An MVP Prototype Story
Burt Hopkins
?? Project Management Leader | Agile Methodologies Expert | Driving Innovation, Team Excellence, & Continuous Improvement
When thinking of an MVP, many things might come to mind. If it's Basketball, maybe Jordan pops up. Baseball? Maybe Mike Schmidt (long shot, but I'm a Phillies fan, no Yankees around in my headspace). But, to an Agile professional, Most Valuable Player isn't what comes to mind when you hear MVP.
You're thinking Minimum Viable Product.
It's counter-intuitive when taking on a big idea. And, let's be honest, big ideas are the name of the game. No one who wants to change minds or make their mark on the world has small beer in mind. Transformation, growth, evolution, game changers are made of those components. But, chances are, outsized effects start with small things. They don't happen all at once. My father (frustratingly) always told my sister and me: "nothing happens now."
And, as much as we hated it then, he was right. At least, he was when it comes to designing that great big idea. That thing that will make everyone re-think what they're doing and come around to a new way of doing things, or maybe even switching to a new kind of thing entirely.
Lots of little things have to happen before that first, and that's where the idea minimum viable product comes in handy.
In fact, I can tell you a story about how I just finished dealing with MVP right now.
How To Build a Game In Seven Weeks
My boss, Daniel Lynn, and I were discussing Dungeons and Dragons (as we often do) a few months ago, particularly about how the game inherently uses cross-functional teams (and always has). I'd started a blog article on it with a narrative style featuring a fantasy-themed heist, and Daniel immediately latched onto an idea. D&D could only reach so many people as a metaphor for teaching cross-functionality - but everyone has seen Oceans 11.
People seem to love heists (my family binged Netflix's Kaleidoscope in two days recently), be they in cinema, television shows, books, or even comics. So we decided to pitch the idea for a workshop this year in Scrum Alliance's Global Scrum Gathering. We would propose a workshop featuring a game. Attendees would take on the role of a Mastermind, assemble a crew of Operators to run a Heist using cross-functional principles, and have fun doing it. How? We'd worry about that part if the talk got accepted.
So, we made our pitch and promptly forgot about it for a month and a half.
Then, it got accepted.
Hello MVP
We realized very quickly that we would have to design, playtest, and print a game in about seven weeks. When a deadline like that comes down, a couple of things must happen.
First, I recommend allowing yourself to panic for about an hour. Breathe into a paper bag. Maybe yell in a safe space for a bit if that's your speed. Cower in a corner with your dog (my dog is super cozy).
Whatever it is you need to do, it's okay. I'm not going to judge you, so long as the next step is putting your Agile pants on and working on that MVP I've been promising to get back to talking about.
I am, first and foremost, an avid lover of games. So, to begin the design of our first MVP, I opened a fresh Miro board and began putting up the names of all the games I love playing; stuff with twists and turns, social elements, and lots of interactions. Once I had a couple on the board, I picked them apart for why I liked them and put them in their own frame for reference.
After that, I started figuring out things our game needed to feature based on what I liked in those games: strategy, speed, and more than a little whimsy. After that, I figured we'd have to incorporate the virtues of what we hoped to impart into the workshop: cross-functionality and the ability to demonstrate as many agile teachings and values as would reasonably fit. Lastly, more than anything else, it had to be fun, easy to teach, and cooperative. Everyone works together, or everyone loses.
The good news is, all of that took about four hours. Concepts are so very rarely difficult. Application is always tricky.
The First MVP Sprint: High Concept, Bare Bones Functions, First Playtest
My Miro board assembled, I took it to Daniel. We started talking about what kind of game it was going to be. We had to quickly figure out how we wanted things to look and feel. A board game didn't feel right because boards and dice would be problematic. Dice are easy to come by, but boards would be a slapdash, handmade affair or expensive to produce if we wanted to put something professional together, to say nothing of the time it would take to make them.
Cards, however, would be doable if we could design them quickly enough. Worse to worse, index cards or printable stock could be used in a pinch (or so we thought). Art was still a problem. Even with Michi, our in-house graphic designer, the amount of art we'd need would be massive. I'm also a designer, but my primary job as a Client Relationship Manager made that moot. I wouldn't have the time either. We ultimately turned to AI, specifically Midjourney to provide the art, guided by my provided prompts.
