I warn you, though, that this is fairly lengthy, so read on only if you're really interested in what makes a good role-playing product.
It's impossible to define precisely what makes a good RPG product, since that's obviously going to change based on the intention and the nature of the product in question. There are, however, several specific qualities that I believe all good RPG products must have; some of these will be more important for some sorts of supplements than for others, but they must all be present to one extent or another.
Utility: This is, bar none, the most important quality-the product must be useful. It doesn't matter how interesting the story being told may be, or how fascinating the concept or the idea; if the product isn't useful to those who are trying to play the game, the author would have been better served in writing a short story or a novel. If the reader's response is "It was cool to read, but I'm not going to use it," the writing may have been good, but the product itself has failed.
Interest: It should go without saying, but the subject of the product must be interesting. A mechanic for determining the current state of the garden in the courtyard of a castle is probably not going to interest anyone who isn't already fascinated by horticulture (unless it is presented as an actual plot point, of course). Some things simply aren't worth writing about, or including in a game.
Adaptability: A good RPG product must allow for multiple different styles of play, and multiple different GMing/DMing/Storytelling techniques. A module that must be set in a particular campaign setting does not possess adaptability; one that is best suited for a particular setting, but can be easily altered to fit others, is a better option. Along those same lines, a supplement designed, for example, to provide a new setting-let us say a city for a fantasy RPG-must allow for all sorts of games to be run in that city. If every plot hook and story concept in the product requires the players to slay the monsters/NPCs of the city, the book does not allow for groups that prefer political intrigue or solving mysteries. On the other hand, if none of the plot hooks can be resolved with combat, the product is inappropriate for players who prefer action-oriented games. A product must be adaptable for multiple styles of gaming.
Creativity: Nobody wants to pay money for a story or a setting that is the same old thing in different wrapping. A good RPG product must offer something new-a new story concept, a setting that is put together in a new way, different ways of thinking about character creation, or even just a strange twist on an old idea. If the reader finishes the product and feels that she's seen everything she just read someplace else before, the product was not sufficiently creative.
Flexibility: A close relative of Adaptability, Flexibility is more of an issue for story-oriented products (such as modules). The story must allow for multiple means of resolution. Though it can be set up to encourage the players to make specific choices, it cannot force them to do so. Any story that requires the GM to railroad players into behaving in a certain fashion, or into making specific choices, is poorly written. There are very few experiences players find more frustrating than the sensation that they are not being allowed to make their own decisions. Similarly, character creation/growth should allow as much freedom of choice as possible. Random rolls are acceptable to determine attributes, hit points, and the like; but they should be avoided everywhere else. Let the player choose the details of her character; never force them on her.
Focus on the PCs: The PCs, and thus the players, must be the focus of the events in the story. Do not require them to stand by and watch while NPCs duke it out; do not put them in a situation from which a powerful NPC will have to step in at the last moment and rescue them. In a novel or a movie, it may be well and good to have the main character dramatically rescued from certain death at the last moment, but this technique does not work in a role-playing game (unless the players are the ones doing the rescuing). It robs the players of any sense of accomplishment, and frustrates them almost as much as being denied the right to choose their own course of action. A product that does not focus the story on the players fails as an RPG supplement.
Inspiration: Most modules are self-contained stories, but all other supplements-setting books, books on monsters and characters, and the like-should inspire the GM to come up with his own ideas. Even story-oriented products should include concepts scattered throughout that an enterprising GM can use to spin his own plotlines. Products that force the GM to use only the specific plot concepts offered, without allowing room for original stories, are too restrictive.
Understanding of the Rules: A gaming product should make full use of the rules for the game it is intended to supplement; more to the point, it must show an understanding of those rules. This is particularly important to a D20 product which is not released by Wizards of the Coast. For example, the D20 system uses a skill check system for determining success and failure. A product should use that mechanic, and no other; saying "There is a 25% that the guard will spot the PCs," when such a thing should be determined by a Spot Check, shows a lack of understanding of the rules.
Professional Appearance: It doesn't matter how fantastic the content of a supplement might be if the product's presentation is haphazard, and an amateurish or clumsy image can destroy a company's reputation. This is not to say that all products must be perfect; typos and errata are going to appear in even the most carefully checked work. But if a product has glaring typos or inconsistencies, and has obviously not been proofread or double-checked, it is not a good product-and probably not a good company.