The Garbage Pail Kids: The Game

Download the game (iOS or Android) here.

I won’t be doing a breakdown of my time on the Garbage Pail Kids: The Game to the same degree as everything else because it’s much more recent and still running, but I will elaborate on what’s in my resume/LinkedIn profile a bit!

Part One: The Launch

From LinkedIn:

Joined as Producer before the game’s soft launch. Worked with a 100% remote team to execute on the CEO’s vision for a polished mobile card-battler on a tight budget. Owned the product roadmap and streamlined production process. Identified tutorial bottlenecks in analytics data and subsequently doubled FTUE completion rate and substantially increased retention. Procured an external QA vendor.

In practical terms: The game was developed as a joint venture between Jago Studios and Galaxy Pest Control (GPC from here on out). I joined on the Jago side initially to help run the game as a live service (thanks to my experience on Star Wars: Galaxy of Heroes).

I joined a few months before we were set to soft launch, but as we did our final sweeps through the game there was clearly a lot more work to do and a lot of bugs to track, so I took over the project management tasks so everyone else could focus on other things. We were using our QA vendor’s JIRA instance for task tracking so I didn’t have 100% control over the configuration, so I had to improvise a bit. The method I settled on was to use the Sprint view (which arranges tickets into rows separated by Sprints) to prioritize tasks, using the topmost Sprint as the current and most urgent/important, with additional “sprints” not representing actual sprints or anything on the release schedule, but different types and priorities of bugs. This is sort of a hybrid Sprint/kanban view, and it worked pretty well!

Once we soft launched, I used Unity’s original (now discontinued) analytics service to set up funnels for the tutorial and immediately saw some sharp drops in places we didn’t expect. There were actually some weird drops that didn’t seem to make sense, so I manually downloaded the analytics data for each day of soft launch and wrote some scripts to parse it with python. Analyzing the resulting output, I was able to determine that some events were firing inconsistently, more than once, or in the wrong order.

Once we ironed out the technical issues (what is soft launch for, right?), we set about reducing the drop-offs in the now-trustworthy data. We iterated on the tutorial for months (not exclusively, of course), and ultimately doubled the completion rate. We got to the point that the biggest drop-offs were during the battles, which made sense because for one, they were long periods of time and two, because that’s the core gameplay. My experience going all the way back to Superhero City is that no matter how much you tweak the FTUE, there’s still a finite number of people interested in even the best possible version of your core gameplay. Because of that, it makes sense that the drop-offs during combat would be large and, for the most part, not moveable.

Part Two: Live Ops And Updates

For reasons I won’t get into, after GPK launched there were no updates for quite a while. Finally, in mid-2020, Reliance Games (who had been doing some QA on the game before launch) agreed to take over development, port the game to Android, and start releasing regular content updates.

From my resume:

Beginning in Q3 2020, development of GPK ramped up in collaboration with an external partner and I assumed all game design responsibilities. Worked with the team to substantially improve KPIs by redesigning the game’s PvP and Event game modes, introducing new characters, rebalancing the game economy, and adding brand new social features as well as a procedurally generated random challenge mode that significantly increased engagement among elder players. Also responsible for writing briefs and approving all art, animation, and sound assets for new characters, as well as interviewing, hiring, training, and mentoring game designers.

Reliance had a full team of programmers, PMs, and a producer ready to go, but no spare game designers for GPK, so I agreed to take over on that end. I generally enjoy game design more than production/project management anyway!

There was still producer-y work for me to do though! I ended up creating spreadsheet templates in Google Docs that included characters, packs/IAP, and live events for every release – similar to what I had done back at EA.

I also had plenty of design work to catch up on. I set about creating a master design doc that included all character stats, event info, packs, PvE loot drops… everything we would need in order to understand the balance of the game and operate it as a live service.

GPK as an IP is really interesting to make a game with because there’s just an absolutely massive selection of source content – hundreds (thousands?) of cards, each with original characters. Picking characters that would work for the game is a bit of work, but grouping them together into themed releases is even harder. I spent a lot of time in the first six months or so of live service operation working with Jago’s CEO, Stuart, to pick the characters for each release. Eventually, the game’s new writer (whom Stuart hired after sharing copywriting duties with me for several months) took over the task of compiling groups of characters for releases, which was extremely helpful.

One of my favorite parts of releasing new characters for GPK was writing the animation briefs. Reliance’s animation team was new to 2D animation, having spent years working on 3D licensed mobile games like WWE, Drones, and Real Steel, so we (at Jago) were asked to write detailed animation briefs for our new characters. I loved the moments during Jago team meetings where I would suggest a character and someone would say “How could that possibly work animated?” and I’d say “I think I’ve got an idea…” and write some notes in our monthly content release doc. Then weeks later, an animation would show up attached to a JIRA ticket that was just perfect.

