The UX of Hades: Why we keep running from hell and back

What makes this modern videogame classic so sticky?

I’ve spent 240+ hours playing Supergiant’s Hades. I’d be more embarrassed if it wasn’t so well-loved by pro and casual gamers alike.

As I’ve been playing recently, I find myself lingering on a burning question. I’ve finished the main campaign, completed the epilogue, wrapped up all side quests, fawned over all the impeccable art, memorised the soundtracks —

— so why am I not bored yet?

What keeps me coming back to this game one year after I first played it?

I’ve been thinking about what makes games sticky, what that says about human behavior, and how to apply those tidbits to building sticky experiences.

How Hades works

But before we talk about what makes Hades sticky, we have to first talk about the rules.

You play Zagreus, the reluctant prince of the Greek Underworld, as he fights monsters to escape. The catch: you can’t save your progress. If you die mid-escape, you go back all the way to the beginning.

Your challenge is to get out of hell in one shot.

To escape the Underworld, you must lead Zagreus out of four main regions. In game design, they call these gauntlets.

In each gauntlet, you must go through 10–12 rooms of monsters to fight.

Almost all rooms and their monsters are randomly generated. For 90% of the game, you can never tell exactly what the next room contains.

In all rooms, a sequence of four events happen in about 30 seconds:

1. You enter the room
2. You fight monsters
3. After clearing all monsters, you earn a reward that aides your escape
4. You choose between 1–3 rooms to proceed to, based on the reward they promise. Rewards are randomly generated.

What makes Hades stick

It’s a simple premise that attracted legions of loyal fans and swept awards during the pandemic. Thinking about why that’s the case has taught me about what we stick to and why:

1. 😲 Novelty: We cling to creature comforts that provide just enough novelty
2. 💪 Control: We seek challenges when we feel like we are in control of them
3. 🏆 Quick rewards: We like taking risks we are quickly rewarded for

😲 We cling to creature comforts that provide just enough novelty

Taking in-game screen caps for science

I made sure to bold every mention of random generation, because I can’t emphasise its importance enough.

In each room, there are at least 2 avenues of randomly generated change:

1. What monsters you fight. Let’s say there’s 5 variations per room.
2. What rewards you are given to choose from. Let’s say there’s 30 variations per room.

So each room has at least 150 versions of itself. The average game has about 40 rooms.

So, the average game has 6,000 possibilities — at least! Because this doesn’t yet include the avenues of change that you can choose: 6 weapons to choose from, 20+ power-up tokens, 15+ challenges you can set for yourself.

At scale, this means that what worked before in one escape attempt, or one region, or even one room may not work in the next one.

With just two variation points and a simple, well-defined set of game rules, no two games ever feel the same.

How you can apply this:

👉You don’t need too much. Even one point of variation with a substantial dataset can do the trick. For example: in an e-commerce app, 5 randomly generated home page sections might make it difficult for users to find things again. However, one “Related Items” section per product page that shows 5 randomly generated items with the same criteria gives users something new and relevant to look forward to.

👉 Ensure you maintain your structure and core workflows. Notice that none of Hades’ core game rules ever change. Avoid adding variation to your core workflows or workflow outcomes. This will cause your system to act differently from how users expect for no reason, which is a recipe for user dissatisfaction. Rather, add variation to the possible choices you can make, or people you meet along the way. In the previous example — a “Related Items” section with randomly generated items doesn’t change a buyer’s workflow, yet still provides meaningful novelty.

💪 We seek challenges when we feel like we are in control of them

Master of your fate, captain of your soul

Imagine if this happened in each room instead:

1. You enter the room
2. You fight monsters
3. After clearing all monsters, you earn a reward that aides your escape
4. You can’t choose between 1–3 rooms to go to next or what reward you get — you just go through the next randomly-generated room chosen for you.

This means that, for the whole game, you must accept the reward in the next room and you must accept the level of monster difficulty.

Now, not all rewards are created equally. Neither are the monsters. So you must accept whatever cards the game has dealt you — regardless of how useful the rewards are, or whether you can handle more difficult monsters.

If we are dealt two cards, at least we can say we chose what felt best at the time. But if we are dealt one bad card, that just feels unfair.

The element of choice is paramount in Hades. It gives you a sense of control and ownership over your escape journey. It elevates the game from “roll the dice” to “choose your own adventure”.

How you can apply this:

👉 Allow users to opt-in to any settings designed make using your app harder or easier. People who seek out games/gamified apps have intentionally challenged themselves. Give them the option to challenge themselves more and trust that they will seek it out when they are ready (or dial it down when they need to take it slow).

👉 Calibrate your challenges so that difficulty doesn’t jump too fast. I like to run, but if I was forced to choose between walking for 5 minutes or running a marathon, I would definitely pick the 5 minute walk — I would just be very unhappy about it. Running a marathon is too daunting for the average person, so the decision feels less like a choice and more like defaulting to the only feasible option. When setting up intentional difficulty, make sure the next step is difficult but still feasible. Testing different difficulty levels with different users will help you find the sweet spot!

🏆 We like taking risks we are quickly rewarded for

I wish I got rewarded for all my runs like this

Something that the diagrams don’t capture is that the game is very quick. With each room taking only an average 30 seconds alongside the great art direction and sound design, it feels like lightning speed.

This means that in the span of 30 seconds or less, you fight off all monsters, earn your reward, and select the next one. It’s a lot of positive feedback per risk in one go!

While it’s true that people who seek games have intentionally challenged themselves, we have to remember that they do so for a promised reward.

So keep them coming quickly — it’s what turns “I’ll stop after the next room” to hours-long gaming sessions.

How you can apply this:

👉 Shorten the feedback loop between user choice and reward. If a user learns that the reward of a choice comes quickly, they are more likely to execute the choice. This can be done in at least two ways:

1. Reduce the amount of steps from choice to reward. Minimise any roadblocks that do not contribute to your overall user goal.
2. Decrease any waiting or load times. Even Google will rank websites higher on search engines if they have a quick load time. Work with your developers and content managers to design around simplicity and smaller file sizes. Your users will be able to tell the difference.

I wrote all this hoping to figure why I can’t stop playing this darn game. However, I think this exercise also gave me a greater appreciation of the strategy and execution of game designers. Knowing the theory is one thing; putting it into practice in another. And I hope spelling it out can help us put these theories into practice to the same elegance and effect.

Up next — what can your dog teach you about UX? what can the competition between running apps Strava and Nike Run Club teach us about user journeys? I’ve been thinking a lot about these things lately and have started to pen my thoughts down. Follow through to get updates when they’re published!