As Lead Producer on Disney Friends, I owned the vision and delivery of an ambitious new game that blended four unique cinematic universes with a single, cohesive gameplay mechanic. We became one of the first Nintendo DS titles to incorporate speech recognition, but had to do so in a way that performed well, integrated seamlessly with gameplay goals, and could ship simultaneously in 7 languages.
I led a cross-disciplinary team of artists, developers and game designers with a total development budget of 1.85 million dollars US to ship on-time in 2008 to extremely solid reviews for a game targeted at a young demographic. This project ran from May 2006 through October 2007.
- Create a new Nintendo DS exclusive game based on a high concept proposed by our client Disney – placing players in the role of “guardian” of one of several recognizable Disney characters.
- Maximize usage of the DS’s unique features while creating a game that competed with Nintendo first-party titles like Nintendogs and Animal Crossing.
- Ensure that our characters lived in thematically appropriate but separate universes, and that players were appropriately immersed in their interactions.
- Ensure our project (the largest in our studio’s history) remained creatively coherent while minimizing extra hours, ensuring on-time deliveries, and maintain our relationship with our Disney client partners.
Jump to: Project Pitch – Pre-Production – Production – UX/Game Design + Research – Impact + Response
My role
As Lead Producer, I owned day-to-day production and vision for the entire project from initial pitch through project closure. My responsibilities included:
- Ensure Disney’s ask was interpreted into a coherent, feasible game concept
- Build the team and hire for open roles
- Build and manage the project schedule and budget
- Day-to-day project operations, studio relationships, and reporting
- Create initial game design specs and contribute to game designs after hiring a design team
- Create UX designs and drive usability testing
- Ensure creative cohesion and cross-team collaboration throughout the project
- Ensure on-time and on-budget completion
- Drive communication with Disney for all project matters
My team
My core Disney Friends development team encompassed the following roles:
- DESIGN: 1 lead designer, 1 scripter/level designer
- ART: 1 lead artist, 1 3D modeler, 3 animators, 2 environment artists
- DEVELOPMENT: 1 lead programmer, 4 programmers
- QA: 1 QA lead, 1 assistant QA lead, 3 software testers
- AUDIO: 1 audio coordinator (shared), 1 composer (vendor), 2 sound effects artists (vendors)
- Additional development: (extension contract): 6 programmers
- Additional art: 5 artists
Key stakeholders included:
- GRIPTONITE: CEO, Executive Studio Director, and department heads (Art, Programming, Design)
- DISNEY: Executive Producer, Director of Game Design, Producer, Game Design Producer, Lead Artists (3), Tech Director, Brand Manager
Project Pitch & Design Interpretation
Coming off of my work on Pirates of the Caribbean GBA/DS, I was offered the chance to lead a team for Disney Friends, a new concept coming out of Disney Interactive we’d been invited to pitch. We were provided with a 5-slide pitch deck which largely revolved around this image, depicting the initial concept.
We reviewed this for key themes: relationships over time, seasons, protector relationship, full use of DS features. As I did not yet have a team, I worked with our department heads as we developed a pitch that turned this idea into something that could be completed at that stage in DS development on our ballpark budget. We were greenlit as the largest DS project ever signed by our studio.
Notably, of course, the pitch included some ideas that didn’t make it into the final game:
- Cross-Pollination: Disney didn’t want characters from different worlds interacting.
- Character count: We opted for fewer, richer characters rather than many shallower characters due to schedule, budget, and cartridge size.
- Music: Music as an interaction was transformed into the concept of “Activities”, and expanded.
Pre-Production: Strategy & Design
As soon as we were greenlit, I began hiring for my lead designer in earnest, while building out the rest of my team with internal resources.
