Challenges In Designing A Casual MMO (Free Realms) – GDC Austin 2009

[THIS POST ORIGINALLY APPEARED ON ADRIAN CROOK’S FREETOPLAY.BIZ SITE.]

Laralyn McWilliams (Creative Director, Sony Online Entertainment)

Designing for a casual MMO.

  • Differences in designing a casual MMO
  • Play sessions are shorter – as short as 5 mins
  • Competition – lots of distractions… new games for this group every single day
  • Skill set and skill level – way different from traditional players… don’t spend hours playing games like we do

Change the way you think

  • Keep the focus on the players
  • Think outside the box
  • Don’t base everything on what you like or prefer
  • Don’t rely on your own judgment over UX testing
  • Question the “way things have to be”
  • Understand your audience well enough that you can speak your audience’s voice. I am speaking as the player here, not what I think – what I think doesn’t matter.

Theory, Practice, Results

  • Design team would have been better off if we had this process from the beginning.
  • Drivers and passengers – analogy of a tour bus experience. You have a captive audience that you want to entertain.
  • Designers are drivers. We are planning where people go. Plan the route, equip bus, pick stops and sights, determnine cost, provide entertainment along the way.
  • Passenger is in control. It’s his trip, money and time. He can get out of the car anytime he wants. He can blog about how your trip sucks. He can never ride with you again. Enough unhappy players will shut you donw.

Focus on passengers:

  • Identify who the passengers are
  • Set guideposts – how do I know I’m on track
  • Look at competing tours
  • Clear the path – get rid of your assumptions
  • Design the passenger’s experience – control exp start to finish
  • Head out on open road

Identify the passengers

  • Who are we trying to attract and entertain
  • How do they spend money: Pre-planned vs impulse (boxed vs online), small increments vs large purchases – determines pricing bundles, retail vs download – assumption is retail is higher value… people may buy online because they saw it in store, convenience vs function vs vanity (maple story has nailed this distinction)
  • What are they interested in?
  • What are they watching on TV?
  • Understand your passengers
  • Don’t base decisions on yourself or your own family
  • No matter how normal or typical you feel, you probably aren’t – we like BSG, general populace likes Desperate Housewives

How to get info

  • Zandl Group Hotlists
  • Iconoculture
  • Others

Identify the passengers

  • Free Realms player is boys and girls equally
  • 10-15 years old
  • Secondary is casual players and parents and family
  • Example of Zandl Hotlist – provides demographic insight into what they are doing, wearing, etc.
  • Pets in Free Realms can wear outfits. For every bionic dog there is a princess cat. Enough options and people feel welcome in your game.

Set Guideposts

  • Short set of goals – guideposts or landmarks. Team goals.
  • Check them with every decision you make
  • Include development goals, when it’s good for players – ex need really robust tools for designers to build content – good for players… equals lots of new content for players. Making content for impulsive players could be key for you.

Free Realms Guideposts

  • Virtual world for teens, tweens, and casual players
  • Support four play styles: Adventure, mini-games, simulation gameplay (housing, pets), Socialization
  • Quick to start – get into game very quickly as first time and returning user
  • Easy to understand
  • Rewarding to play – a lot of online games have you working toward longer term goal
  • Never assumes based on age or gender – won’t assume your girl character will start as a cook… can alienate ppl and close off market

Free Realms Key Design Decision

  • Create a game with a wide variety of activities that are all optional but all equally rewarding
  • Players do what they want to do and feel like they are playing a game made just for them

Free Realms Interaction Reward Cycle

  1. Player Need (Wealth, relationship, personal skill, etc)
  2. Interaction (NPC, object, Mini-game, etc)
  3. Reward (money, friendship, leaderboard, pet level up, etc)

Assess competing tours

  • Tours that explore similar landscapes in terms of demographics
  • Free Realms competing tours are: Runescape, Maple Story, Habbo, etc
  • Other games: EQ, WoW, Animal Crossing, Nintendogs, WCIII, DotA, Viva Pinata, The Sims, Gears of War, Cooking Mama, Puzzle Quest, Gears of War (reload mechanic will show up as a part of housing)
  • Inspiration comes from anywhere

Clear the Path

  • Discard assumptions you may have about how to make game your making
  • Look at assumptions from passengers POV
  • Analyze each feature
  • Challenge every assumption
  • “In writing you must kill your darlings” – William Faulkner

Flow Chart

  • Is it fun?
  • No? Is it essential? – No, Get Rid of it; Yes, Improve it!

