[{"data":1,"prerenderedAt":231},["ShallowReactive",2],{"e9a489c0-b052-4414-a39c-6b4ad90adfe0":3,"jobs-count":75,"lateste9a489c0-b052-4414-a39c-6b4ad90adfe0":76},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"date":8,"id":9,"collection":5,"content":10,"coverImage":7,"author":46,"categories":47,"card":52,"order":18,"slug":59,"title":60,"uri":58,"url":58,"meta":61,"canonical":18,"social":64,"_id":70,"_type":71,"_source":72,"_file":73,"_stem":74,"_extension":71},"/stories/e9a489c0-b052-4414-a39c-6b4ad90adfe0","stories",false,"","2024-02-20T23:00:00.000000Z","e9a489c0-b052-4414-a39c-6b4ad90adfe0",[11,14,20,22,26,28,32,34,38,40,44],{"type":12,"text":13},"text","\u003Cp>Using data has been essential to Outfit7’s success. As the company grew, it started relying more and more on those of us in the Analytics department in key decision-making processes. It was obvious that our department would have to grow to keep up with what the company needed, but we didn’t want to bloat the team and face diminishing returns. The way we were able to keep the Analytics team compact despite an increasing workload is by automating repetitive high-level tasks.\u003C/p>\u003Cp>At the moment, our analytics team is comprised of 12 data scientists from a mix of backgrounds, like mathematics, psychology, computer science and finance. A big part of our job is analyzing game feature performance so we can identify opportunities for improvement. But we don’t just track basic stats. All of us try to dig deep and understand why users behave the way they do. In order to have the time to do this properly, we needed to establish processes that work for us, not the other way around.\u003C/p>\u003Ch3>How we optimized AB testing\u003C/h3>\u003Cp>In the gaming and web industry, AB testing is a very common part of our lives. Outfit7 has up to 350 million monthly active users, so we don’t want to make a change that would negatively affect even the smallest percentage. That’s why we always test different scenarios to create the best user experience.\u003C/p>",{"type":15,"items":16},"slider",[17],{"caption":18,"asset":19},null,"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/ibi_feb2024_21_4.png",{"type":12,"text":21},"\u003Cp>\u003Cem>Fig. 1: Here is an example of an AB test about icons. We want to use a new icon, but first we want to see which one would work best for our users.\u003C/em>\u003C/p>\u003Cp>When testing the performance of our games and app store listings, we rely on many different metrics, but three of the most important ones are:\u003C/p>\u003Cul>\u003Cli>\u003Cp>\u003Cstrong>Retention:\u003C/strong> the percentage of users that returned to the game several days after install\u003C/p>\u003C/li>\u003Cli>\u003Cp>\u003Cstrong>Session duration:\u003C/strong> how much time users are spending in the game\u003C/p>\u003C/li>\u003Cli>\u003Cp>\u003Cstrong>Lifetime value:\u003C/strong> how much revenue the average install is generating\u003C/p>\u003C/li>\u003C/ul>\u003Cp>We include these three metrics in every report and look at them from different perspectives by creating multiple visualizations and analyses. This used to mean that we spent a large chunk of our time plotting metrics for AB test reports. Looking back, we were constantly answering the same types of questions and performing similar tasks every time we did a test.\u003C/p>\u003Cp>Actually, we still do all of that, but now it only takes one simple query or even just a click to prepare the data. Before automation, one data scientist could analyze a handful of tests per month, along with one additional bigger task. Now we can do all that in a week.\u003C/p>",{"type":15,"items":23},[24],{"caption":18,"asset":25},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/ibi_feb2024_21_3.jpg",{"type":12,"text":27},"\u003Ch3>The road to automation\u003C/h3>\u003Cp>Broadly speaking, these are the most important steps to automate a process in data analytics:\u003C/p>\u003Col>\u003Cli>\u003Cp>Select what you want to automate\u003C/p>\u003C/li>\u003Cli>\u003Cp>Make sure everyone on the team can safely update the code\u003C/p>\u003C/li>\u003Cli>\u003Cp>Create a daily job to execute the code\u003C/p>\u003C/li>\u003Cli>\u003Cp>Create an easy setup environment \u003C/p>\u003C/li>\u003Cli>\u003Cp>Create a nice UI to display the results\u003C/p>\u003C/li>\u003C/ol>\u003Cp>The first step was clear for us: we knew we wanted to automate AB test reports as much as possible. First, we selected all the metrics that were in every report and weren’t app-specific – I already mentioned the three main ones above. These metrics form the high-level basis that we start from before digging deeper into specific user data.