The box.
That sound like a stupid thing to say, but hear me out. Somebody has to make a box. Said box is usually made by taking paper, printing, adding the fluting (the wavy bits), gluing, shearing out the form, and then shipping hundreds of flat boxes on a pallet. The two largest expenditures are the printing and the shearing, because the hardware doing it is complicated and needs to be manufactured. As such, their goal is to make as few types of packaging as possible, to minimize the huge financial outlay. So, you might have a half dozen SKUs of processor, but you want to minimize the cost of packaging because it's not considered value add to the customer but must be present.
For example, we had a customer decide that they were going to change the placement of a barcode label on the box. They said this was absolutely necessary to make it easier on their staff, and my company handed them the bill to do it. They laughed at the number, and we had to explain that the MOQ was about 6 months of product, that they had vastly overestimated sales figures on and was actually closer to 8 months of packaging. They then asked why the supplier had such a huge associated fee, and we broke down the machinery cost of rebuilding the screen printer that they had specified be used...because the cost per unit was so low. Once they'd consumed all of the box inventory, and had the screen print rebuilt, it was over $100k to move two labels 6". That's why the boxes are the same...because the wasted space without a cooler is much cheaper to them than the frankly silly amount of cost wasted on customizing packaging to account for minor differences.
It sounds silly, but by the powers that be it's actually much more expensive to have a slightly cheaper box that introduces a new SKU versus simply using the boxes already made and shipping empty air.
. That would have been an even worse decision, making it unnecessarily harder to replace a faulty/lost power brick, or preventing the creation of convenience like a power station.
