Packaging Procurement Guide

Variable-Data Packaging Procurement: 3 Practical Boundaries—Data Ownership / Version Freeze / Batches and Warehousing—Which 3 Details to Confirm First for a “Every Box Is Different” Inquiry

📅 2026-11-09 ✍️ Wuxi Lexiang Printing & Packaging ⏱ 3min read

💡 💡 At a Glance

The risk of a variable-data order lies not in printing, but after order release: variable data may become different quantities, different packaging specifications, or different delivery addresses. Procurement teams should prepare 3 tables—a data delivery table covering the provider, timing, format and encoding, and row-count validation; a version freeze table covering the freeze point, sealed sample method, quantity changes, and waste clauses for already-printed output; and a batch and shipment table covering batch allocation, serial number range ledgers, and shipment verification. Teams should also compare the costs of a fully variable solution with separate-version printing.

“The code on every box in this batch will be different. Can you do it?”

Today, most printers answer “Yes.” Variable-data printing through digital printing is a mature technical process. For serial numbers, QR codes, personalized copy, and one-item-one-code applications, printers typically combine variable merging at the RIP stage with backend inkjet or laser coding.

Therefore, the real risk lies not in printing. The real risk is that—what the customer describes as “variable” can turn into something else during production.

This creates 3 order-release risks for variable-data printing VDP—variable data becoming different quantities, different packaging specifications, or different delivery addresses. When a printer accepts its first VDP order, what it most often underestimates is not printing, but layout and warehousing.

This article approaches the same issue from another perspective: from the position of the procurement party, what tables should a variable-data order be divided into when it is placed, and which details should be locked down? The most useful support a printer can provide is to present these tables during the quotation stage.

The First Boundary: Who Provides the Data, When, and at What Level of Detail

The fundamental difference between a variable-data order and a standard order is that: standard-order data is complete before printing, while variable-data order data may continue changing during production.

Four fields must be clarified at the data level:

FieldQuestionWhy It Determines Scheduling
Who provides itDoes the customer provide a file, or provide an interface that allows the printer to export the data?A customer-provided file is the simplest option. An interface means the data resides in the customer’s system, and the printer must retrieve it before each production run. The retrieval window becomes the scheduling boundary.
When it is providedIs the full dataset provided once after artwork approval, or is the daily batch provided 2 hours before production?The latter enables roll-to-roll production. The press cannot stop or change batches, so the scheduling logic is entirely different.
Level of detailIs it item-level detail with one record per unit, or does it follow a regular generation rule, such as serial number + fixed prefix?With regular rules, the printer can generate the data using a variable algorithm, resulting in an order-of-magnitude difference in efficiency. Without regular rules, every record must be imported and checked individually.
FormatExcel / CSV / direct database connection? What encoding is used?Garbled Chinese text, inconsistent encoding, and mixed full-width and half-width characters are the most common sources of incidents in these orders.

The third field deserves further explanation. For regular rule-based data, such as serial numbers running from 1001 to 5000 with a fixed prefix, the printer can perform variable merging in the RIP, with high processing speed and low error rates. For irregular item-level data, where every row corresponds to a real customer’s name, address, and card number, the data must be imported and checked row by row. Row-count validation is mandatory, because if one row is missing, duplicated, or shifted, the result is not one fewer printed item—it is incorrect shipment.

When it comes to format, the customer’s original description is often casual: “We can just export an Excel file.”

The printer’s response is often: “The Excel file has several thousand rows. What should I open it with?”

Then an entire night is lost. This field must therefore be finalized before quotation, including the encoding, whether it is UTF-8 or GBK, the delimiter, whether there is a BOM header, and the date format.

The Second Boundary: Version Freeze—The Most Counterintuitive Rule for Variable Data

Customers often assume that variable-data orders do not require final artwork approval because “every item is different anyway.”

Quite the opposite is true: the artwork for a variable-data order must be frozen earlier and more firmly than for a standard order.

This is because two layers change simultaneously in a variable-data order:

  • The variable layer: the code, name, and copy on each box. This layer is designed and does not change.
  • The fixed layer: the layout itself, color scheme, dimensions, finishing processes, and materials. Any change to this layer affects all N variable units.