\u003C/p>\u003Cp>However, there was another problem we needed to solve before we could automate reporting. We generate terabytes of data stored in billions of rows every day. That means we need to constantly optimize to avoid bloated storage and long query times.\u003C/p>\u003Cp>Because we’re usually only interested in a limited number of metrics, we created user aggregates. That way, we only have one table per day that stores each active user in one row. So, for instance, instead of having sixteen rows for a user that opened the app 16 times, we only have one row with a column that tells us how many times the user opened the app. This allowed us to reduce the size of daily tables to 0.5%, resulting in faster execution time for queries and lower costs for both querying and storage – win, win, win.\u003C/p>",{"type":15,"items":29},[30],{"caption":18,"asset":31},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/ibi_feb2024_21_5.png",{"type":12,"text":33},"\u003Cp>\u003Cem>Fig. 2: An example of aggregated data. One user from Brazil had 5 sessions (sessNum) on 18th June 2019 and stayed in the app for 19 minutes (calculated from sessDayTime in milliseconds).\u003C/em>\u003C/p>\u003Cp>As awesome as these aggregates are, they have one problem: we lose the order in which a user performed actions. We can’t know if they watched a video and then fed Talking Tom, or the other way around. Because we may need this information, we still save all the raw data. \u003C/p>\u003Ch3>Getting everyone on the same page\u003C/h3>\u003Cp>After we decided which metrics we want to have calculated automatically, we had to make sure everyone on the team can update the code and do it safely. The first step here is to decide where and how the code will live. This is important, because all team members should be able to change or add something to it, so everyone should know the language. \u003C/p>\u003Cp>We were already using BigQuery from Google, as it allows us to handle the massive amounts of data we get every day. However, half of our team was using Python for additional pivoting and visualization and the other half was using R. In the end, we all sold our hearts to R, because of its awesome ggplot package.\u003C/p>\u003Cp>Now that we decided to use R, we had to make sure everyone can update the code safely. We used to rely on iPython notebooks for that, but that could quickly get messy as there’s no version control. We started developing and testing our code locally and – finally – migrated to Git.\u003C/p>\u003Ch3>Let the job do the job\u003C/h3>\u003Cp>Having set up all the code, we still needed to run it every day. Since our goal was to decrease manual work as much as possible, we had to find an automated solution. When we migrated to Git, we also started using a scheduling tool to run our code. Besides doing a lot of work for us, this also made our code updates safer and improved error handling.\u003C/p>\u003Cp>However, since we had so much data, the daily jobs set up in AirFlow still took forever to execute. When you need to import more than a million rows of query results to the server, it can take a long time. Part of our solution was to move the calculation of tests to BigQuery, but it wasn’t enough. The solution was to stop calculating metrics from raw data. Since we were already creating user aggregates every day, we just started using them for AB tests as well.\u003C/p>\u003Cp>By moving calculations of statistical tests to BigQuery and using user aggregates, we decreased execution times from more than one hour per test to just a couple of minutes. \u003C/p>",{"type":15,"items":35},[36],{"caption":18,"asset":37},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/ibi_feb2024_21_1.jpg",{"type":12,"text":39},"\u003Ch3>Set up should take minimum effort\u003C/h3>\u003Cp>Our daily job now knew what to plot for each metric, but it didn’t know which AB test this needs to be done for. To add an AB test to the daily job, we need to write info about it somewhere, but preferably not directly in the code. A detached file or database works best.