How can a moderated-mediation exist without mediation or moderation?

[Yuri]

So I'm running this analysis with several variables conceptually this is my model:
So here we have A predicting B, both with a direct effect, and with a indirect effect through M, where W is moderating.

Here I have a moderated mediation, also known as conditional indirect effects, where the effect of variable X on an outcome variable Y via a mediator variable M differs depending on levels of a moderator variable W. Specifically, either the effect of X on the Y, and/or the effect of Y on M depends on the level of W.

Now in my analysis the moderated mediation is significant.

However:
a, the moderation on it self is not significant.
b. Neither is the mediation on it self.

BUT the moderated-mediation is!

How it this possible?

I'm not sure how to explain these findings...

Best,

Yuri