Per-item authenticity on labels and linings, DPP access | Sewn-in or label NFC, QR for passport |
| Cigars | Authenticity at point of sale, grey channel exposure | Tags on boxes and bands |
| Furniture and premium audio | Registration, warranty, resale over long product lives | NFC or UHF, ownership certificates |
Leather goods and fashion carry particular weight. They make up the largest share of seized counterfeits, and textiles are among the first product groups expected under the DPP delegated acts, indicatively from 2027.
For owners in every category, the same tap can deliver an ownership certificate to a phone wallet. That gives the brand a consented route to a person it would otherwise never meet.
Published converter prices for NTAG 424 DNA-class labels cluster around USD 0.45 to 0.65 per unit at five-figure volumes. On-metal versions add 7 to 30%, and quotes at 50,000 units and above are usually by request (public catalogues, 2026). Standard non-secure NTAG213-class tags list around EUR 0.19 to 0.23 at 1,000 to 10,000 units. They cost less because they cannot prove authenticity.
| Constraint | Issue | Solution |
|---|---|---|
| Metal | Standard antennas detune on metal surfaces | On-metal tags with a ferrite layer |
| Liquids | Liquid near the antenna reduces read range | Tags on closures, necks or labels, away from the liquid |
| Phones | People must be able to read without effort | iPhone XS/XR and later read in the background; Android natively with the screen on |
| Read reliability | Every failed read is lost data and a frustrated owner | Placement testing; Selinko deployments report 99% read reliability |
The tag is rarely the largest cost line. Application on the line, integration and the experience that gives people a reason to tap usually matter more. A full cost model for NFC authentication breaks it down.
Scan data gains most of its value when joined to what the brand already holds. The unit identifier is the join key.
| System | Contributes | Joined with scans, it answers |
|---|---|---|
| ERP | When and where each batch and shipment went | Was this unit ever allocated to the country where it was scanned? |
| PIM | Product content by model | Which content to show, in which language |
| PLM | Materials and components | Passport data for this unit’s model and batch |
| CRM | Identified, consented customers | Which registered owner holds this unit, and which scans became relationships |
The CRM link deserves the most careful design. Scans should flow into the CRM only when a person opts in, for example by registering or claiming an ownership certificate. Until then, a scan identifies a unit, a time and an approximate place. That is enough for authentication and diversion analysis. The practical steps are set out in connecting ERP, PIM and PLM data to product identities.
The same item record can later serve a Digital Product Passport. The central EU DPP registry has been live since 20 July 2026 (European Commission). Brands that already log item-level events have the structure in place.
No. Product events describe the unit, not the person: unit, time, country or city, authenticity result, tamper status. They are kept for the product’s life and can be shared with partners as needed.
Personal data is a separate layer: registration, ownership, contact details. It is collected only with clear notice and consent under GDPR, and an owner’s identity is never shown to later owners or partners.
Keeping the two apart from the start lets scan data serve protection and supply chain freely, while consent governs any link to an individual.
| Team | Uses scan data to | Measure |
|---|---|---|
| Brand protection | Find counterfeit and clone clusters, refill hotspots | Failed verifications by region, cases opened |
| Supply chain and commercial | Detect diversion and real sell-through by market | Scans outside intended market, shipments implicated |
| Marketing and CRM | Reach unidentified owners, turn scans into consented relationships | Scan-to-registration rate, repeat scans |
| Client services | Support resale, warranty and service by unit | Ownership transfers, service events recorded |
A sensible start is one product line where counterfeiting, diversion or resale is a real problem. Choose a carrier that cannot be copied. Give people a reason to tap: authenticity, provenance, usage or ownership. Join scans to ERP shipments so every tap can be compared with where the unit was meant to go.
40+ brands · 12 years · €1.5B protected
A CRM shows who has introduced themselves. Scans show what is happening to every unit, including those whose owners never will. Brands that join the two on the unit identifier, and link to people only with consent, see counterfeits, diversion and silent loyalty well before a customer database does.
Product scan data is the record created each time someone reads the digital identity of one specific item. Each event holds the unit, the time, an approximate location and, with secure NFC, an authenticity verdict. It describes what happens to a product, whoever holds it, without requiring the holder to identify themselves.
CRM data is declared by identified people: purchases, contact details, campaign responses. Scan data is produced by products whenever they are read. A CRM misses gifts, resale, diversion and counterfeits. Scans capture them, because they follow the unit, not the customer. The two are complementary and join on the unit identifier.
Only with a secure carrier. An NTAG 424 DNA-class chip signs a fresh one-time code with an AES-128 key on every tap, and the platform verifies it and rejects replays. A plain serialised QR code can be photographed and reused, so its scan shows an identifier but proves nothing about originality.
No. iPhone XS/XR and later read NFC tags in the background without an app. Android phones read them natively when the screen is on. The tap opens a web address that carries the signed message, so the person sees a verification result directly in the browser.
No. A scan records a unit, a time and an approximate place. It becomes linked to a person only when that person opts in, for instance by registering or claiming an ownership certificate. Personal data is collected with notice and consent under GDPR and kept separate from product events.
Yes, with the right design. Standard tags detune on metal, so on-metal tags with a ferrite layer are used. On liquid products, tags sit on closures, necks or labels away from the liquid. Placement testing matters, and Selinko deployments report 99% read reliability.
The ESPR (Regulation (EU) 2024/1781) is carrier-neutral, but a GS1 Digital Link QR is emerging as the default visible carrier. NFC is the trust layer, proving the item is genuine. Many premium programmes use both on one product, resolving to the same identity.
Talk to our team