Why didn't you allow multiple people to play this together?
Why didn't you add this system to the entire game?
Why didn't you make this introduction mandatory?
Because I'd be losing something else on the other end of the game.
Because I would pay the price in subtle psychological differences that you may not consider as important as I do.
...Because I didn't.
I think one of the most important things for anyone interested in design to learn is that perfect designs don't exist. Design is about trade-offs. You gain something, you lose something else. You make one aspect of the game stronger, you make another one weaker. If you don't think you're giving something up, then there is an aspect of your design you aren't seeing yet.
Perhaps the only perfect design decision that can ever be made is the first one on a brand new game. That one is pure. That game is exactly what it intends to be. There are no other design decisions yet to contradict or complicate it.
Okay, perhaps this view sounds extreme. Let's start with examples.
Example: Investment vs. Potential Frustration
Let's give a game a Death Penalty.
Let's start off harsh, making sure that you take death seriously and try not to let it happen. When you die, you lose everything you've gained and have to start over from the beginning of the game.
But that's frustrating right? When it happens, some of your players are liable to quit outright and may not come back.
Okay, what if death isn't a penalty at all? Let's just bounce you back a few seconds before you die to try again, with no loss.
Now your players don't care about death. They charge recklessly into the same challenge over and over, dying repeatedly. They're no longer invested in the game, and over time, some leave citing the game as boring.
And you can spend all the time you want looking at all of the compromises in between, looking at a game without death, looking at a game that follows death up with entirely different results, and no solution you derive will ever be perfect. It's a trade-off. You have to decide which end of the spectrum matters to you more, and in what amounts. You can soften the design and supplement it, but you will never achieve both of those things in full-measure.
Example: Convenience vs. Social Bonds
Let's create a social game where groups of players overcome challenges together.
Let's start off by making it free-form. Present the group challenges, and let players find ways to group up and face them.
But players find it inconvenient to form groups. Some wait around hours and can't find others to play with. Others form groups only to have someone leave in the first 5 minutes, making them start over from scratch. Some complain and quit.
Okay, let's create a system that will group people together, with the right number and right composition, removing all of the hassle.
Many players find this incredibly convenient and compliment it highly, but now they're not talking to each other. They're not seeking other players out for activities, they're not remembering the names of the people they were paired with, and they're not forming lasting bonds in their play. They taper off and start leaving because there's no one they're invested in sticking around for.
And again, you can spend all the time you want trying to offer the convenience while cultivating the social bonds at the same time. You can try it in different proportions, or with different systems layered on top of each other to compliment weaknesses. But ultimately, you can't have both fully. You are forced to decide what is more critical to your game, convenience or lasting social opportunities, and aim most of your designs in that direction.
Example: Deep Story vs. Smooth Gameplay
Let's create an action-adventure game.
Let's start by giving it the best story we possibly can, and immersing players in it at all times. The characters are deep and nuanced, and interact with each other in fascinating ways that betray deeper feelings they have towards each other. The plot is original, turns at just the right points, and has amazing impactful climaxes. The end is surprising, but wraps everything together in that perfect way that only the best stories do.
But players start to voice concerns about the amount of dialogue. That they don't understand why the characters are acting the way they are. That they missed the plot change and are confused about what happened. That they can't figure out how to complete the game in a way that lets them see all the possible character interactions.
Okay, let's take an approach on the other end of the spectrum then. We'll focus in on a tighter, easier to understand plot. We'll touch on the critical aspects of it a few times, to make sure that people don't miss them. We'll give the characters personalities, but we'll keep them more background so that they don't overwhelm the player. We'll keep the dialogue focused and light, so it's noticeable when it happens.
Now players are following your story more consistently, and speeding through it smoothly. You start to see less players posting excitedly about your characters, and less players writing fan-fiction about their backstories. Players consume the game far more quickly, and leave it behind once they've completed the main story-line. You get less follow-through from players when you release additional expansions of content further down the line.
And no matter how much time you spend trying to land the perfect middle ground, it doesn't exist. Some players will always be overwhelmed by your story. Some players will always leave without being as invested and excited about your story as they could have been in another mythical approach you might have taken. You have to choose what matters to your game.
And on and on...
Add a new ability, and you add complexity for the player to learn.
Place a new spawn, and you lengthen the time to completion, losing some attention along the way.
Add an awesome new vista, and you distract from the clean central focus the area previously had.
Nothing is free. Everything has a cost. Some costs are small and smart. Some are much bigger than you think they are, especially when you start dealing with social components and overlaying systems.
How to Design in an Imperfect World
This isn't all doom and gloom, it's just important in game design to recognize that you're making these trade-offs. Some are big and obvious, some are small and psychological, but you're always trading something off. Even in the best case scenario, you're favoring one kind of audience over another.
The way you manage this is by knowing what matters. Don't act on feedback blindly. Know what makes your game fun, and use that as the lens you look through to decide what you change and what you don't, what matters and what doesn't.
A great way to do this is by identifying the core tenants of what make your game great. Is it a world about immersive exploration? Or a world about social play with friends? Is it a game that non-gamers can get into as their first game, or is it a game that aims to be deeply interesting for a more experienced audience? Is it a game that is meant to last 10 hours or a game that is meant to last 200 hours?
Set your stakes in the ground, and use them to guide your toughest decisions. Does the trade-off make the core focus of your game stronger, or does it hurt it?
And when another designer inevitably challenges why you didn't design the game the way they think is right, you can obnoxiously respond, "...Because I didn't."
Or I suppose you can have a constructive conversation where you explain your reasonings. That's okay too.
[My thoughts and opinions are my own. They are not those of Blizzard Entertainment, and they do not necessarily represent Blizzard design philosophies.]