Character Selection
I immediately began discussions with Disney about which characters would be included in our game, with input from my leads who had already started on the team. Disney was offering us almost unrestricted access to their catalog: we were even offered Mickey and Minnie. But the original game concept was a child caring for a Friend as a guardian. Characters like Mickey and Minnie are incredibly self-sufficient. I felt strongly we needed characters who would plausibly need regular assistance. We knew Stitch was a home run, but to land on our other three characters we partnered with Marketing for focus groups and conducted our own research.
After many spirited discussions, many movies watched, and some very enlightening focus groups we landed on:
- Stitch: extremely popular in our target market, often needs help regulating moods or cleaning up
- Dory: struggles with memory issues and benefits from the aid of others in her film
- Simba (young): needs guardianship from Zazu in his film, so this relationship is on-brand
- Winnie the Pooh: basically spends half his life stuck in a honey pot; his neighborhood is an extended support network
- Toy Story Alien: Used as our tutorial character (fun, but not verbal enough for a full interaction).
Dory was actually the toughest to land on. We felt an underwater character would be a nice contrast, and initially wanted Flounder but the film was older AND Disney had just recalled it to the Vault pre-streaming times. Our first guess from Finding Nemo, Nemo himself, went over very poorly in focus groups – but I called an audible and asked Marketing to propose Dory as an alternative, to significant positive reactions.
Once we had selected our characters, I partnered with my lead artist to ensure we were well-connected with Disney’s brand resources and were fast-tracked to character art approvals.
“It seems like a weird batch of characters, but it makes sense. Stitch is a ravenous little imp that needs constant supervision. Simba is a cub, not yet trained to do stuff on his own. And Poor and Dory are… kind of dumb. So everyone in the game has a valid reason for needing a Guardian.”
IGN Review of Disney Friends
Immersion and Innovation
The core of our game was “Interaction Mode” – where players are 1:1 with a character, building a relationship. Normally, the DS only renders 3D on one screen, and the other screen is used for UI elements. In our case, we wanted to use touch for direct character contact, so the 2D screen would be fairly useless informational UI. I felt that would be a distraction. How could we be more immersive? Could we make it feel like there wasn’t useless UI?
In partnership with my lead programmer, we tore down the technical architecture of the DS and discovered a large texture buffer that could be used to store a full 3D rendered frame. We worked with our lead artist to redistribute our texture art across the remaining buffers, freeing this large buffer up to give us 3D on two screens by double-buffering. We’d halve our frame rate, rendering one frame to the buffer and then one frame direct to screen, displaying both.
Innovating through Gameplay
ARTIFICIAL INTELLIGENCE: We sought to create a more interesting and dynamic personality system for our characters than traditional Sims “motives” or the binary Nintendogs happy/sad. We (my Lead Game Designer, my lead developer, and myself) landed on a system of 4 binary need sliders, and a way to specify a set of 4 thresholds for the motives as a specific “emotion”, like ‘excited’, ‘naughty’, or ‘lonely.’
VOICE CONTROL: In order to aid in the immersion during interaction mode, we supported voice control with an approximate 20-word grammar and equivalent touchscreen dictionary. Very few Nintendo titles had ever shipped with voice interactivity, and the effect on gamers and games was fairly uncharted territory.
Production: Leadership & Relationship Management
Working towards a Shared Vision
- I ensured that each of our pre-alpha project milestones had a central functional goal to make assessing success easier. (Prototype, Vertical Slice of functionality, etc)..
- I structured our design documents for ease of collaboration and working within a source control system, breaking each of our character-specific design documents out of the Master Game Design Document which covered system design.
- I also felt it was important for us to document as much as possible about how we worked together and not just the game design itself (which I also helped maintain). I built out the project Wiki and documented my conversations, our client asks, contact lists, and other work logistics there for the team.
- This proved invaluable when a few months in I was diagnosed with ovarian tumors and needed emergency surgery. My documentation allowed my leads to pitch in during my stead and keep things going – although I’ve perhaps never seen 3 people more happy to see me return a few weeks later!
