Company swag inventory: SKUs, barcodes or individual asset IDs?

Use a variant SKU to count interchangeable merchandise. Add an individual asset ID only when you need the history of a particular physical item. A barcode is the scanning mechanism—not the decision about what you are tracking. For a company swag closet, confusing these layers can make a fast scan produce the wrong inventory update.

A navy medium polo can share a variant identity with dozens of identical polos. A loaned sample jacket may need its own identity because someone must return that exact jacket. Neither the employee’s gift code nor the carrier’s tracking number answers those inventory questions.

Brand Sauce publishes this guide as a company-store provider. The design below is our recommended operating framework, not a claim that our stores include barcode scanning, asset management or any particular integration. Confirm software, labeling, hardware and implementation scope separately; unverified costs are quote-needed. Official sources were checked October 15, 2026.

Choose the question before choosing the label

Start with the consequence of mixing two objects up. If exchanging one identical notebook for another changes nothing, a count by variant may be enough. If exchanging two jackets would lose an alteration history or an outstanding loan, the system needs to distinguish the objects.

Keep the system proportionate. A small, occasionally used closet may work with readable shelf labels and a controlled spreadsheet. Barcode scanning becomes worth testing when repeated identification or transcription is a demonstrated problem. Individual serialization adds work at receipt, issue and return; do not add it merely because the label printer can produce unique numbers.

Separate SKU, barcode and supplier identity

A SKU is your internal stock-keeping reference. A barcode carries machine-readable information. A GTIN is a standardized trade-item identifier. These can cooperate without being interchangeable fields.

GS1 explains that a GTIN identifies a trade item so systems can look up its associated information. That is different from assigning every physical copy its own history. GS1’s serialization explanation distinguishes a base product identity from an individual instance, including the combination of GTIN and serial number.

For your internal program, maintain explicit mappings rather than inventing a number that looks like an official product code. If an external marketplace, trading partner or fulfillment service requires standardized identification, ask who owns allocation and which rules apply to the finished item. An internal closet SKU is not a substitute for that requirement.

Shopify’s SKU guidance recommends distinct SKUs for product variants and keeping the same variant SKU across locations while tracking location quantities separately. This is a useful concrete platform example, not a guarantee about every company-store system. Ask your provider which identifiers its imports, exports and scanners actually use.

A supplier’s barcode may describe the undecorated garment. Your picking operation may need to distinguish that blank from the same style with a division logo, an outdated logo or a different approved print. Preserve the supplier reference, but map it to the finished configuration instead of assuming it contains information added later.

Write a finished-variant contract

Before printing labels, define what makes two units interchangeable for fulfillment. Record the underlying style, fit, size, color and approved decoration configuration. If an artwork change means old and new pieces cannot fulfill the same request, give operations a reliable way to distinguish them.

You do not need to encode the whole specification into a long SKU. A short stable identifier can point to a descriptive record. The important requirement is that a picker, importer and support agent resolve that identifier to the same finished product.

• Internal variant ID: a stable reference that will not be reused for an unrelated item.
• Supplier mapping: supplier name, style and supplier variant reference, plus applicable barcode values.
• Finished configuration: size, fit, color, decoration version and other fulfillment-relevant differences.
• Unit of measure: one garment, one kit or one defined pack—not an unexplained “unit.”
• Location and state: stored separately so moving or inspecting an item does not change its product identity.
• Label ownership: who creates, approves, applies and replaces the operational label.

Illustrative configuration, not a customer record: one polo style offered in two colors, four sizes and two noninterchangeable logo versions creates 16 finished variants. Keeping those variants at three offices still gives 16 variant identities, with up to 48 location-level stock records. It does not require 48 different product SKUs.

The same logic helps during a rebrand: historical references should continue to resolve to the old configuration rather than silently displaying the new logo. Decide whether old stock remains issuable, but do not rewrite what an earlier order actually contained.

Serialize exceptions that need individual history

Consider individual asset tracking for returnable try-on samples, altered garments or reusable event equipment when the business needs to identify the exact object. This is a different decision from tracking ordinary gifts after employees receive them. Do not impose a return obligation on a gift merely because it carries an inventory label.

