Ep. 22 | Leveling Up: Gamification in Software Development  ## Metadata - Author: [[Overcommitted]] - Full Title: Ep. 22 | Leveling Up: Gamification in Software Development - Category: #podcasts - URL: https://share.snipd.com/episode/afc38f8e-b1f2-4b89-8402-b194e99d7930 ## Document Notes ## Highlights - **Reframing Failure Changes Behavior** - Framing failure changes behavior: participants who saw failure as losing points tried less and succeeded more than those who just got an error message. - Reframing failure as part of progress can reduce fear and increase experimentation. Transcript: Erica I first came across this idea reading Feel Good Productivity, which introduces a study conducted by Mark Rober, NASA-trained engineer who recruited 50,000 people to try a new computer Challenge. Participants were divided into two groups when they wrote code that didn't work. One group received a generic error message that said, you have failed, please try again. And the other group received the message, you have failed, you've lost five points, you now have 195 points, Please try again. And the results are really interesting. The participants in the first group made an average of 12 attempts and had a 68% success rate. And those in the second group made an average of five attempts and had a 52% success rate. So the takeaway here is that we are afraid of failure even when it's arbitrary and left to chance. When he was sharing his findings, Rober posed the question, if we could frame our learning process so that we weren't so concerned with failure, how much more could we learn and how much More could we exceed? ([Time 0:00:35](https://share.snipd.com/snip/ba9d3603-0be5-43c9-8dae-cf11ad5e4b15)) - **Run Safe Incremental Experiments** - Try safe, incremental experiments like Ship to Learn to validate ideas in real systems without catastrophic risk. - Use data and metrics to define success criteria before running experiments so you can judge outcomes clearly. Transcript: Bethany Janos Yeah. So at GitHub, I think we have a really cool, I don't know what you'd call it, I'm saying, but it's Ship to Learn, where we're actually encouraged to try to try things out in production and Validate our experiments or our thoughts. I think there is definitely a lot of trial and error, but I think that you also can leverage data and metrics and things to understand what you're doing better and gain more criteria on What is success, what is failure. And so I think while there is a of that, there is a way to safely introduce these tests and experiments without necessarily making the product worse or making it a bad experience or whatnot. ([Time 0:02:46](https://share.snipd.com/snip/229e286d-1494-4d70-8be7-8ac2474225a0)) - **Pre-Mortem To Clarify Success** - Erica described running a pre-mortem to define technical success and failure early in a project. - She found clarifying criteria helped reduce spinning wheels and personal anxiety about slipping dates. Transcript: Erica Sometimes we do and that's good, but I recently did a pre-mortem for a project, and it was exactly this thought experiment of what does success look like? What does failure look like from a technical level? Because I think the first thing that my mind goes to with failure is, and what I personally experience with the fear of failure is, this project doesn't complete on time or this project Doesn't complete at all because we take too long and leadership shifts priorities or leadership shifts priorities in the middle of working on it and it's going perfectly fine. And like, how does that reflect on me? Like, was it because of something I did? And yeah, definitely, like you said, when you're leading a project, you feel like you have a lot of investment in it. And it does feel like a personal, like, yeah, I, I personally feel it personally when things don't go well. ([Time 0:07:41](https://share.snipd.com/snip/a0763418-67cf-4ccb-a661-f5d00d9aa5f9)) - **Layered Gamification Lowers Fear** - Gamification can be layered: sandboxes, feature experiments, and quest-like learning break complex topics into achievable steps. - Structured levels lower fear of failure and create clearer milestones for progress. Transcript: Erica Or I know we have tooling for deploying feature experiments in production and viewing performance and behavior related metrics. Or establishing a system where you break down unfamiliar complex topics into quests and levels and use that to motivate achievement. So somewhere that might be familiar, somewhere like Code Academy, where you have a topic broken down into levels and you check them off and win badges and stuff like that. So have any of you experienced any of these approaches in your work? And if so, how did they impact your motivation? Maybe I'll pass it over to Brittany first. Brittany Ellich I think I have a direct one-to with like the levels, but that's often, I think, the feedback that I'm looking for from completing tasks and completing ethics. Where I'm, you know, completing these things that lead up to this big thing that we can close. I feel like I get a lot of the same satisfaction that I get out of completing a level in a video game. Just like, yes, that to do list is complete. And I really enjoy that feeling. And I maybe I think maybe that was like kind of the reason behind the term epic to begin with. I don't know. I feel like that's very, you know, somewhat game related. Yeah. But yeah, I would love to see more gamification of those types of things going forward. I know at GitHub, anyway, we don't really do any story pointing. I think every team does everything their own way, but we don't really do any story pointing. ([Time 0:12:27](https://share.snipd.com/snip/0511b1f6-7a48-479c-9e2b-9cb1943721e6)) - **Reward Meaningful Progress, Not Vanity Metrics** - Use visible completion feedback like story completions or tracked L&D to motivate people; those small wins mirror game-level progress. - Beware that any metric you reward can be gamed, so pick measures that map to real impact. Transcript: Brittany Ellich I think I have a direct one-to with like the levels, but that's often, I think, the feedback that I'm looking for from completing tasks and completing ethics. Where I'm, you know, completing these things that lead up to this big thing that we can close. I feel like I get a lot of the same satisfaction that I get out of completing a level in a video