\u003C/p>\u003Cp>Since we were already using Google services, we decided to use Google Sheets to store internal test configuration. Back when we were using iPython notebooks, access to Sheets was really difficult without writing your username and password on the server. But since we migrated to BigQuery, we were able to use an awesome built-in feature to directly access Sheets.\u003C/p>\u003Cp>To add an AB test, all we need to do is write down a short description of groups and wait until the next morning to let the daily job do its thing.\u003C/p>\u003Cp>That’s a broad overview of how we optimized the way we create reports on AB test. We also built in a few other useful features, like a column where you can create different data selections – for example the platform used (“iphone”, “ipad”). This filter is then applied to all plots automatically and displayed in addition to non-filtered ones. We use this to see what kind of effect the test has on different segments. You can filter by platform, country, install date or a mix of factors. \u003Cem>Fig. 3: If we need results from one AB test on different segments, we add rows with defined clusters. In this AB test, we were interested in users from the US using iOS only.\u003C/em>\u003C/p>\u003Cp>\u003C/p>",{"type":15,"items":41},[42],{"caption":18,"asset":43},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/ibi_feb2024_21_6.png",{"type":12,"text":45},"\u003Cp>\u003Cem>Fig. 3: If we need results from one AB test on different segments, we add rows with defined clusters. In this AB test, we were interested in users from the US using iOS only.\u003C/em>\u003C/p>\u003Ch3>Last step: looking good\u003C/h3>\u003Cp>Now that we have the code in R, everything in Git, a daily job to create plots and an easy way to add new AB tests, we just need to easily display everything so we can create reports. \u003C/p>\u003Cp>We use a package in R called Shiny to create a dashboard where we can view, browse and group metrics. This isn’t just nice to have – because we have a huge amount of different metrics, being able to quickly sort through them in a neatly designed dashboard is essential for efficiency.\u003C/p>\u003Cp>This entire system took some effort to implement and perfect. However, it saved us a much larger amount of time in the long run. When we run a new AB test, we add one line to a Google Sheet and return the next morning to collect our neatly generated plots.\u003C/p>\u003Cp>The daily job collects all aggregated data for that test, manipulates it as required, calculates statistics and plots everything, preparing enough plots to create reports as long as 50 pages. Instead of losing time on performing high-level tasks, our team can now dig deeper for additional insights or do advanced analyses on things like game performance or user churn. \u003C/p>\u003Cp>By automating as much of our work as possible, we freed up time and resources to dig deeper into our field and get to all the tasks that used to get pushed down the priority list. It makes us more efficient and it makes our jobs a lot more satisfying.\u003C/p>\u003Cp>Disclaimer: This article does not address personal data collection, processing, storage or sharing. Please note that nothing in this article should be interpreted as commenting or assessing personal data compliance. Outfit7 handles personal data and privacy with the utmost seriousness and is in compliance with GDPR and all relevant compliance regulations.