The customer’s instinct is to “start printing and revise the copy as we go.” The reality for the printer is: once the layout changes, the entire press must stop, color calibration must be repeated, and proofing must be rerun, wasting everything printed previously.

Therefore, a variable-data order should specify the following 3 matters in the contract:

1. The version-freeze point. The data file is considered final only after the customer has approved the layout, approved the proof, and confirmed the production quantity. Any subsequent customer request constitutes a change and must follow the change-control procedure, with the customer bearing the costs of restarting the press, wasting printed output, and repeating color matching.

2. Sample tracking and sealed samples. Variable-data orders especially require physical sealed samples and a signed sample archive because every item is different, meaning that subsequent accountability can only rely on “the item printed at that time.” Among the 3 practical methods for creating a digital-printing sample archive—physical sealed samples / color cards / sample photographs—physical sealed samples have far greater value for these orders than for standard orders, because they prove not only the color but also the combination of “this data batch + this artwork version.”

3. Traceability rules for quantity changes. This is the core requirement. The customer orders 5000 units and, after 2000 units have been printed, requests a change to 3000 units. The 2000 units already printed exist, and where should the serial number range for the remaining 3000 units begin? Should it continue from 2001, or should the customer specify a new sequence?

If the customer requires resequencing—for example, units 1-2000 were intended for customer segment A but are now intended for customer segment B—then all 2000 printed units must be wasted, because serial numbers cannot be reused or mixed across different batches. This is the most expensive type of change in a variable-data order.

The rules for quantity changes must therefore be specified in the contract, including how the already-produced portion will be handled when a change occurs.

The Third Boundary: Batches and Warehousing—The Most Underestimated Half

The variability of variable data ultimately appears in warehousing and shipment.

A typical scenario is a customer ordering 5000 packaging boxes with unique codes. After printing, they are not shipped all at once but are shipped in batches according to sales progress. The printer is then managing not “one batch of goods” but “5000 independently identified units of goods.”

There are 3 specific warehousing challenges:

First, separating same-version and different-version print runs. Within the 3 practical boundaries for splitting packaging orders, same-version split printing means printing the same artwork in several runs because other jobs have been inserted into the schedule, while different-version split printing means printing different specifications separately. Both often occur simultaneously in variable-data orders—for example, 1000 units of the same box type are printed in 3 runs, with the customer changing the copy in between.

Second, slow-moving finished goods and WIP occupying space. Within the 3 practical management accounts for a printing plant’s warehouse, slow-moving finished goods and WIP occupying space are two areas that are easily underestimated. The special feature of a variable-data order is that every item produced is non-interchangeable. A box with a unique code can only be sold to the corresponding customer. Once printed, the printer cannot hold it as generic inventory while waiting for the next order.

Third, continuous management of numbering. This is the real operational task. The printer must be able to answer the customer’s questions at any time: “What is the code range for this batch of 3000 units?” “Where did the previous batch end?” “Do these 2 batches contain any duplicate numbers?”

In practice, a ledger is needed containing the batch number, serial number range, printing date, finished goods destination, and shipment records. During customer inspection, 10 units are randomly selected and scanned. If every code matches and agrees with the system records, that is the acceptance action for delivering a variable-data order.

Another issue is easily overlooked: the cost of shipment errors. If standard packaging is shipped incorrectly, the customer can send a replacement. If variable-data packaging is shipped incorrectly, an address for customer A may be applied to a box for customer B. For some product categories, this involves personal information, turning a “shipment error” into a “data breach.” Shipment verification must therefore be stricter than for standard orders.

The 3 Tables to Present Before Placing an Order

Required TableContentsParty Confirming
1. Data delivery tableData provider, delivery timing, format, encoding, delimiter, row-count validation rules, and whether generation rules applyCustomer IT/data team + printer production team
2. Version freeze tableVersion-freeze point, sealed sample method, quantity change rules, waste clauses for already-printed output, and change feesCustomer project team + printer production team
3. Batch and shipment tableBatch allocation rules, serial number range management, finished goods storage method, batch shipment rules, and shipment verification actionsCustomer warehouse + printer warehouse

Presenting these 3 tables during the quotation stage is far more professional than having the printer state after the fact that “there is a problem with this data format.”

