Serialization

Serialization is the practice of assigning a unique, non-repeating identifier to each individual unit of a product, instead of identifying only the lot or the SKU. While a regular product code tells you "this is a package of such-and-such medicine," the serialized code tells you "this is unit number 0000457 of that package." That serial number, combined with data such as lot and expiration date, turns each box, bottle or pallet into an object that can be tracked individually.

In practice, serialization is the foundation of unit-level traceability: without a unique code per unit, you cannot reconstruct the exact path of a specific item or verify its authenticity. It is the base that a traceability system records and unifies, one of the concepts that traceability tools such as Trazzo put into practice, connecting each code scan with movement events (receiving, shipping, sale).

The identifier usually takes the form of a two-dimensional barcode (typically a DataMatrix under the GS1 standard) that encodes the product's GTIN plus the serial number, the lot and the expiration date, so a single scan returns the unit's full identity.

Why serialization matters

Serialization solves a problem that lot-level identification cannot: unit-level granularity. When you work only with lots, an entire group of thousands of units shares the same code. If a quality problem, a theft or a counterfeit shows up, you cannot isolate the affected unit: the whole lot has to be treated as suspect. With serialization, each unit carries its own "fingerprint," which makes it possible to recall, block or verify individual items without halting the rest.

This has three core uses for companies in Latin America. First, regulatory compliance: in the pharmaceutical industry, agencies such as ANMAT in Argentina require medicines to be tracked at the unit level. Second, fighting counterfeiting and diversion: a unique code makes it possible to detect duplicates (the same serial number scanned twice is an immediate alert). Third, operational efficiency: with every unit identified, recall, receiving and inventory reconciliation processes no longer depend on manual counts.

How it works, step by step

1. Serial number generation: a central system creates unique identifiers, guaranteeing that none is repeated within the product's universe.
2. Printing and marking: the code (usually a GS1 DataMatrix) is printed on the unit on the production or packaging line.
3. Verification: a camera or scanner confirms the code is readable and correct before the unit moves on.
4. Event capture: at each point in the chain (plant, warehouse, distributor, point of sale), the code is scanned and the system records what happened, when and where.
5. Aggregation: serialized units are linked to their larger packaging through a logistics unit code (SSCC), so scanning a pallet reveals all the units it contains without opening it.

A concrete example in Argentina

A pharmaceutical company that distributes medicines in Buenos Aires needs to comply with the traceability required by the health authority. Each printed blister pack receives a DataMatrix with its unique serial number. When a pharmaceutical wholesaler receives the goods, it scans the boxes and the system confirms that those specific serial numbers left the manufacturer and were never reported as sold or expired. If a pharmacy later scans a unit whose serial number is already recorded as dispensed in another province, the system triggers an alert for possible counterfeiting or diversion within seconds, something impossible with lot-only identification.

Serialization vs other concepts

It is easy to confuse serialization with traceability or with simple lot identification. The difference lies in the level of detail and in the role each one plays.

In short, serialization generates the data, aggregation organizes it hierarchically and traceability uses it to tell the story of the product.

Common mistakes

A frequent mistake is believing that serializing is just "printing a bigger barcode": in reality, it requires an infrastructure to generate serial numbers without collisions, store them and query them in real time. Another mistake is serializing without aggregation, which forces you to open every box to scan unit by unit. Finally, many companies underestimate print and read quality: a poorly printed DataMatrix that does not scan breaks the entire data chain and forces costly reprocessing.