Shot of Disneyland theme park map – they are the best at controlling experience. Lot to learn from that.

Design the Passenger’s Experience

  • We use SCRUM on Free Realms
  • Traditonal view of user story: written desc used for planning, conversations about the story, etc
  • Should be written from player perspective (“don’t want to run out of inventory space”)
  • User stories as the foundation
  • Orient design docs toward user stories… inventory design doc could start with user story
  • USer stories + solutions + implementation details + game design
  • Design the passengers experience

Servers

  • Assumption: a character is locked to a server
  • Sucks for my friends on different server – sucks when server is down
  • USer story: want to play with friends on diff servers
  • User story: want to play when server is down
  • Other games, Runescape, Wizard101
  • Solution: Play on any server, any time
  • Hindsight: should have included server transfer while in-game before launch
  • All languages + all servers = new problems (great from player perspective… but lots of delay in translation)
  • Need server recommendation logic – take server select out of experience… to do that you need a system that recommends a server based on where their friends already are
  • Jumping to a friend may mean a long download – if you stream your content… if you haven’t been to your friend’s part of the world, you will spend time waiting for it

Classes

  • Assumption: a character is locked into a class at character create
  • Problem: I can’t understand the choice before  play
  • User stories: As a player, I don’t want to make a long term choice before I play
  • User story: As a player, I want to try different classes
  • Solution: jobs
  • Hindsight: too many choices, all at once
  • Making a job cool = lots of investment in that job = expectation that subsequent jobs will be that robust
  • Putting job choice up front in new player experience
  • Create a stronger link between fave jobs and identity
  • Gleam and Gloam are really important to give a sense of purpose

Inventory

  • Assumption: have to limit inventory space… buy additional space.. inventory tetris is good
  • Problem: I have to decide what to throw away, I have to spend money to carry things I earned
  • User story: don’t want to have to choose to destroy something to pick up something else
  • Other games: none
  • Solution: unlimited inventory
  • Going to buy items in Free Realms is like going to target – shields, swords, but also housewares and automotive
  • Hindsight: unlimited inventory + no delete? ooops; single char is 2mb and growing, before housing… a lot to xfer to and from client;
  • Implementing a super high inventory limit… as well as a way for players to sort inventory and give players a closet in their house where they can store inventory items

The Fun

  • Assumptions: MMOs are about systems and rewards and not gameplay
  • Problem: audience expects moment to moment fun, so I expect more polish
  • User story: want to have fun ACTUALLY playing your game; all play styles need to be available to me
  • Other games: puzzle pirates, puzzle quest, etc
  • Solution: emphasize interaction and reward equally; For each min–game target a specific gender and age with the mechanics… mining (boys), harvesting (girls)
  • Solution: have a sense of humour… anything is more fun if it’s funny
  • Hindsight: Game developers are not normal… what we thought was fun was not; inveting more in 2D games, improving 3d games so they are easier to play, improving camera, exploration should have been optional – will change it to be a playstyle not mandatory, activities need to be clearly marked;
  • All play styles and mini-games need reward and progression… our tower defense game had no rewards and wasn’t being played… once we added rewards it went through the roof
  • Players want to make their own fun. Parties are very popular in Free Realms.  Always a party going on. Need to give players tools to make their own fun and just hang out.

Stats

  • Assumption: MMOs have to have stats.
  • Problem: how do I choose between items with lots of confusing stats?
  • Problem: I don’t want to have to use a calcuator
  • User story: I want to understand my choices without having to learn these stats
  • Solutions: Only derived result stats – all explicit about what they do for you, very few stats, shard system
  • Hindsight: Stick to only derived stats, but add more depth; find ways to better separate appearance from stats, enhance “walking leaderboard” variables (need to look at someone and know what they’ve done)
  • Players who walk around in banana or hotdog suit get challenged to more duels.

