The instinct when work is going well is to add: another size, another finish, another variant. Each addition looks small at the point of deciding, and the cost of it only becomes visible much later, spread thinly across everything else.
A range that has doubled has more than doubled in effort. Every item needs its own description, its own photographs, its own stock decisions and its own answers when something goes wrong. None of that work appears on the item's own line; it is absorbed by everything already there, which quietly gets less attention than it had before.
The counter-argument is real: a narrow range turns away people who wanted the thing that was not made. That is a genuine loss and it should not be waved away. But it is a visible loss, counted at the moment it happens, while the cost of a wide range is invisible and paid continuously. Losses that can be seen tend to get overweighted against losses that cannot.
The useful question is not how many things to make. It is which additions would still look worthwhile if their full ongoing cost arrived on the same day as the decision.
