There is most certainly an aesthetic principle at work in the selection of characters and incidents [in La Gioconda] -- though an aesthetic principle which we have come to regard as foreign to serious art. It is, approximately, that the course of a work should be determined by whatever adds up to the most effective sequence of events, rather than by what is organic and necessary to the donnees of character and situation. There is a great deal of this in nineteenth-century Italian opera; people show up at the most improbable places and times simply because the theatrically effective thing is to have them show up...
Among the things judged essential in this sort of opera is the presence of set numbers and arias -- in this case, a major aria for each of six principal singers, plus an extended ballet and several big choral episodes. This means that the elements will again be manipulated in such a way as to provide these things in appropriate proportions and spacings. This is, it is probably needless to say, also true of works that quite seriously pretend to greatness, the difference being that in such works, the author is at great pains to disguise his "pacing" and "structure," to make it one with the logical motives and actions of the characters. In Gioconda, as in many earlier Italian operas, there is hardly any such pretense: now comes a mezzo aria (the tenor leaving the stage for no good purpose other than that of letting her sing it), now comes a Confrontation Duet, etc., etc.
The passage I have bolded still seems to me a great explanation of one of the differences between works that achieve greatness and those that don't really achieve it (or even aspire to it). Almost all commercial theatre, television, film, novels, use certain tried-and-true devices. And nearly all of them space out the use of these devices to the moments when they will be most dramatically effective. Musicals need spots for comedy songs, ballads, an "11 o'clock" number; movies and plays need certain set-pieces or types of scenes depending on the genre; TV episodes need to incorporate certain tricks to keep the audience's attention through the episode and especially during breaks.
The distinction between first-rate works and merely good or effective ones is not that the former doesn't use those proven, familiar devices. It's that the former tries to make those devices seem like an organic part of the story. It wants to convince us that a certain thing happens not because convention requires it, not because this is the moment in the evening when a certain type of response needs to be obtained from the audience, but because the characters would logically do this at this point. Of course they're also doing it because convention requires it, because the actor/singer needs a showpiece, and many other reasons. But the writer is trying to hide this and make everything seem natural. If the writer does not succeed in making the events seem like they are driven by the story, and instead makes it too obvious that the story is constructed around the set-pieces and tricks, then the result may still be something entertaining and good. But it's probably not going to work on the highest level
So what Osborne says about two operas that he considers great works of theatre, Cavalleria Rusticana and Pagliacci (two Italian operas often accused of being lowbrow, but which are superb examples of theatrical writing where the big moments seem to come naturally), goes for first-rate writing of any kind:
[They] make use of all the formal conventions... carefully breaking them at key points to give the impression that they are not controlling factors.
But an entertaining middle-of-the-pack work, whether it's La Gioconda or an episode of a fun but cheesy action TV show, we know that something will happen at a specific point because that's what always happens in this kind of piece. The seams are visible, and we can see that we're being toyed with. If it's well done, we'll enjoy it anyway, but it's likely to fall short of greatness.
And then you have outright bad works, where we are able to see the conventions driving the story and they're not even well-executed enough to make their effects.