3 ms·

It was a hypothetical example. I know nothing about warehouse automation (though I know it is not using pad-and-paper as I assumed any intelligent reader would

It was a hypothetical example. I know nothing about warehouse automation (though I know it is not using pad-and-paper as I assumed any intelligent reader would understand I was making up a hypothetical world where warehouse automation didn't exist yet), but I know it's complex and I know that if a company with a lot of employees trains them in a complex system, the odds of them changing that system are very low. And, as I said above: It's not a monopoly. But, it is a comfortable place to exist profitably without a lot of risk of losing your customers. Also, you've wholly missed my point about complexity. Some problems are hard because of external, uncontrollable, unpredictable, complexity. Import/export is that kind of complexity. I'm talking about predictable complexity. Again, to come back to warehouse automation as an example, which you hate, but makes my point. A warehouse has a lot of moving parts (people, machines, inventory, shelves, whatever TF), but it is finite and is a mostly closed system. The inputs and outputs usually match up, and you can predict what's going to happen with ins/outs and all the people and things that make the things move in and out. It is complex because it is large and there are a lot of people involved and many of them would need to be trained to use any new system that automates it. It is not, however, chaotic, and involving many competing factions, like import/export.

hyperpallium 9y ago
Unpacking that a bit, training costs are to do with interface complexity (in the sense of "necessary" complexity) due to integration with other systems, and other information. i.e. it's not 100% automated; you can't just press a button. It's more "amplification", where you must be trained and practised to use it effectively - a tool for skilled labour. If you're super-successful, yours becomes a de facto standard, which others copy (usually, you still command a premium; but for the life-style idea here, we'd want to avoid that). This might be included as "installation" costs, along with and engineering integration with other information systems, Related is the organizational purchasing process, expensive for larger transactions: determining requirements, evaluating alternative vendors and solutions, sign-off from legal and upper management, and so on. Change itself is disruptive, there'll be teething problems, other systems will have to adapt, and other knock-on effects apart from training itself. These are one-off costs that you'd prefer not to incur again, unless there's a very large benefit (your example of 10x is worth it, but not (11/10)x). All are switching costs.

hyperpallium 9y ago

Unpacking that a bit, training costs are to do with interface complexity (in the sense of "necessary" complexity) due to integration with other systems, and other information. i.e. it's not 100% automated; you can't just press a button. It's more "amplification", where you must be trained and practised to use it effectively - a tool for skilled labour. If you're super-successful, yours becomes a de facto standard, which others copy (usually, you still command a premium; but for the life-style idea here, we'd want to avoid that). This might be included as "installation" costs, along with and engineering integration with other information systems, Related is the organizational purchasing process, expensive for larger transactions: determining requirements, evaluating alternative vendors and solutions, sign-off from legal and upper management, and so on. Change itself is disruptive, there'll be teething problems, other systems will have to adapt, and other knock-on effects apart from training itself. These are one-off costs that you'd prefer not to incur again, unless there's a very large benefit (your example of 10x is worth it, but not (11/10)x). All are switching costs.