\u003C/p>\u003Cp>\u003C/p>","Sara Krk, Senior Director of Analytics",[48],{"id":49,"title":50,"slug":51},"stories_categories::behind-the-fun","Behind the fun","behind-the-fun",{"title":53,"copy":18,"backgroundImage":54,"cta":55},"How We Optimized AB Testing at Outfit7","https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/card_feb20234_21.jpg",{"title":56,"type":57,"href":58},"Read the story","link","/blog/stories/automating-repetitive-tasks-how-we-optimized-ab-testing-at-outfit7","automating-repetitive-tasks-how-we-optimized-ab-testing-at-outfit7","Automating Repetitive Tasks: How We Optimized AB Testing at Outfit7",{"title":62,"description":7,"canonical":63,"robots":7},"Automating Repetitive Tasks: How We Optimized AB Testing | Outfit7","http://localhost/blog/stories/automating-repetitive-tasks-how-we-optimized-ab-testing-at-outfit7",{"open_graph":65,"twitter":65,"site_name":66,"title":67,"description":68,"image":69},true,"Outfit7","Automating Repetitive Tasks: How We Optimized AB Testing","And handle 4TB of data every single day.","https://cdn-o7.o7web.com/img/asset/YXNzZXRzLzRfYmxvZ19uZXdzL2xpZmVfYXRfb3V0Zml0Ny9mYi1saXBvc3RfZmViMjAyNF8yMS5qcGc=?w=1200&h=630&q=70&fit=crop&s=bbd866169c6b29a0c39eb068479c31e0","content:stories:e9a489c0-b052-4414-a39c-6b4ad90adfe0.json","json","content","stories/e9a489c0-b052-4414-a39c-6b4ad90adfe0.json","stories/e9a489c0-b052-4414-a39c-6b4ad90adfe0",13,[77,126,182],{"_path":78,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"date":79,"id":80,"collection":5,"content":81,"coverImage":7,"author":104,"categories":105,"card":110,"order":18,"slug":115,"title":116,"uri":114,"url":114,"meta":117,"canonical":18,"social":120,"_id":123,"_type":71,"_source":72,"_file":124,"_stem":125,"_extension":71},"/stories/fcb7ee7f-cb90-45fe-a8ca-9aed12b0428d","2024-11-14T23:00:00.000000Z","fcb7ee7f-cb90-45fe-a8ca-9aed12b0428d",[82,84,88,90,94,96,102],{"type":12,"text":83},"\u003Cp>One of my favorite things about working at Outfit7 (Ekipa2 subsidiary) is that if you identify something that your team needs, or that another team needs, or even that the company needs, everyone will hear you out and help make it happen. All it takes is enthusiasm, passion, and belief!\u003C/p>\u003Cp>And I know this from experience because this is exactly how UX (User Experience) research started at Outfit7, and how I forged the path to my current role as Senior Lead UX Researcher.\u003C/p>\u003Cp>The UX Research team is a bridge between Outfit7 players and the product team. By understanding Outfit7 users and their needs, we can make Outfit7 games clear, intuitive, and enjoyable to play. While on the other hand, we also have to understand what the product team needs to help them make informed decisions.\u003C/p>\u003Cp>And I love doing this! But I must admit that at the start of my career I had a very different career path in mind.\u003C/p>",{"type":15,"items":85},[86],{"caption":18,"asset":87},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_ibi_oct24_30_arhar5.jpg",{"type":12,"text":89},"\u003Cp>\u003Cstrong>Exactly what I’d been missing\u003C/strong>\u003C/p>\u003Cp>I studied marketing communication studies and, when I was a student, my dream was to work on marketing campaigns and TV advertising. And that’s why I started working at an advertising agency when I was still a student. But I pretty quickly figured out that research was actually the aspect of marketing and advertising that appealed to me most. It turned out that I was more interested in understanding the audience than in the creative process of producing ads and marketing campaigns.\u003C/p>\u003Cp>So, after a while I switched jobs to a market research agency. At first I was mostly working on data-driven quantitative analysis. But then the agency’s management realized they needed more support with qualitative research, and asked me, “Can you jump in and help us here?” And I said, “OK, let’s try it.” It didn’t take long to discover that I loved this kind of work. After running just one focus group, I was hooked!\u003C/p>\u003Cp>I enjoyed my time at that agency and progressed steadily through the ranks, but something was missing. I wanted to follow through on my work, to make sure my analysis was being implemented properly. But I couldn’t do that because I wasn’t part of the client’s company. And this is why, in 2017, I applied for a market research position at Outfit7 (Ekipa2 subsidiary).\u003C/p>\u003Cp>During the hiring process, when I was discussing my background with the hiring manager, Jernej Česen, qualitative research and web page testing came up. So, we started discussing it, and if and how we could apply some principles from those fields to games. In the end I was hired as a market researcher, but sure enough, after only about 6 months, I switched to UX research.