Packaging, Cartonization and Dimensional Weight
Parcel carriers charge for the space a package occupies, not only its mass. For light, bulky shipments the box sets the price, and every cubic inch of air is billed.
Cube usually sets the price, not mass. For light bulky parcels the dimensional weight exceeds the scale weight, so the shipper is billed for volume.
Void fill counts as empty space. Under the European rule, filling a too-large box with cushioning does not solve the problem. Using a smaller box does.
Surcharge thresholds are step functions. Crossing a cubic threshold by one inch costs the same as crossing it by a foot, which makes box dimensions worth engineering to the limit.
Cartonization is a hard computational problem. It is a bin packing variant, which is NP-hard, so commercial systems use heuristics rather than provably optimal packing.
Discount the savings percentages entirely. Every circulating figure on cost reduction or the share of parcels billed on dimensional weight comes from packaging vendors or parcel auditors.
Market overview
The short answer
Parcel carriers price on the greater of a package's actual weight and its dimensional weight, which converts the volume the package occupies into a billable weight using a divisor. For dense goods the scale wins. For the light, bulky items that dominate e-commerce, the box wins, and the shipper pays for the air inside it. That makes box right-sizing the primary cost lever, ahead of negotiating rates or changing void fill. Cartonization is the computational side of the same problem: choosing which boxes to stock and which items go in which one. As of August 2026 the argument has a second front, because the European packaging regulation now applies and introduces a limit on empty space that turns right-sizing from a cost decision into a compliance one.
KEY FACTS
Verified August 2026. Each statement below is complete on its own and cites its source in section 08.
What is cartonization, and where does it sit?
Cartonization is the decision of which carton or cartons an order should ship in, and which items go in each. It answers three questions at once: how many boxes, which sizes from the available range, and how the items are arranged so that they fit and arrive intact. It is invoked at order release, typically by the warehouse management system or the order management system, both of which SCR covers separately. It is a decision layer rather than a product category, and buyers frequently discover during evaluation that what they wanted was better cartonization logic inside systems they already own.
Underneath it is a well-studied computational problem. Choosing how to fit items into a fixed set of container sizes is a variant of bin packing, which is strongly NP-hard, and the three-dimensional version is harder still: the difficulty was established formally in the operations research literature, and it means that solution time grows explosively with item and container count. Commercial systems therefore use heuristics such as first-fit decreasing and best-fit decreasing, or metaheuristics, and produce good packings rather than provably optimal ones. That is entirely adequate operationally, and it is worth knowing when a vendor describes its packing as optimal.
The decision that matters more than the algorithm is the box range itself. A cartonization engine can only choose from the sizes a facility stocks, so a poorly designed range guarantees wasted cube regardless of how good the logic is. Adding sizes improves fit and increases handling complexity, storage space, and replenishment burden, which is the real trade-off. Analyzing the actual order profile against candidate box ranges is the exercise that produces most of the available benefit, and it can be done before any software decision.
Two adjacent constraints belong in the same analysis. Damage risk sets a floor on how tightly items can be packed and how little protective material can be used, and it should be measured rather than assumed, because a right-sizing program that raises damage rates can cost more than it saves. And pack station throughput matters: a scheme that requires more careful arrangement slows the line, which is a labor cost that rarely appears in the packaging business case.
How is dimensional weight calculated, and what changed in 2026?
The arithmetic is simple. Multiply length by width by height in inches, divide by the dimensional divisor, and round up to the next pound. Compare the result against the actual scale weight and bill the greater of the two. A lower divisor produces a higher dimensional weight, so a change from 166 to 139 raises the billable weight on the same box by roughly nineteen percent, which is an arithmetic consequence rather than a rate increase.
Figure 1. One parcel produces two weights, and the higher one is billed. The right-hand ladder shows the cubic thresholds that trigger surcharges in 2026. Because these are step functions, a box designed just under a threshold and one designed just over it carry materially different costs.
Several things changed across 2025 and 2026 and are worth holding together. Both major private carriers moved to rounding each fractional dimension up to the next whole inch from August 2025, which raises billable weight slightly on almost every non-integer box. Both set headline general rate increases of 5.9 percent for 2026, the third consecutive year at that figure, though the effective increase experienced by a given shipper is usually higher once surcharges, residential fees, and fuel are layered on. The postal service moved its divisor to 139 in July 2026 for packages over a cubic foot, narrowing a gap that had made it the cheaper option for bulky lightweight parcels.
The surcharge thresholds deserve separate attention because they are where large single charges appear. Additional handling now attaches above 10,368 cubic inches and the large package or oversize charge above 17,280 cubic inches or 110 pounds, with both major carriers sharing those numbers. One carrier also restructured its distance zones and raised its charge for packages exceeding size limits substantially. Because these are cliffs rather than gradients, the packaging engineering question is not merely how to shrink boxes but where the cliff edges sit relative to the products actually shipped.
Table 1. Dimensional weight rules as of August 2026. Every figure here should be verified against the carrier's own tariff or service guide before it informs a rate negotiation, because these documents are reissued at least annually.