game. Just like, yes, that to do list is complete. And I really enjoy that feeling. And I maybe I think maybe that was like kind of the reason behind the term epic to begin with. I don't know. I feel like that's very, you know, somewhat game related. Yeah. But yeah, I would love to see more gamification of those types of things going forward. I know at GitHub, anyway, we don't really do any story pointing. I think every team does everything their own way, but we don't really do any story pointing. But in the past, I've worked on teams where we estimate things and assign story points. And I feel like having those story points, I was like, ooh, I want to complete the most story points in this sprint out of my team or something like that. I feel like that's another sort of way that people kind of gamify getting things done. But obviously, if there's a way to game it, then there's a way to exploit it and make it not totally accurate. ([Time 0:13:12](https://share.snipd.com/snip/3ae74bd8-81a4-4ca3-afed-c258ba35bcfc)) - **Chasing Contributor Rankings** - Jonathan admitted he chased rankings like the contributor graph to motivate himself and get 'gold stars'. - That visible leaderboard drove short-term activity but framed work as a competitive game. Transcript: Jonathan Tamsut I'm being honest, there have been times, like actually, like when we were, you know, working on GitHub stuff as a team, you know, when we were on the same team or I would look at the contributor Graph and be like, I got to I got to climb number one. And I would like be like, surpass that person today. Okay, got to make a couple more commits. And I mean, it was just it was really motivating to me. And so yeah, I mean, I, you know, I like my gold stars for sure. ([Time 0:14:19](https://share.snipd.com/snip/61acd444-c0f2-4ef1-9135-679ff07c78cb)) - **Rewards Create Perverse Incentives** - Any incentivized metric creates perverse incentives because people will optimize the rewarded measure instead of the true goal. - Designing rewards requires anticipating how people can game the system and choosing signals that reflect real impact. Transcript: Jonathan Tamsut I think brady you brought up so many good points like i think the two that kind of stood out to me were like one you know i think it's important to realize you know if you're that you know when When you're playing a game so i think people you know you know with games there is a typically a singular or narrowly defined objective that you're optimizing for at the expense of all Others, you know, all other sort of things. And so like, you know, I think, you know, you know, that could lead to a bunch of, you know, sort of perverse incentives or, you know, outcomes that aren't necessarily desirable moral Hazard and i think that's like important to know is like you know realize you're playing a game realize you're optimizing for this and realize i guess realize like like know when to quit Or you know um and then i think the other thing is like intrinsic versus extrinsic motivation this is something i've struggled with a lot and like i've talked a lot about and you know and You know with people and I think for so much of my life like doing things to impress people or to spite people even prove people wrong has been so much of a motivator and I've recently been Trying to develop more intrinsic reasons for doing things and I think you know you ultimately are going to be happier when you're kind of just doing things for the love of the game. Brittany Ellich I do have one thing to say related to realizing that you're playing a game. I think one thing that comes up really frequently with software development is trying to like John had said have the most commits or to like close as many issues or mark off as many as you Do items just close as many pull requests And that can often be a game that people are playing within their team or within their organization to try and say like, look, how good our team Is. You should give us all the resources. And one thing that I found recently is I've been looking into my own stats around this. And I think that I've found some stats that are a little bit more important to me. And I think identifying like, okay, what do those stats actually mean? Like closing the number of pull requests doesn't necessarily mean anything because like you could have really tiny pull requests that are just updating docs, but like it's still a Number. But then I've realized that the number of pull requests I've reviewed, like that's actually a more interesting statistic where I can say, look at my impact across this team or how I've Unblocked other members of my team. So recognizing which numbers are actually meaningful to try to gamify, because if you're going to gamify something, try to find something that actually means something instead of, Erica You know, lines of code, which we all know is a terrible metric to measure anybody by. Yeah, that reminds me of the sort of like conversation around AI enabled software engineering and the idea that AI is a superpower um or like you know llm assisted coding is a superpower And it's like i feel like every time i hear this conversation it's around creating code and you're like okay like you know but how long does that code take to get to production and how many Like incidents does that code cause like there are so many other questions that are not answered in that one very narrow view that also need to be considered for like effectiveness yeah So i don't know it's an kind of an interesting like related conversation strictly gamification but a little bit because you know like you said jonathan like to some extent it always Kind of feels like you're playing some kind of a game like working for a company like yeah there's there's motivators intrinsic and extrinsic and you know a job inherently has external Motivation by how much you're paid and when your payment is related to your performance like yeah like how well you perform does matter directly to that extrinsic motivators um yeah I i think um yeah it's uh it's a it's a tough question maybe for another day of like you know how we how we define success for ourselves i think that's a topic in and of itself yeah totally, Totally. But yeah, at least recognizing that there might be like different, different goals at play. And I think your point, Brittany, was like defining those goals for myself gives me agency