Bluntly, few approaches will result in automatic success, so having a good skill bonus and using approaches that leverage those skills is immensely beneficial when the very likely skill check is asked for.
I don't know how often I have to say that approach is used so I don't have to assume anything, to set DCs appropriately, and to set up appropriate consequences. Apparently, it's at least once more.
"Bluntly, few approaches will result in automatic success, so having a good skill bonus and using approaches that leverage those skills is immensely beneficial when the very likely skill check is asked for."
That seems to be lacking necessary info to be a necessarily valid conclusion or even a meaningful description that tells us anything. That is I think part of the communication issue here, very ambiguous claims.
So let me ask, did you mean "few challenges will be seen in play to be resolved by auto-success approaches" or did you mean "challenges will have only a few approaches that lead to auto-success"?
If there are only 1 or 2 approaches per "challenge that matters" that produce auto-success without checks and character stats coming into play, but lots more of approaches for each that would require checks too, you could still see every single "challenge that matter" resolved by players choosing those 1-2 autos never once having to "leverage those skills".
That's why I broke it down to how often do you as GM setup (or do your players see resolution in play) of the challenges that matter having (being solved by) non-skill auto-success "approach wins" vs "leverage skills checks immensely wins?
Might be good to separate out vombat challenges thst mstter from non-combat, given even some of the "approach wins advocates seem to see combat challenges as "uncertain due to the whole roll-a-1 thing.
My baseline estimate in my games was roughly
60% require/resolved-by some form of checks (possibly passive, possibly auto-success due to stats (DMG proficiency.), approach/method reflected by advantage/disadvantage)
20%* require/resolved-by either some form of checks (as above) *or* some key approach/methods.
20%* require/resolved a combo of checks (as above) *and* some key approach methods used in tandem.
Practically none of my "challenges that matter" are resolvable by method/approach alone regardless of stats.
* Honestly this 20-20 varies and is more WAG than anything else. Not more than 30-10 or 10-30 I figure.
So, this means my folks see in play when it matters that most of the time the charscter's skills that the players chose to highlight and the player's in character choices go hand in hand to reach an outcome. They see in a few cases both are needed and in a few cases they can get by on method/approach slone.
That sounds like a very different thing and gameplay outcome than one gets when an "approach gm" goes into how it's actually "cheating yourself" to go mostly with skill checks for outcome as a player.