Which levers actually move parcel cost?
Three levers do most of the work, and they are not equally available. Box right-sizing reduces the dimensional weight directly and can move a parcel below a surcharge threshold, which is where the largest single savings appear. Reducing void fill helps only insofar as it permits a smaller box: filling a too-large carton with less cushioning changes nothing about the billable weight and increases damage risk. Consolidating multiple items into fewer cartons reduces the per-parcel charges that apply regardless of size, though it interacts with damage and with pick sequencing.
A fourth lever is contractual rather than physical. The divisor itself is negotiable, and a higher negotiated divisor reduces dimensional weight on every shipment without any operational change. Shippers focused entirely on packaging engineering sometimes leave this untouched, and it is worth establishing what divisor a contract actually specifies before designing a box program around the published default.
On the size of the prize, this page will not give a number, and the reason is worth stating plainly. Figures circulate widely describing the share of parcels billed on dimensional rather than actual weight, and the typical multiple between the two. Every one that SCR could trace originates with a packaging manufacturer, an on-demand box vendor, or a parcel audit firm, all of which sell services whose value rises with the number. No carrier, regulator, or independent researcher publishes a population-level statistic on how much of a typical parcel invoice is driven by cube. Carrier tariffs specify the rules; they do not publish the distribution of outcomes.
What a shipper can do instead is measure its own. The data required is already in the carrier invoice: for each parcel, the actual weight, the dimensions, and the billed weight. Computing the share of shipments where dimensional weight governed, and the aggregate difference between the two, produces a defensible number specific to that operation. It takes a few hours against a month of invoices and it is worth more than any published benchmark.
What does packaging design trade off?
The central tension is protection against cube. Every reduction in box size or protective material lowers shipping cost and raises the probability that something arrives damaged, and damage is expensive in a way that compounds: the replacement item, the second shipment, the handling of the return, and the customer relationship. A right-sizing program that does not measure damage rates before and after is not measuring its own effect.
On-demand box making, where equipment produces a carton sized to the order from fanfold material, addresses the box range problem directly by removing the constraint of a fixed set of sizes. It carries a capital cost, a footprint requirement, and a throughput characteristic that has to fit the pack line. It also frequently carries a commercial structure worth examining: some vendors tie the equipment to a proprietary consumable on subscription, which is capital-light and creates a supply dependency. Neither structure is wrong, and a buyer should know which one is being proposed.
The returns interaction is the trade-off most often missed. Packaging for a category with a high return rate may need to survive a second journey, because the customer will reuse the original box, frequently after opening it inelegantly. Minimal packaging optimized purely for the outbound leg produces returns that arrive damaged and unsaleable, converting a restockable unit into a liquidation or a write-off. SCR covers the disposition economics of that outcome in its returns guide; the packaging consequence is that outbound cube optimization and inbound recovery value pull against each other for returnable categories.
The fair case against aggressive right-sizing deserves stating. Optimizing cube in isolation is a local optimization that can raise total cost: it increases damage, it can slow the pack line, and on-demand equipment carries capital and consumable commitments that outlast the rate environment that justified them. The honest framing is that right-sizing is the dominant lever on cube-driven invoices and must be evaluated against damage, labor, and capital rather than sold as a standalone saving.
What do the packaging regulations require?
The European Union replaced its long-standing packaging directive with a regulation, which entered into force in February 2025 and applies generally from 12 August 2026. Because it is a regulation rather than a directive it applies directly rather than through national transposition, which removes much of the country-by-country variation that characterized the previous regime. Its provisions phase in over several years.
The provision most relevant to this page is the limit on empty space. For grouped, transport, and e-commerce packaging, the regulation sets a maximum empty space ratio of 50 percent, applying by 1 January 2030 or three years from the relevant implementing acts, whichever is later, with the Commission required to adopt the calculation methodology by February 2028. The detail that changes packaging practice is that void fill counts as empty space: air cushions, bubble wrap, foam, and paper cuttings do not reduce the ratio. The compliance response is a smaller box, which is the same response the parcel economics already reward.
In the United States the picture is state by state and less settled. Seven states have enacted packaging producer responsibility laws, and as of August 2026 only two had live producer fees, with a producer responsibility organization approved in six of the seven. Programs are at different stages, obligations and fee schedules differ, and there is no federal statute. Litigation in one state during 2026 introduced further uncertainty that could affect how other programs proceed. Any organization planning against these should verify state by state rather than treating them as a single regime.
Table 2. Packaging producer responsibility as of August 2026. The European position is settled in principle and unsettled in calculation detail; the United States position is unsettled in both, which is the opposite planning problem.
Frequently asked questions
What is billable weight?
The weight a carrier actually charges for, which is the greater of the package's actual scale weight and its dimensional weight, rounded up to the next pound. For light bulky parcels the dimensional weight usually governs, so the shipper pays for volume rather than mass.
What is the dimensional divisor and what is it in 2026?
The number that converts cubic inches into a billable weight. Major carriers publish 139 for United States domestic and export daily rates, with 166 at the retail counter for one carrier, and 5,000 for metric calculations. Negotiated contracts frequently specify a higher divisor, which lowers billable weight.