Look & Feel

  • Assumption: high fantasy is cool
  • Problem: fantasy is not cool, it is for nerds; embarassed to talk about; I want a player that looks like me
  • Free Realms has to appeal to the guy who beats up the kid who plays WoW
  • Smedley’s kid decided he wasn’t going to talk about WoW anymore at school because he wanted to get dates
  • User story: As a player, I want the choice to look cool and wear real clothing that looks cool to my friends –
  • Solution: FR is a mix of real world and fantasy and player can look the way they want to look
  • Matrix: Costume, Freestyle vs Combat, Non-combat
  • Critical shift to go from fantasy to real
  • Hindsight: Need more clothing choices in character create for girls; need more bad ass appearances for boys; need more elaborate outfits for high character progression

Progression

  • Assumption: characters level up only by playing combat
  • Problem: game looks cool, but I don’t want to fight; things I like (crafting) aren’t important enough to level up on
  • User story: As a player, I want to level up for what I do – don’t tell me what I like isn’t important)
  • Solution: level up for multiple things
  • Hindsight: Need more meaningful items; need more robust and accurate leaderboards, achievements will be really significant in progression when they come online in a week; need more consistency across jobs
  • Figuring out how a postman and a ninja level up in a consistent way was designer hell. If you are considering a game that has multiple play styles that are equally rewarded, think of it a lot up front otherwise your game will feel like a carnival.

Sessions

  • Assumption: Should take many hours to get the best rewards in the game
  • Problem: I like this game, but I don’t get anywhere playing it 2 hours a week; Why won’t you let me spend more money on your game?
  • User stories: Don’t want to commit all my time to one game; want to decide whether or not I want to buy game changing items
  • Solution: Gameplay is in 15 min chunks, interactions give frequent rewards; game-changing items avail in store – add exp, etc
  • Hindsight: we shipped the game too easy… will make combat harder; will change game to make sure best gear is dropped not bought, putting more limited time items on marketplace, adding a wheel of prizes… helps short sessions

The Open Road

  • How do you know if any of your ideas will work? You don’t. Get to the point where you can try it on people outside of your live audience… really key.

Solutions lead to new problems

  • How do leaderboards work if you char is not linked to a server?
  • How do you set difficulty when your char progression is so shallow? There will be more diff between level 10 and 20 character in the future.
  • How do you deal with hackers when everyone can create a free account? Player char dressed as referee. All our C/S agents wear this outfit when are in game – only our GMs can wear these. Players know they can trust this person. Guy in police outfit is the enforcer… you know you are going to get hit with a ban stick – players will stop doing something when they see him. Only 1 enforcer per server – they are a real person the GM needs to “check out”
  • How do you help players understand unlimited inventory.
  • How do you keep stats light and meaningful? Majority of our audience is level 5.
  • How do you satisfy casual and dedicated players?
  • How do you balance earning coins as postman vs pet trainer, etc? Best you can do is to try to do things in a way that doesn’t alienate people. Boots for postman were way overpowered. Designers nerfed it, we patched it, players freaked out. We unnerfed it – let the 100 players who had the boots keep them, but made new boots fixed.

Design Cycle

  • Identify passengers: who is actually playing? Are the ppl we wanted to play, playing? Do we need to retarget?
  • Set guideposts: What now is important to players? What are they asking for?
  • Competing tours: Never stop looking. After we released, several competitors changed web flows. No we are changing.
  • Clear Path: What of our assumptions were wrong.
  • Design passengers experience: 3 levels of data: Stats (how many ppl play Postman), Trend (are more people playing postman now), Correlation data (most important: how many ppl who play postman now but didn’t before are now buying items?)

Parting thoughts

  • Stay focused on goals
  • Find creative solutions
  • Understand that each solution creates new probs
  • Play game in your head and look for edge cases (what happens if bought a job but my membership lapsed and I tried to interact with something)
  • Evaluate each decision against your guideposts
  • Design every system to be as flexible as you can – easy to make changes without people noticing
  • Be willing to take risks – but be willing to cull it early if it looks like it will fail
  • Be willing to kill your darlings
  • Stay in touch with your passengers
  • Keep focus testing! Ends guesswork and arguments.