saltburnman said:
Isn't it by definition already a black hole
Another way of answering this, and the OP question in this thread, is to note that the definition of "black hole" you are implicitly using is not correct. A black hole is
not "a bunch of matter that is inside the Schwarzschild radius for its mass". It is "a region of spacetime that cannot send light signals to infinity".
In the Schwarzschild spacetime geometry, which describes a compact, central massive object surrounded by vacuum, "sending light signals to infinity" means something like "the light signals can get as far away from the central object as you like, if you wait long enough". (The more technical definition is that the light signals end up at future null infinity, and a black hole is a region of spacetime that cannot send light signals to future null infinity.) And it turns out that the surface area of a black hole region of spacetime in this geometry will be ##4 \pi r_s^2##, where ##r_s## is the Schwarzschild radius corresponding to the mass of the hole. So it makes sense to think of the hole as confined within a region of "size" equal to the Schwarzschild radius. (Note, however, that the hole is vacuum inside; it is not "made of matter". It's made of spacetime curvature.)
In the expanding FRW spacetime geometry, which describes our universe as a whole,
there is no infinity to send light signals to. (The more technical statement is that there is no "future null infinity" in FRW spacetime the way there is in Schwarzschild spacetime; the conformal boundaries of the spacetimes are fundamentally different.) So, strictly speaking, it is impossible to have a black hole at all, according to the technical definition of that term, in FRW spacetime. The best you can do is to have a region of spacetime that, locally, looks like a black hole. And "looks like a black hole" means "is approximated well by a piece of the Schwarzschild spacetime geometry", which means it has to be a compact, central massive object surrounded by vacuum.
In our current universe, there are plenty of objects, like planets, stars, or, if we are willing to accept a coarser level of approximation, even galaxies or groups of galaxies, that can be approximated by a piece of the Schwarzschild spacetime geometry. And of course that set of objects includes the objects that we ordinarily call black holes.
But in the early universe,
there were no such objects. There was
no region in the early universe that could be approximated even coarsely by the Schwarzschild spacetime geometry. The main reasons for that were that, as has been noted, the early universe was rapidly expanding, and, combined with that, that the early universe was very uniform--there was no gravitational clumping the way there is now. The matter in the early universe was a uniform fluid at very high temperature (and, as noted, expanding rapidly), with no boundary. The Schwarzschild spacetime geometry is simply nothing like that, and cannot describe that.