Tracking products with serial codes means giving every product its own number and recording where that number is scanned from production to sale. It helps detect counterfeits, but not on its own. A code is a print: it can be photographed and printed on another pack, and the copy scans just like the original. What exposes a counterfeit is not the code but the record behind it: whether the number was ever produced, whether it was activated at line exit, how many times and where it has been scanned, and whether those scans match the channel it was shipped to.
This post looks at where the method is enough and where it needs another layer. How a scan works behind the scenes is explained step by step in our serial number verification post.
Why is a serial code not enough on its own?
A serial number is an identity, not proof. If a counterfeiter prints a genuine product's code exactly, a scanner sees the same number on two packs. The lookup finds the number in the record and says "valid" on the first scan, because the number really does belong to the manufacturer.
Two things follow. First, a check that only asks whether a number exists cannot catch a copy. Second, the person who sees the warning is not always the one holding the fake. If the counterfeit is scanned first, whoever later scans the genuine product gets the "scanned before" warning. The warning does not say which pack is fake; it says the same number is circulating in more than one place.
What makes detection work?
The path from a serial code to counterfeit detection is built from five parts:
- A registry. Every number enters the record when it is created, together with the product and batch it belongs to. A number that is not in the record is caught on its first scan.
- Activation at line exit. The label is read as it is applied, and the number is activated as "produced". Labels that were printed but never applied stay separate, and an inactive number appearing in the field came from outside production.
- First-scan and duplicate-scan signals. Every scan of the number is added to the record. A number that belongs to one product being scanned by different people, in places far apart or at short intervals raises a copy alert.
- Channel data. Once you know which distributor and region a number was shipped to, a scan outside that shipment is a separate signal. Copied codes are often scanned in channels the genuine product never reached.
- Clear result wording. "Valid", "scanned before", "not in the record" and "recalled" should be shown as different results. Collapsing them all into "genuine" or "fake" misleads the user.
When is a serial code enough?
A serial code and the record behind it are a strong tool for spotting counterfeits when:
- Scans are frequent. If the product is scanned often along the chain and by end users, the second scan of a copied number arrives quickly.
- Copying is at volume. When one genuine code is printed on thousands of packs, the same number is soon scanned in many places and leaves a clear trace.
- The product is sold through a controlled channel. With an authorised retailer network and handover scans, a number that turns up outside its shipment stands out fast.
- Labels are accounted for. When labels printed, applied and destroyed reconcile, the risk of genuine labels leaking from waste onto fake products shrinks. See our label accounting post.
When is a serial code not enough?
The same method has known limits:
- Products that are never scanned. If nobody scans, the record says nothing. In channels where scans are rare, a copy can go unnoticed for a long time.
- A different genuine code on every pack. If a counterfeiter copies a different genuine code for each pack and each copy is scanned once, the duplicate-scan signal weakens. The alert only comes when the genuine product is scanned as well, which is a late signal.
- Predictable numbers. Sequential or patterned numbers make it easy to produce valid-looking numbers without ever seeing a genuine product. Numbers should be hard to guess.
- Refilled packaging. If an empty genuine pack is refilled with fake contents, the code is real and may be scanned only once. Catching this needs a separate measure that shows the pack has been opened.
The EU's approach to medicines shows these limits are well understood. Since 9 February 2019, the outer packaging of medicines in scope has carried two safety features together: a unique identifier and an anti-tampering device. The identifier includes a random serial number unique to each pack. Before handing the pack to the patient, the pharmacy scans it and the number is set to "supplied" in the national repository. A pack whose number already has that status cannot be verified and raises an alert. Serial number, registry and pack integrity were designed to work together.
What layers can you add to a serial code?
- A copy-resistant code. A production method that makes the code itself hard to copy raises the bar before any second scan is needed.
- Pack integrity. A cap, tape or seal that leaves a trace when opened reveals refilled packaging.
- A layer that verifies the pack itself. Overt or covert security features check whether the pack, rather than the code, is genuine.
- Field checks. Store scans by field teams and test purchases make the record denser in channels where consumers rarely scan.
What should you ask when evaluating a serial code system?
- Is the number activated at line exit, or treated as valid as soon as it is printed?
- Does the lookup check only that the number exists, or its scan history too?
- Does the record include the distributor and region the number was shipped to?
- Does the result screen clearly separate "valid" from "scanned before"?
- Are numbers generated so that they are hard to guess, and is the code itself copy-resistant?
- Are printed, applied and destroyed labels counted?
- When an alert fires, is it clear who acts, on which record and how?
How does a counterfeit suspicion show up in Tagvex?
Tagvex is a traceability system that gives each product, or each batch, its own copy-resistant code, tracks the distribution path and makes channel diversion visible. The code is produced to make copying harder and is read and recorded at line exit. Handovers at warehouse, distributor and dealer points are gathered in the same record. When a company team, a dealer or a consumer scans the product with the mobile app, the result is based on that record, and a code scanned in an unexpected place or again and again is flagged.
Because copied-code signals are meaningful at product level, product level is the default in Tagvex. A batch-level code helps with recalls and channel tracking but cannot separate a copy from the genuine product; we explain the difference in our post on unique unit codes versus batch codes. How to read diversion in the record is covered in our channel diversion guide.
Frequently asked questions
Can a product with a serial number be fake?
Yes. A serial number can be copied, and the copy is found in the record on its first scan. What points to a fake is repeated or out-of-shipment scans of the same number, and numbers that were never activated at line exit.
Does a "scanned before" warning mean the product is fake?
Not on its own. The warning describes the number's history: the product may have been scanned earlier by a dealer, a family member or a counterfeiter. Where and when that scan happened is weighed together with where the product was bought.
Can batch numbers detect counterfeits?
To a limited extent. A batch number that is not in the record, or one seen in a region the batch never went to, is a signal. But every product in a batch shares the number, so repeated scans are expected and do not separate a copy at product level.
How do you measure whether a serial code system works?
Track three ratios: how many products are scanned, how many applied labels were activated at line exit, and how many alerts are confirmed by field checks. The last one shows whether thresholds and result wording are set correctly.