Fighting Crunch
I’d specifically constructed our schedule in an attempt to fend off the need for extra hours, as I feel strongly the games industry does not need to burn people out. Each milestone included at least one week of slush time to accommodate inevitable bugfixes, regressions, and client asks.
However, crunch was still a deeply ingrained practice at the studio, especially amongst the programming teams. We noted a growing trend of late hours amongst our programmers and in tandem with my lead programmer, our department head, and studio head we engaged in a concerted effort to break the cycle. Late hours lead to bugs and late starts, which perpetuate the cycle.
- I tracked late hours and late starts and repeatedly encouraged individuals to leave at the end of the workday, while top-down support came from department leads.
- I also held a firm line against feature creep – we had very enthusiastic developers who wanted to put in pet features, but those introduce risk, complexity, and scope we hadn’t planned on. I usually pushed back against these asks, or captured the ideas and brought them to Disney for funding (which resulted in an extension to our contract.)
Client Relations
BRAND ALIGNMENT: Disney is a client with high standards. We regularly received deliveries of model sheets or other reference materials from the Disney Vault and were expected to ensure our work was on-model; this meant an extra layer of approvals for all assets. All character-specific design decisions also needed to be evaluated for brand alignment. I was the primary driver of all of these conversations, and the broker of compromises when Disney desires conflicted with the capabilities of the DS or the limitations of the schedule.
ROOT CAUSE ANALYSIS: There were also larger, macro level considerations to track when working with Disney. I noticed a pattern partway into our partnership: after a delivery, they’d drop off the radar for a day or two, then deliver a list of apparent demands without context, and many of them were out of scope for the contract. I began to realize this was our client’s way of expressing concerns while trying to helpfully provide solutions. I partnered with them to shift communications to the root problems, rather than suggested solutions. This led to more constructive conversations where my leads could use their expertise in order to solve any problems noted in more efficient ways.
RESPONSIBLE SCALING: At one point in the process, it became clear that Disney’s asks indicated they wanted more content than was covered in the original scope – and they were also asking for a major overhaul of a previously approved minigame. I transparently communicated the schedule impacts of these asks, and worked with my leads to put together a proposal for a reasonable contract extension that addressed their concerns by paying for additional resources while keeping us on schedule without excessive crunch hours or introducing excessive risk. We successfully negotiated a $350K addition to the $1.5M contract.
User Experience Design & Game Design
A subset of my responsibilities included the interaction design and research for our game, as this was not a distinct discipline for the studio. While I delegated the majority of game design tasks to our talented design lead, I did take on a subset of other design tasks like writing, cutscene scripting, AI tuning, and directing voice recording sessions.
Interaction Design
Menu designs for DS games may seem simplistic, but console games are subject to a complicated series of requirements and constraints from the platform level. We were also localizing to 7 languages simultaneously, so all work had to be created with the globe in mind. Multiplayer interaction design is particularly challenging as it is rife with error conditions, and mishandling any of these errors can cause a product to be rejected by the platform.
UI flow diagrams for our core gameplay experience and our multiplayer lobby experiences.
UX Research / Playtesting
While we did have a dedicated QA team, I felt it important we have an additional track of evaluation focused on gameplay and not just bugs. Our UX/gameplay research encompassed 3 techniques:
- In-house playtesting with local children, followed by discussion and surveys
- Formal lab playtesting in a focus group setting, followed by discussion
- Longitudinal diary study (3rd party, hired by Disney)
Our playtesting program yielded a variety of really important insights that had significant impact on the gameplay.
IMMERSION AFFIRMATION: Some of the insights affirmed our design choices: for example, our experience was so immersive that kids didn’t refer to game-breaking asserts as a bug. Instead, they said things like, “Stitch turned the game off because I didn’t give him a sandwich.” This seemed to affirm that our 3D-on-2-screens approach was working, keeping kids focused on the characters.
