Some people already said it, but an RPG is a game which is entirely built around some sort of progression system (through levelling up, equiping items, unlocking skills or whatever). That’s its staple, what really defines it as a genre from a design perspective. The rest is there (or should be there) just to enhance the former, although some features are traditionally more common than others.
A narrative, for example, gives that progression system a context and helps making it more meaningful, and when done really well it can even add to it, by making the character(s) evolve in more refined ways than just tweaking numbers or options. But there can be good RPGs without a story.
Combat is usually the way the progression system is realized, because it’s the most straightforward "excuse" to make the player grow (generally by obtaining experience or loot when defeating enemies). But there isn’t a particular style of combat that an RPG needs to have in order to be an RPG —choosing one over another merely results in a different subgenre, like action RPG or tactical RPG. And, of course, it’s entirely possible to make an RPG whose character(s) evolve through means other than combat, although it’s very infrequent because it’s harder to think of something that ticks that box while at the same time works well as an interesting and satisfactory mechanic (and thus it’s more risky).
In other words, you can take away the narrative, the combat and all the things that are usually taken for granted in any of the multiple RPG subgenres, and if the game still keeps a progression system as the core of its gameplay loop, then you still have an RPG. But take away the progression system from a game like (for instance) Chrono Trigger, while leaving intact all the other elements, and what you have right there is no longer an RPG —it’s an adventure (in this case). And, if a game has some sort of progression system, but it's not the core of its gameplay, then that game is only partially an RPG: enough to call it that or say that it has RPG elements, but not entirely.
So, taking that into account, what you can do to enhance an RPG is designing the progression system before anything else. That means that you must be able to test it and refine it in isolation, ideally polishing all the details in the process (not necessarily in an engine, it could be in a different platform or tool, or using any other method that is accurate enough). The result should be a robust progression system that feels rewarding and gratifying, and is in itself an interesting game loop. In order to do so in a competent way, of course, you need to know very well what you’re doing, so acquiring a deep knowledge in the matter should be the first step if you really want to go that route. If you can do it, however, it’s quite probable that, while designing the progression system, you come up with some interesting ideas about how the rest of the game should be, and those ideas will most likely intertwine in a very natural way with the progression system that you just created (since they were born from it), making the whole game way more cohesive as a result.
As for what can make the game worse, a difficulty curve is quite tricky to execute correctly and can hamper the overall experience if done badly. That’s true for all genres, but it’s arguably worse in RPGs because of grinding: a particularly sharp spike in difficulty in other games can feel unfair or frustrating for a while, which is generally bad, but that same sharp spike in an RPG has the potential to lead to hours of tedious and boring gameplay, and this description is perhaps the worst thing that anyone can say about any game loop. Besides, the progression can easily start to feel artificial if you’re forced to do the same repetitive tasks over and over just to move forward, and since the whole core of an RPG is the progression system, a badly designed difficulty with lots of mandatory grinding can make the whole game fall apart very quickly, causing it to feel like a chore. Grinding, of course, can be an integral part of the progression system, but, if that’s the case, it needs to be done in a way that feels natural and interesting, organically implemented, and not just some sort of secondary effect caused by an invisible wall between the player and the fun. And, if you want to create challenges in some parts of the game, but don’t want the players to just grind their path to success, you have to not only make sure that the players have the necessary tools to beat the challenge at that point without any backtraking or side-questing, but also ensure that they actually know that they can do it and that that is in fact the intended way to play that part of the game (whether they then decide to do it that way or not). And transmitting that information to the players is way harder that what it may look like, specially in a genre where many people are pre-conditioned to grind even before they start playing. (Note: making parts of the game way too easy is just as bad as making then way too hard, and designing a way too linear difficulty curve is also equally bad.)
In any case, I strongly advice you against making an RPG, unless you don’t have any intention of developing more games after that and you are just doing this as a long-lasting hobby on your free time. The reason for this is that the most likely scenario (EXTREMELY likely) is that you won’t get close to finishing it. And, even if you do it, many many years will have passed and the game will not be anywhere near the quality that you expected it to be at the beginning. So, as a piece of advice, do this only if it’s a pure hobby and you don’t mind finishing it or don’t have any concrete expectations. If that’s your case, then I strongly encourage you to keep going (it's really fun!) =) and you can completely ignore the following spoiler box.
If you’re somewhat more serious on this, it’s way better to start creating arcade-y (or extremely small) games that might not even be fun to play, but which will be something much more important that than: complete. They will give you the full experience of developing a game from start to finish, even if just at a small scale, and a sense of fulfillment and motivation when you actually have them done. They will also give you a better understanding of the time and effort it takes to make every aspect of a game and will eventually become either a nice portfolio or a good starting point to grow as an indie developer. And, more importantly, you will learn in a much smoother way than by making a single, longer game, because if you get yourself into a long development without any prior experience you’ll often find yourself refactoring code, redesigning assets or rethinking how the flow of your game works, instead of creating new stuff. If you instead devote yourself to creating smaller games, you'll still see what a mess your first code was or how ugly your graphics were or how dull and boring the gameplay was, but in that case it won’t matter at all, because, by the time you realize that, you will already be done with it and want to move on to your next game, where you’ll have the chance to put into practice all you have learnt with the previous ones, until some (distant) day you finally feel like you have the necessary experience to consider making an RPG. And even then you’ll probably need to hire people to help you either with the programming, the graphics, the music... or all of that, because, generally speaking, RPG is one of the most expensive and time-consuming genres that you can make. That will depend on your scope too, of course, but it's definitely not the easiest to pull off.
Also... My experience tells me that, when someone has an idea for an RPG, that idea is either more vague than they think it is or just too unrelated to game mechanics. In fact, in many cases it is actually NOT an idea for an RPG —it’s rather an idea that they have adapted in their minds to be an RPG (because they like games and specifically RPGs). And that is often a really bad starting point for any game development. That said, I’m not saying that this is your case, just that it is usually the case with unexperienced people who want to start developing games by developing an RPG.
In the end, if you really feel like that’s what you want to do, just do it. =) I’m not saying this to demoralize you or something, but I think it's important to be aware of the potential risks or problems that can arise from such an endeavour.
Anyway, if one day you publish this RPG (or a different, perhaps smaller, game), I'll be curious to play it if you share it.
I'm mostly a lurker now.