\u003C/p>\u003Cp>Up until that point, UX research hadn’t formally existed at Outfit7. Product managers would user-test games, but mostly just before release, only really at the surface level, and in a way that was not as structured as we eventually started doing it.\u003C/p>\u003Cp>I felt that as a UX researcher I \u003Cem>was\u003C/em> having an impact. It was eye-opening in a  lot of ways, and we quickly started changing things. I got exactly what I’d been missing in my previous job.\u003C/p>",{"type":15,"items":91},[92],{"caption":18,"asset":93},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_ibi_oct24_30_arhar2.jpg",{"type":12,"text":95},"\u003Cp>\u003Cstrong>Initiative made it happen\u003C/strong>\u003C/p>\u003Cp>At around that same time another colleague joined the company as a dedicated UX designer. Inevitably, our paths crossed and we realized that we should connect our work. I’m proud that we connected on our own, figuring out how we should collaborate together as we went along. It was our decision to work closely together, and we used our initiative to make it happen. We realized that we needed each other and that we could support each other, and that’s how the UX team started to take shape.\u003C/p>\u003Cp>So, at that point, both the research and design sides of UX were “one-person bands”, and we needed to figure it out between the two of us. We wanted to try running a more detailed test of a game to see how the process and the results would look. It went well, and we got buy-in, but our role was still “on-demand.” We would be asked to test a game when it was almost done, so my role was reactive to the needs at a given time.\u003C/p>\u003Cp>But then we started to become more and more proactive, trying to establish some processes whereby user research could support the game development process. We didn’t just want to be reacting to needs as they arose, we wanted to figure out what UX needs were going to be \u003Cem>before\u003C/em> game features were fully implemented, because of course it takes much more time to fix issues on a near-complete game. So the role, responsibilities, and focus of the UX team developed over time. \u003C/p>",{"type":15,"items":97},[98,100],{"caption":18,"asset":99},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_ibi_oct24_30_arhar1.jpg",{"caption":18,"asset":101},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_ibi_oct24_30_arhar4.jpg",{"type":12,"text":103},"\u003Cp>\u003Cstrong>Putting user testing to the test\u003C/strong>\u003C/p>\u003Cp>I remember that at the start of the Talking Tom Hero Dash project, there was a lot of discussion about whether players would understand that they were supposed to hit the Rakoonz, and not avoid them. Half of the team said that players wouldn’t hit them  because they wouldn’t want to hurt them, and the other half said that they would just hit all of them because they wouldn’t care. I just said, “Guys, why don’t we test this?” We had never tested prototypes up to that point, but it turned out to be quick and easy to do. \u003C/p>\u003Cp>And the players in the user test \u003Cem>did\u003C/em> hit the Rakoonz, but they rationalized what they were doing and put it into nice words. So, they were \u003Cem>picking up\u003C/em> the Rakoonz, and \u003Cem>catching\u003C/em> them, not \u003Cem>hitting\u003C/em> them or \u003Cem>hurting\u003C/em> them. And we wouldn’t have known that without testing it. Everyone was just arguing about it, and no one quite had it right.\u003C/p>\u003Cp>When I started, assumptions like that often came down to who could present their opinion most strongly. But now when a team is having these kinds of discussions, they’ll say, “Let’s ask UX Research if they can test this and find out what the users think.” This is such positive progress, and I’m so proud that it started happening because just two of us had the enthusiasm to take the initiative. I love that things like that can happen here!\u003C/p>\u003Cp>I also love the people at Outfit7. I can honestly say that I’ve made a lot of really good friends thanks to this job - like, really good, close friends - and I hope I will keep meeting new people and making more new friends. You can talk to basically everyone, and everyone can talk to you, and you can be really honest and open. So, I like this little community that we have, both as a whole Outfit7 group, and also within the environment of the people I have most contact with.