GENDER NORMS AND GAMEPLAY GOALS: In early playtests, we had Stitch as a playable character. Stitch is massively popular with young, pre-teen boys. But in the studies, the girl players LOVED the game and said they could play it indefinitely, but the boys just didn’t “get it,” and reported they didn’t know what to do. This was a big red flag: how did we mess up Stitch for boys? I talked with my lead designer about a theory: are these boys used to playing for mastery? At first we worried we’d have to add objectives and a bunch of UI, which might turn off the girls. We decided to start with just a single UI element: a Friendship Points meter that provides a daily goal, along with particle effects that indicate when you’re earning those points. We retested and the result was night and day: boys LOVED playing with Stitch, and the girls still loved it as well.
Voice UX Design
This was my first foray into the world of voice user experience design. For each of our player characters, I had to design an approximately 20-word “grammar”: the set of words or phrases the character would recognize when delivered using our push-to-talk feature.
VOICE AND NORMS: I had an epiphany when I caught my lead programmer trying to test the “I love you” interaction. He thought he was alone, but still whispered “I love you” to the device too softly to be recognized. He tried a little louder, still no luck. Finally he went full voice. What was behind this behavior? Why was he embarrassed to use that phrase when he knew we knew what he was doing, and what did that mean for little boys? Two things came out of this exploration:
- All voice commands could also be triggered via touch equivalents from a menu that also served as a dictionary of commands. In this way, our voice UI was multimodal and adaptive.
- For particularly vulnerable commands like “I love you” where social norms might interfere with willingness to deliver, we provided alternatives – so instead of “I love you” you could also say “Friend” to get the same response.
ACCURACY AND LOCALIZATION: Of course, it took a fair bit of work to ensure English success rates were acceptable, choosing acoustically unique words. But that’s not all. We shipped Disney Friends simultaneously into 7 languages. This was particularly challenging because grammar-based systems are interdependent. If all the words in one language start with the same sound, performance will be poor. Even if we had the English dialed in to 20 very different words, that didn’t mean these 20 words would work well in all 7 languages. We had to work back and forth thoughtfully to ensure reasonable success rates across ALL languages.
PROMPTS AND DIRECTION: As part of my voice design, I wrote all of the character responses to the voice commands in English (subject to brand review), and also participated in the direction of the voice recording sessions for several of the characters including Stitch.
Impact and Response
The game shipped on the timeline provided by Disney (they pushed the original timeline back on their request) – but we managed to cut off the crunch spiral and keep the team balanced. I still hear from team members that this is their favorite game development project to date.
We were originally slated for a MASSIVE PR push, but due to a major restructuring at Disney, our marketing budget was cut and reallocated to Turok. This was despite strong results coming in from our diary study on a Beta build that, according to the independent research company, were “on par with Nintendo first party titles.”
“It’s one of those games that’s mystifying in its invisibility, the sort of thing that should be advertised until we desperately hunt down the remote just so we don’t have to see that damned commercial again. Instead, it slipped by with nary a wave in the pool of DS games, and that’s just criminal.”
DS Fanboy Review, 2008
The Good
While this was a game targeted mostly at tweens and younger, it managed to pull solid review scores, including a 9/10 from USA Today and a 70 from IGN.
“It was only a matter of time before Disney did a pet simulation title. The genre has been ripe for the plucking with little other than puppy and baby games keeping it afloat. But rather than dumping out some schlocky shovelware, Disney got the guys over at Amaze to make a game that actually manages to be fun.”
IGN Disney Friends Review
“If you are a fan of these types of experiences you will be hard pressed to find a better offering than Amaze’s Disney Friends.”
-ZTGD
Could Be Improved
If I had to do it over again, I’d take a much harder cut at our text in general. Griptonite was used to making narrative Sims games for teens, and we leaned on those habits – too much text for some of the kids who wanted to play our games and who could have expanded our market. Furthermore, the font suffered from readability issues we just didn’t see because we’d all been working on GBA and DS games for too long.
