Breakage in transit is not a mystery. It is one of three things: the cushion is too thin, the jar is moving inside the box, or the box wall is carrying more weight than its flute can hold. The fastest way to tell which one you have is a drop test, run in your own shop, before a customer ever files a claim.
The drop test is the cheapest lab you will ever own
The standard most of the freight industry runs its cushioning off is ASTM D5229, and the number that matters for a single carton is 30 centimeters, the drop height for a package under about 34 kilograms. ISTA, the transport simulation program most shippers quote, is more aggressive and more expensive. You do not need it to find whether your packaging works. You need the 30 cm drop, and you need to run it on the finished unit: the jar, the lid, the cushion, and the carton, all assembled the way it will actually ship.
Set up is a flat concrete floor, a wooden drop rig, or a pallet at the right height. Drop on the edge first, then the corner, then the face. Three units per orientation is enough to see a pattern. If the glass is intact, the lid sealed, and there is no wax on the bottom of the carton, the unit passes. If any of those three are off, you have a failure mode to name.
A single jar in an open box will pass the corner drop more often than a full carton of eight. The eight-jar box concentrates load and shifts the point of impact. Test the configuration you actually ship, not the one that is easy to demonstrate.
In our factory runs, the corner drop has failed the same way three separate times: a hairline crack at the jar lip, wax seeping through the bottom of the jar, and a lid that cracked but kept the wax inside. All three show up within the first 24 hours of transit. If your packaging passes the drop test and still arrives broken, the failure is not in the box. It is upstream, in the glass lot or the lid press.
Three numbers decide whether a jar survives
There are three variables you actually control. Cushion depth is the first. For a glass jar that is 9 centimeters tall, the foam or paper insert needs to hold at least 2.5 centimeters above the lid and 1 centimeter on each side so the jar cannot touch the carton wall. If the lid is loose in the cavity, the cushion is doing nothing, and the box is doing all the work.
Clearance between jars is the second. Two glass jars with zero gap between them will crack each other on any corner impact. The gap has to be at least the thickness of the filler material, and if your filler is paper, it needs to be the double or triple layer, not the single sheet.
Wall flute is the third. Corrugated board is coded by three digits, 33, 44, 55, where the first number is the edge crush strength in pounds per square inch. A 33 flute will hold one layer of cartons on a pallet and fail at two. A 44 or a 55 will carry the three-layer stack that most carriers expect on an un-stacked carton. If you are shipping 8-ounce jars in a single-wall 33 flute box and stacking three high, you are asking the wall to do a job the flute was not rated for.
These three are not independent. A deeper cushion lets you use a thinner wall. A tighter jar clearance lets you use a 33 flute instead of a 44. The drop test tells you which combination is failing.
The breakage rate is the number your supplier will not give you
Track breakage as a percentage: damaged units over total units shipped, per month, per carton type. A single broken jar is an accident. Five broken jars in a month of one carton type is a design problem, and it is the same problem every month until you change one of the three numbers above.
Cost per carton is the second number to track. Take the replacement cost of a broken jar plus the glass it came in, divide by total units shipped, and you have a number that is comparable across carton sizes and filler types. This is the number to put in front of a packaging vendor when they tell you the 44 flute is 20 percent more expensive. The 20 percent on the wall is a rounding error compared to a 3 percent breakage rate on a full order.
In our runs, paper filler has been the quiet cost driver. A single sheet of crinkle paper between jars passes the edge drop and fails the corner drop with enough regularity that our breakage rate on paper-filled cartons has stayed in the low single digits. Switching the same carton to a molded foam insert dropped that rate to under 1 percent over the same quarter. The foam costs more per carton. The breakage it prevents costs more per month.
Before you change the carton, change one variable and re-run the 30 cm drop. If the breakage rate does not move, the failure is not in the packaging, and no amount of thicker wall will fix it.