Did the postal service really change its divisor?
Yes. It moved from 166 to 139 on 12 July 2026, applying only to packages over one cubic foot, and adopted rounding of each fractional dimension up to the next whole inch. On the same box, that change raises the dimensional weight by roughly nineteen percent as a matter of arithmetic.
Why does my light package cost so much?
Because its dimensional weight exceeds its actual weight, so the carrier bills the larger figure, and because cubic thresholds may have triggered an additional handling or oversize charge. Both are functions of the box rather than the contents.
Does adding void fill help or hurt?
It does not reduce the billable weight, since the box dimensions are unchanged, and under the European empty space rule it does not count toward compliance either. Void fill protects the product; reducing the box size is what reduces cost and empty space.
What is the European empty space rule?
The packaging regulation sets a maximum empty space ratio of 50 percent for grouped, transport, and e-commerce packaging, applying by 1 January 2030 or three years from the relevant implementing acts. Void fill counts as empty space, so the compliance response is a smaller box.
Which United States states have packaging producer responsibility laws?
Seven have enacted them, and as of August 2026 only two had live producer fees, with a producer responsibility organization approved in six of the seven. Obligations and timelines differ by state and litigation in 2026 added uncertainty, so verify state by state.
Is cartonization part of my warehouse system or a separate product?
It is normally a decision layer invoked at order release by the warehouse or order management system rather than a separate category. The more consequential variable is the range of box sizes stocked, since the logic can only select from what exists.
Is on-demand box making worth it?
It depends on order profile variability, pack line throughput, and the commercial structure offered. It removes the fixed box range constraint, and it carries capital, footprint, and sometimes a proprietary consumable commitment. Evaluate it against damage rates and labor, not against a vendor savings percentage.
How much will right-sizing save us?
No credible independent benchmark exists, and every circulating figure comes from packaging vendors or parcel auditors. Compute your own from a month of carrier invoices: for each parcel compare actual against billed weight, and total the difference. That number is defensible and specific to your operation.
Method, sources, and where to go deeper
Method
Dimensional weight rules and thresholds were taken from carrier tariffs and service guides and from trade reporting of carrier announcements, and are stated with their effective dates.
The computational characterization of cartonization follows the peer-reviewed operations research literature on bin packing rather than vendor descriptions of packing engines.
Regulatory status follows the European Commission and the regulation text, with legal analyses used to corroborate provisions and flagged as interested-party-adjacent.
Supply Chain Research is independent and vendor-neutral. We accept no payment from the vendors or categories covered, and this page names no products.
Caveats
SCR publishes no benchmark for packaging cost reduction, nor for the share of parcels billed on dimensional weight. Every figure SCR could trace originates with packaging manufacturers, on-demand equipment vendors, or parcel audit firms, none of which discloses a method.
The nineteen percent effect of a divisor change from 166 to 139 is an arithmetic illustration on a stated box size, not a statement about invoice impact across a population of shipments.
Carrier tariffs and service guides are reissued at least annually and surcharge thresholds change with them. Every figure in Table 1 should be verified against the carrier's current document before it informs a negotiation.
The European calculation methodology for empty space was not yet adopted at the time of writing, and the United States state programs were at different stages with active litigation. Both should be reverified before compliance planning.
Figure 1, Table 1, and Table 2 are structural and status summaries rather than measured research findings.
Where to go deeper
Readers scoping the systems that invoke cartonization should read the SCR guides to WMS, WES, and WCS and to order management and distributed order management. The returns and reverse logistics guide covers what happens when packaging has to survive an inbound journey and how disposition economics work. The Scope 3 and carbon accounting guide covers packaging emissions, which this page deliberately leaves to it. The last mile and transportation management guides cover the wider parcel cost picture, and readers scoping across categories should start with the SCR supply chain software category map.
Sources
Sources
- UPS. Tariff and terms and conditions of service, United States, 2026. Primary carrier document. Source of the published divisor and terms.
- FedEx. Service guide, 2026. Primary carrier document.
- Supply Chain Dive. Reporting on 2026 cubic surcharge threshold changes. Trade press reporting carrier announcements.
- European Commission. New EU rules on packaging enter into application. Primary regulator source for the application date.
- European Commission, circular economy portal. Packaging and packaging waste regulation overview. Primary.
- Latham and Watkins. Summary of the packaging and packaging waste regulation provisions. Interested-party-adjacent: a law firm advising on compliance. Used for the provision summary including the empty space article.
- Holland and Knight. Update on United States state producer responsibility laws. Interested-party-adjacent; used for state-by-state status.
- Mayer Brown. Producer responsibility packaging laws moving to compliance. Interested-party-adjacent; corroborating source on program status.
- Martello, Pisinger and Vigo. The three-dimensional bin packing problem. Operations Research, 2000. Peer reviewed. Establishes that the three-dimensional problem is strongly NP-hard.
- Garey and Johnson. Computers and Intractability: a guide to the theory of NP-completeness, 1979. Foundational reference for the complexity of bin packing.