Animation Fun

In my opinion, there’s a threshold of weirdness of character design where as you approach it, the animations get really fun and interesting, but if you cross it they just stop making sense and the whole thing doesn’t work anymore.

There’s also the question of “how much can this character actually DO?” – Given the constraints of time, what we can rig using Spline, and how many overlapping or complex pieces are in the image, sometimes we have to throw out an idea for a card adaptation because it would just be too much work. Lots of GPK cards feature characters mid-explosion, and to adapt those we’d have to draw new, un-exploded version of the characters and then find a way to blow them up (and rig them) to approximate the exploded state on the card… it’s just too much work!

Here are a few characters I am especially proud of.

Shirley Temple

Reuben Cube

Bruised Bruce

This was part of our big Christmas 2021 release, and we really went all-out. I recommended that we redesign the original card quite a bit to modernize it (and add more movie references). Check out the side-by-side!

Sorry about the quality of the scan on the left, but the official high-resolution version Topps sent us is confidential. Anyway, we re-did the background and gave him a gun somewhat like the one from the movie (but a bit more cartoony so we don’t get in trouble with Apple regarding the game’s age rating!). I’m really happy with how this one came out!

Character Ability Design

The original set of characters in the GPK game were assembled with tried-and-true archetype builds, but I felt that they didn’t necessarily feel like they were always appropriate to the characters and the IP. While we had limited resources to create new mechanics, I wanted to make sure that new characters had ability kits that felt much more like they were telling the story of the character.

Telling the character story through abilities isn’t always easy (or possible), both because of time constraints and the base rules of combat, but it’s always worth it to at least try. Here are some examples of ability sets I’m particularly happy with:

• Reuben Cube
  • Basic attack: Blockheadbutt
    • Reuben attacks one enemy. If Reuben has any debuffs, apply copies of each to the enemy.
  • Special attack: Solved
    • Reuben taunts for 2 turns if not already taunting and converts all debuffs to equivalent buffs.
  • Special attack: Scrambled
    • Reuben taunts for 2 turns if not already taunting and applies 3 debuffs to himself.
  • Leader ability: Geeky Squad
    • While Reuben is taunting and has no debuffs, Geeky teammates receive +5% Physical Damage.

(These are all at level 1 – Once upgraded, all these numbers go up)

The basic idea here is to imitate real life: You scramble the cube first, then solve it. Players can choose never to use the Scrambled ability, but it’s an interesting way to end up with three buffs if you use Solved next, OR to apply those debuffs to enemies via his basic attack. Plus, using Solved will always result in teammates receiving bonuses to Physical Damage.

• Spicy Spencer
  • Basic Attack: Special Sauce
    • Spencer attacks all and applies Offense Down for 2 turns
  • Special Ability: What Smell?
    • Spencer Taunts for 2 turns. He also removes up to 2 debuffs from each teammate, adds them to himself, and heals 2% of his max health for each debuff added.
  • Leader Ability: Desensitized
    • Smelly teammates get +1 Tenacity.
  • Passive Ability: Skunked
    • Whenever Spicy Spencer is attacked, 25% chance to apply either Offense Down or Defense Down to the attacker for 2 turns.

Spicy Spencer clearly enjoys the smell of skunk, so everything about his ability design is about how smelly he is, and just how much you wouldn’t want to touch him. He actually enjoys it so much that he’s able to transfer debuffs from teammates to himself and it heals him in the process – like having something that most would consider a huge disadvantage is still good for him. And of course his passive ability shows that anyone who touches him is just gonna end up stinky too.

I think if I could rebuild the game from scratch, ideas like being smelly or slimy would be their own specific status effects that would function in some specialized way… but since we didn’t have anything like that and there were other priorities, I had to simulate those ideas using the existing mechanics and systems.

Part Three: Moving Up And Moving On

Eventually, we were able to hire an entry-level game designer part-time to take over a bunch of my work on content updates. Monica was able to take over the nitty-gritty of planning week by week, finding eligible characters, and most of the data entry (thankfully, as I had been doing that kind of work going all the way back to Heroes of Dragon Age!) so I could spend more time writing specs for new features and working with Jago’s founder, Stuart, on pitches for potential future projects.

In the summer of 2022, Monica got a full-time job elsewhere and Reliance Games found a designer to take over full-time from their end on GPK. I recorded several videos walking through the process of programming character abilities (a task I had been performing ever since Reliance started in late 2020, because it required both design skills and programming skills), had some final handoff meetings, and bowed out of the project for good.