Link each asset ID to a variant, then store its own status and history. A replacement label should preserve the asset record rather than create a second apparent jacket. If the old label can still be scanned, define whether it resolves to the same record or produces an explicit retired-label warning.

Do not put an employee’s name, email or home address inside the scan payload just to locate a loan record. Use an opaque internal reference and authorized access to the relevant details. The label identifies the object; the application decides who can view its custody information.

For hotel workwear specifically, our assigned-versus-pooled uniform guide covers care and custody states. Here the implementation question is narrower: can the identifier survive reassignment without losing the identity of the garment?

Make every scan an explicit transaction

A scan should first resolve an identity, then apply an intentional operation. “Receive,” “transfer,” “issue” and “return for inspection” are different actions. A scanner left in the wrong mode can correctly recognize a product while recording the wrong movement.

Show the resolved description, size, decoration version, unit and location before confirmation. For individual assets, show enough context to detect a duplicate issue or an unexpected return. Unknown codes should enter a resolution queue, not quietly create new products or increment a generic merchandise total.

Handle packs explicitly. In an illustrative receipt of two cartons containing 24 shirts each, the expected total is 48 shirts. A carton scan must either identify a recorded contents list or request a verified quantity. Do not let an ambiguous label alternate between “one carton” and “one shirt” depending on who is holding the scanner.

Separate an accidental repeated scan from an intentional second unit. Variant counting often needs repeat scans; asset checkout usually must reject a second checkout of the same object. Define confirmation and correction controls for each mode. If an integration retries a completed transaction, it should not add another receipt merely because the message arrived again.

Record the transaction reference, operation, object or variant, quantity, location, timestamp and authorized operator. Corrections should leave an understandable history. Our order-sync recovery guide addresses the related problem of duplicate and missing updates between systems.

Test the label where the handoff happens

GS1’s barcode overview describes one-dimensional and two-dimensional symbols as machine-readable data carriers. Choosing a QR-shaped symbol instead of bars does not by itself supply inventory software, permissions or a reliable transaction workflow.

Test the actual scanner or phone, application and printed label together. Include the lighting, package surface and handling distance of the real receiving station. Put readable text beside the code so a damaged label has a controlled fallback. Keep product, shelf and carton labels visually distinguishable.

For merchandise being given away, a removable packaging label may be sufficient. For a reusable garment, test the proposed attachment and readability through its approved care route before choosing permanent labeling. Do not cover required manufacturer information or assume that an adhesive suitable for a shipping carton belongs on fabric.

Specify what staff do when both human-readable and machine-readable identification fail. Set the item aside, compare approved product evidence and have an authorized person restore the mapping. Guessing the nearest-looking size or logo version defeats the purpose of scanning.

Use a seven-case acceptance test

Run this with representative test stock before expanding. Keep expected and actual outcomes, plus an owner for every unresolved discrepancy. This is a proposed acceptance card, not a report of measured Brand Sauce results.

Finally, compare the records with a physical count. A scanning demonstration proves little if the system cannot explain the closing balance. Use the classification and reconciliation approach in our company swag inventory audit to distinguish available reserves, blocked stock and unexplained differences.

Match tracking effort to the production model

Bulk stock can justify variant-level receiving and picking controls when repeat events or immediate availability warrant a reserve. A hybrid program may track local event stock while sending less predictable employee selections through per-order production. Keep supplier availability separate from stock your organization actually owns.

On-demand production reduces some local inventory tasks, but order-to-finished-item mapping still matters. Conversely, serializing every low-cost giveaway can create more handling than the program needs. Compare labeling, data maintenance, hardware, training and correction effort against the specific mistakes the proposed workflow is meant to prevent. There is no universal savings percentage.

Frequently asked questions

Does every company gift need its own serial number?

No. Use individual identity when a particular object’s history matters. Interchangeable gifts can usually be managed as quantities of an identified finished variant.

Can we scan the manufacturer’s existing barcode?

Possibly, if the application supports it and the mapping resolves the correct finished configuration. Test whether decoration, packaging or other differences require an additional internal reference.

Should we buy scanners before choosing software?

Define the transaction and supported identifier first. Then test the actual reader, label and application together before committing to a rollout.

The useful purchase is not a roll of codes. It is a dependable answer to “what is this, what happened to it, and can we issue it now?”