\u003C/p>\u003Cp>And finally, I’m still really passionate about Talking Tom & Friends - I always have been. I love this brand and its audience. They are the most appreciative audience out there. When I see them playing Outfit7 games, and I see their genuine reactions and how passionate they are about it; that’s when I know \u003Cem>why I do this\u003C/em>. It’s so rewarding and fulfilling to see Outfit7 games making their audience feel good, and that’s why I’m so happy I took the UX research path.\u003C/p>","Anja Arhar, Senior Lead UX Researcher",[106],{"id":107,"title":108,"slug":109},"stories_categories::my-outfit7-story","My Outfit7 story","my-outfit7-story",{"title":111,"copy":18,"backgroundImage":112,"cta":113},"How Initiative Shaped the UX Team","https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_card_aug24_23_oct24_30_arhar.jpg",{"title":56,"type":57,"href":114},"/blog/stories/how-initiative-and-enthusiasm-paved-the-way-for-the-ux-team","how-initiative-and-enthusiasm-paved-the-way-for-the-ux-team","How initiative and enthusiasm paved the way for the UX team",{"title":118,"description":7,"canonical":119,"robots":7},"How initiative and enthusiasm paved the way for the UX team | Outfit7","http://localhost/blog/stories/how-initiative-and-enthusiasm-paved-the-way-for-the-ux-team",{"open_graph":65,"twitter":65,"site_name":66,"title":116,"description":121,"image":122},"Makers of Talking Tom & Friends – Fun Games for All Ages","https://cdn-o7.o7web.com/img/asset/YXNzZXRzLzRfYmxvZ19uZXdzL2xpZmVfYXRfb3V0Zml0Ny9mYi1saXBvc3Rfb2N0MjRfMzBfYXJoYXIuanBn?w=1200&h=630&q=70&fit=crop&s=0ef94a251edc986559da2e13a8411f31","content:stories:fcb7ee7f-cb90-45fe-a8ca-9aed12b0428d.json","stories/fcb7ee7f-cb90-45fe-a8ca-9aed12b0428d.json","stories/fcb7ee7f-cb90-45fe-a8ca-9aed12b0428d",{"_path":127,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"date":128,"id":129,"collection":5,"content":130,"coverImage":163,"author":164,"categories":165,"card":167,"order":18,"slug":172,"title":168,"uri":171,"url":171,"meta":173,"canonical":18,"social":176,"_id":179,"_type":71,"_source":72,"_file":180,"_stem":181,"_extension":71},"/stories/6d8856a2-a369-4fed-839a-eb33be748477","2024-10-09T22:00:00.000000Z","6d8856a2-a369-4fed-839a-eb33be748477",[131,133,137,139,143,145,149,151,155,157,161],{"type":12,"text":132},"\u003Ch2>Why There’s Nothing Quite Like a Game Jam\u003C/h2>\u003Cp>We’ve just wrapped up our latest internal game jam, which was dedicated to one of our newest apps. Which app was it? That’s a secret, \u003Cem>for now\u003C/em>. However, seeing as Outfit7 has a long history of game jams in various forms, I’d like to show you how we host our game jams, and share some insights and stories from some of their participants. So, let’s go! \u003C/p>\u003Cp>The process of making a game can vary a lot and become really messy and crazy, yet for game jams that’s normal. Finding and assembling a team, generating a bunch of ideas, narrowing them down to just one per team, and then implementing it in a very short space of time can get pretty intense. So, it’s important to be surrounded by the right people. Luckily, at Outfit7 that’s pretty much everyone! \u003C/p>\u003Cp>“It truly felt like controlled chaos,” says \u003Cstrong>Daria Akulova, 3D Artist at Outfit7 (Ekipa2 subsidiary)\u003C/strong>\u003C/p>\u003Cp>Have you ever thought about what it takes to organize a game jam? Where do you start? What’s the theme going to be? How are you going to make it happen? Over to \u003Cstrong>Marko Spasojević\u003C/strong>, \u003Cstrong>Senior Game Designer\u003C/strong> at Outfit7 (Ekipa2 subsidiary) who is the go-to-guy for Outfit7 game jams, to guide us through this process. \u003C/p>",{"type":15,"items":134},[135],{"caption":18,"asset":136},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/outfit_gamejam-portraits_websize-(50-of-100).jpg",{"type":12,"text":138},"\u003Cp>“The first question you need to ask yourself is, what are you trying to achieve with this particular game jam? Is there a team that could benefit from some new ideas or gameplay mechanics? Or do you just want people to have fun, connect with each other, and do something unusual? In the former case, you consult closely with other game designers and product teams to see what they would like to get ideas for. Otherwise, you just let people unleash their creativity!