The term MVP is often used to justify shipping less. Fewer features, a shorter timeline, a rougher edge. The first word gets all the attention.
We read it the other way round. A minimum viable product has to be viable first: it must solve one problem well enough that someone would choose it over what they use today. Minimum only describes how little else it carries.
That changes how we scope. We pick one job the product does and one audience it does it for. Everything that does not serve that job waits, however good the idea.
Inside that scope, quality is not optional. If the core task is slow, confusing or unreliable, the test tells us nothing about demand. People leave because the product is weak, not because the need is missing.
A viable first version gives cleaner signals. When people use it, come back and start asking for more, we know the need is real and where to build next. When they do not, we know it was not the craft that failed.
That is the product side of one: not the smallest thing we could ship, but the smallest thing people actually use.