It's more complicated than that. It's literally that sometimes two of the exact same item last for radically different times. It's not a different design or manufacturing process, just an amorphous series of random factors lining up we call luck.

Mean time between failures is something they do actually measure in manufacturing, and you see interesting results like what hard drive manufacturers do to increase reliability: stress test the drives until the ones destined to fail early fail, and then sell the others.

There are things that can increase reliability, but a lot of the things that make the extreme outliers are just random, and no one documents what they were because they didn't know it was going not have an effect, good or bad.

Well, sometimes, actually, yes. Unfortunately you have to do the research to figure out if anyone does it right and the people doing it wrong tend to out-marketing spend the good ones.

For example, house came with a garbage disposal. Within 5 years it rusted through, because why would you bother with stainless steel in something that is constantly soaked with water? Did research, didn't have to spend too much more to get one actually made using stainless steel and that one is still going 15 years later.

Of course sometimes you just don't have options. Like if you want a microwave, it doesn't matter what microwave you get from what brand, all of them are just cosmetic variants/brands applied to the exact same microwave made by one company in China.