From the procurement party’s perspective, the value of these 3 tables is that they turn “we can print variable data” from a technical promise by the printer into a management process with shared responsibility. Most of the risk exposure in variable-data orders lies not in printing, but in data handover and version management, and both parties can manage these areas.

There is also a judgment often overlooked by procurement teams: does this order actually need variable data?

If 200 out of 5000 boxes are special, with different copy or addresses, using variable data for the entire batch is more expensive and slower than producing 4000 standard-version units and separately processing 200 units. The 3 practical cost boundaries of using one die for multiple packaging specifications—multipacks / size changes / channel-specific versions—address the same principle: the fewer versions, the shorter the chain, and the lower the risk.

When the printer presents the costs of both a “fully variable” solution and a “separate-version processing” solution during quotation, customers will often choose for themselves.

Further Reading

#Variable Data #VDP #One-Item-One-Code #Serial Numbers #Packaging Procurement #Version Freeze #Warehouse Management

FAQ

Where do variable-data orders most commonly go wrong?

Not during printing, but after the order is released. Variable data may become different quantities, different packaging specifications, or different delivery addresses—each of which can render printed output unusable or mismatched. When a printer accepts its first VDP order, what it most often underestimates is not printing, but layout and warehousing.

Is it acceptable for the customer to revise the copy while printing is underway?

Technically possible, but extremely costly. In a variable-data order, the data layer changes while the layout layer remains fixed. Once the layout changes, the entire press must stop, color calibration must be repeated, and proofing rerun, wasting everything printed previously. The artwork must therefore be frozen earlier than for a standard order. The contract should specify the version-freeze point and change fees.

The customer orders 5000 units and changes the order to 3000 units after 2000 have been printed. What happens to the printed units?

It depends on whether the customer requires the serial number range to be resequenced. If the customer changes the customer segment or encoding rule, all 2000 printed units must be wasted because unique codes cannot be duplicated or mixed across batches. This is the most expensive type of change in a variable-data order, and the contract must specify in advance how the already-produced portion will be handled when a change occurs.

What data-file format should be specified?

At minimum, specify 4 items: encoding, whether it is UTF-8 or GBK; the delimiter; whether there is a BOM header; and format conventions for dates and numbers. Garbled Chinese text, inconsistent encoding, and mixed full-width and half-width characters are the most common sources of incidents in these orders. Row-count validation rules should also be agreed, and the customer should confirm the total row count.

What should be considered when warehousing 5000 boxes with unique codes for batch shipment?

There are 3 key points: account for same-version and different-version print runs separately; finished goods cannot be treated as generic inventory because each item corresponds to a unique customer; and a batch-to-serial-number-range ledger must be maintained so the printer can always answer questions such as “Where does this batch’s code range end?” and “Are there any duplicate numbers?” Shipment verification must be stricter than for standard orders, since an incorrect shipment in a variable-data order can become a data incident.

When should variable data be used, and when is separate-version printing better?

It depends on the proportion of variation. Using variable data for an entire batch when only a few items are special is generally more expensive and slower than separate-version processing. The same principle applies when multiple packaging specifications share one die: the fewer versions, the shorter the chain, and the lower the risk. When the printer presents the costs of both options during quotation, customers can usually make the decision themselves.

How does sealed-sample control for variable-data orders differ from standard orders?

It has greater value. A physical sealed sample for a standard order proves the color and finishing processes. For a variable-data order, it proves the combination of “this data batch + this artwork version.” Because every item is different, subsequent accountability can rely only on the item printed at that time. Among sample archives comprising physical sealed samples / color cards / sample photographs, physical sealed samples are indispensable in this scenario.

If the customer provides a database interface that the printer must retrieve data from before each run, what issues may arise?

The data resides in the customer’s system, making the retrieval window the scheduling boundary. If the customer cannot retrieve the data, the printer’s press remains idle. The contract should specify whether the delivery date will be extended if data is delayed or whether the customer must provide the data separately, as well as provisions covering repeated retrieval failures. Such situations commonly lead to disputes over press occupancy time.

Need a Custom Packaging Solution?

Learn more about packaging, or consult directly for a custom solution and quote