and sort of like takes me out of the hamster wheel of the game that I see being played. So I at least know like I am being successful in these ways regardless or maybe hopefully in combination with how like I'm expected to be successful from the company perspective. Bethany Janos I think too whenever something defines a success criteria or a reward or like you're trying to hit certain KPIs or you're trying to hit this many commits or this many XYZ, there's always Going to be a set of people that will further game that system. And so I think that might be another trade-off of gamification, because if you're knowingly saying, I'm going to set a reward for doing that, there's always going to be a subset of people That will be like, okay, well, I'm going to hyper fixate on this exact reward. And you might not get the expected results that you're trying to get for having that reward in the first place. ([Time 0:21:29](https://share.snipd.com/snip/6ac93df2-b2c1-4513-ad69-c022893775d0)) - **Celebrate Effort To Reduce Shame** - Celebrate effort and learning aloud to normalize risk-taking and reduce shame around mistakes. - Give constructive feedback plus positive recognition to reinforce growth culture immediately. Transcript: Erica Yeah, I think another way to kind of take back power too is like being a good human to people around you. I really appreciate compliments and like celebratory moments and like anyone can make that happen right like if you see somebody who like man they got stuck in a terrible incident and Like you know they probably feel like this was awful like I didn't know how to handle this or you know maybe they were stuck in a really tough situation like celebrating that person for Like stepping in trying or like trying something new like you can be that voice that's like it's okay you did a great job um or like on a pull request review like hey like you know this approach Needs change but like here's positive feedback as well for trying and i i experienced that and feel like that goes a long way. Jonathan Tamsut Well, Erica, I think you're doing a great job leading this podcast. I celebrate you. Erica Thanks. I'm looking for a gold star at the end of this. Jonathan Tamsut Yeah, it is funny how some people, yeah, I mean, you know, you deal with, you work with all different personalities and some people are more withholding of their praise. Cool. Erica Thank you all so much for exploring this topic with me. ([Time 0:26:56](https://share.snipd.com/snip/607454dd-1971-429b-b676-0866172514a1)) - **Hosts' Favorite Game Worlds** - Each host named a favorite game character: Erica chose Street Fighter and Neopets, Jonathan chose Mario, Brittany recalled Star Wars Galaxies, and Bethany picked Stardew Valley. - These choices reflected nostalgia, comfort, and different motivations like chill or challenge. Transcript: Erica Because we're talking about gamification and inspired by video games. So I can start and this, my answer, I have two answers to this question which both reveal how long it's been since i've meaningfully played any video games so the first answer is street Fighter i think it's really fun to like live in that grungy universe and like kick ass all the time so that's my first answer and then the other complete opposite end of the spectrum is Neopets, because everything is so cute and like fluffy things all the time. Jonathan Tamsut Do you still play Neopets? Because I know someone who's actually pretty involved in the Neopets community. Erica Do you really? I know somebody who like recently revived their account. Jonathan Tamsut I have not tried but um yeah i might it's a robust community i've learned um i can go yeah i i'm yeah i'm also not i'm not a huge gamer um i i mean growing up i played video games but yeah i'd pick Mario because you know i like that he's in the trades he's a plumber i think you know working with your hands is kind of cool he's also like an italian guy but he was invented by jap you know A japanese company which i think is like kind of interesting funny and i watched the super mario brothers movie not with a child just like with my friend we just went to the theaters and Saw it it was it was honestly one of the best movie going experiences i've had i recommend a movie that's's good. Brittany Ellich That would be a good one to see in theaters. I'm sad that I didn't see that one in theaters. My answer is also reflecting on how long it's been since I've played a lot of video games. I thought this is like one of the earliest ones that I played. I used to play a lot of MMO RPGs, which are like the massive multiplayer online games like World of Warcraft and one of my very favorite ones uh was called star wars galaxies it was like A really long time ago but it had one of the best skill trees of any game like i've ever played and it was like the most dynamic and interesting skill trees and ways to like get new skills And i still think back to it all the time when i'm like you know closing out issues and like trying to knock off things on my checklist like I felt like it was a very satisfying way to think About building skills so I still think about that game a lot so I think it'd be fun to be in that game world plus then you could go to be in Star Wars you could you know fly around on what rockets They're not called rockets they're called something else in Star Wars but yeah anyway that's what I would yeah yeah you could be a Jedi how cool would that be? Like, I want Jedi powers. That'd be so cool. Erica I feel like this is making me think of a GitHub feature request that would never actually fly, but maybe we could like crowdsource it as like contribution graph skins. Like, if you could create like a Star Wars galaxy skin for your your github contributions that would be really cool we could i mean we do we do change it to spooky theme for halloween that's Brittany Ellich True that's true but i like that idea of like okay if you get so many you know commits in a specific repo then you can yeah i like that idea bethany which would you be in i would probably choose Bethany Janos Like stardew valley it's just such a cute game i oh it's my comfort game i always go back to it and love it so you can do so many things you can just chill or you can go fight monsters in the in A cave and i love that the duality ([Time 0:28:18](https://share.snipd.com/snip/39975248-a915-4a9d-b43a-9ee802d6afa6))