(We wrestled with the choice to use AI art and went with it because this is presently a prototype game and not for profit. Should Oceans42 go to market as a sellable game, a full redesign is intended.)
Art sourcing handled, we made the call for our first sprint goal and MVP. The game's system would call for each player to take on the role of a Mastermind (represented by a single card) to draft a Crew of Operators (represented by a series of several cards) to pull off a Heist (represented by a placard). All players must succeed at their heists. If any Crew fails, they are arrested and then rat out the remaining Crews (no Prisoner's Dilemma here).
领英推荐
The first MVP for our game would comprise of:
We completed all those goals by the end of the sprint, though the actual playtest wouldn't happen until the following Monday.
The Second MVP: More and Better Everything, Plus Math
Now that we had completed the first sprint, we ran the first demo in Miro (which worked far better than I thought it would), and the system's flaws become obvious. The game, as we'd made it, was impossible to win. There just weren't enough cards. I had to make more cards on the fly in Miro to make the game playable. Even then, the Operators' skill distributions still weren't enough to make the game possible to win with secondary mechanics in place from Mastermind cards. I knew trusting my gut was dicey when it came to statting out the Operator cards. I went to college for art, after all. In no way should I ever have been allowed near complicated math.
The good news was that our playtesters liked the concept of how it worked. They immediately grasped the strategy. It engaged them and made them think right from the Drafting phase about exactly how they'd have to pick the right Operators to pull off their Heists cross-functionally. They said they enjoyed our broken game!
As we played, I made a handful of Miro stickies, and after, we ran a brief retrospective. I learned I'd need a clearer rule opening, cheat sheets, a handful of clarifications, and playspace tweaks. It gave me everything I needed to bring guidelines for the next MVP back to Daniel. We settled on:
I began to throw myself into the work and found that I had finally become something I'd always wanted to be: a game designer. I began actively promoting the upcoming Portland workshop and the game itself on LinkedIn. Colleagues and friends began to take note. A great deal of output began to pile up, and I put in hours beyond the usual to get it done, knowing there was not much time left.
It was around this time, my enthusiasm absolutely grew beyond what was probably healthy, and I suggested a massive uptick in the number of cards for the game (somewhere around double the number of Operators). Daniel, being the more experienced and wiser Agilist, brought me back to - you guessed it - Minimum Viable Product. We weren't looking to get that big world-changing thing out of the sprint. He checked my ambition and re-scaled things. We had no budget. More cards meant more money to spend on printing the game, more time trying to get an AI to make art we could use (and that aligned with conventional reality), and more math.
We were ending week four as we began the second round of playtesting. We did end up with more cards than planned, but only what we needed:
While my playtest seemed to go well (my players won and generated incredibly valuable feedback), Daniel's did not go so well (his players lost and also generated good feedback). More algorithmic tweaks would be needed (as well as a handful more Operator cards), and we would still have more to do as we entered week five.
The Third MVP: The Last Card Additions, Last Minute Print Plans, and Final Algorithm Tweaks
The game was still broken, but not nearly so bad as the last time, and that was all we were looking for with exhaustion starting to settle in for both of us. Daniel was finishing the algorithm's final iteration. Michi came through in the clutch with some much-needed print art experience, providing a template that would make print setups faster for me on the print production side. I spent the better part of 36 hours going through the print production process.
The cards went out to our printer two days ago. This is the end of week five.
Our work is not yet done.
We still have one more MVP sprint left in which we finalize our Heist placards, then balance and print them. It's the one phase we don't need a complicated print process. We can take our time with it (relatively). I think last night was the first time I wasn't designing something in my sleep for the past three weeks.
If not for the concept of MVP, I don't think we could have got this far. If we didn't know where to limit that scope and how to appropriately set the goals to something achievable, testable, and measurable, we'd be in a bad way.
And since we knew all of that, we're excited to see you all in Portland in two weeks, when we hope you'll join us to pull off some heists.
Oh, and if you're wondering, here's the final spec on our awesome game:
Leader in Agile Product Management | Driving SaaS innovation through Agile best practices, customer- centric solutions, and product-led growth.
1 年I really appreciate that your first MVP was not a fully functioning game for Portland haha. Well done.