\u003C/p>\u003Cp>“What makes game jams so special to me is that they allow people to express their creativity beyond their everyday duties and responsibilities. A developer can become an artist, a UI Artist can become an animator, an animator can become a game designer, and they’re all working together to create something fun.”\u003C/p>\u003Cp>The announcement of a new game jam begins the countdown to all the fun. Teams start assembling and imaginations start generating ideas. You might think that in order to make a game in a limited amount of time, you need to be really good at what you’re doing. Well, \u003Cstrong>Jure Vukoja\u003C/strong>, a member of the winning team of our latest game jam, spills the beans on how this experience was for them.\u003C/p>",{"type":15,"items":140},[141],{"caption":18,"asset":142},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/20240831_100811-(1).jpg",{"type":12,"text":144},"\u003Cp>“We took this as an opportunity to do something that we usually do not do, hence the name - Level 1, indicating that the majority of us were new to our roles. Quality engineers were writing code, developers creating sounds, UI artists making concept art, and so on. So basically the condition to join was that you had to be willing to do something you are not the most comfortable with.”\u003C/p>\u003Cp>“When you have only two days to implement the game, nothing goes according to plan. Time pressure is very tight and we had to make the most of what we had. One of our team members having to get emergency surgery halfway through the game jam definitely was not in the plan, but you have to adjust quickly. We had to reshuffle responsibilities a bit and used a bit of modern technology, namely AI, to speed up the process. In a game jam the only rule is you have to deliver a game by the final call, so you have to use every resource you have to the absolute max.”\u003C/p>\u003Cp>“We assembled a better team than The Avengers,” says \u003Cstrong>Tomaž Rojc Software Engineer at Outfit7 (Ekipa2 subsidiary)\u003C/strong>\u003C/p>",{"type":15,"items":146},[147],{"caption":18,"asset":148},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/outfit_gamejam-portraits_websize-(44-of-100).jpg",{"type":12,"text":150},"\u003Cp>“We had so many ideas during brainstorming that we ran out of sticky notes to write them down on,” says \u003Cstrong>Stefan Ivašković\u003C/strong>, \u003Cstrong>Senior Animator at Outfit7 (Ekipa2 subsidiary)\u003C/strong>. “Since the topic was Different Perspective, we were trying to come up with the most weird ideas that we could imagine. It took us about three hours just to select one idea that everyone would love to work on. And as soon as we started development, we realized that we did not know how to implement some of the core gameplay, and that we did not have time to learn how to do it, so we needed to think outside the box. A core element of the gameplay was different shaped 2D characters on surfaces inside of a 3D cube. Nobody was familiar with 2D animation, so we had to invent a whole pipeline to have our characters animated and looking good while still delivering on time.”\u003C/p>\u003Cp>\u003Cstrong>Mikhail Pomakhin\u003C/strong>, \u003Cstrong>Senior Software Engineer at Outfit7 (Ekipa2 subsidiary)\u003C/strong> says that the biggest challenge in a game jam is the time limit. “You have only 24 hours in one day and two days in total to make it, so you have to work as a true team. There were challenges that we unfortunately did not have time to solve properly, one of which was a tutorial. We did not have time to implement a proper tutorial that would explain all the game mechanics and controls thoroughly, so we had to make an image that would appear before the game starts and explain everything, just like you’d often get in older games. This did not work out well, as people were often just closing the pop-up, yet we were not disappointed as we knew that in the end we achieved our main goal - we made an amazing game in such a short time!”\u003C/p>",{"type":15,"items":152},[153],{"caption":18,"asset":154},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/outfit_gamejam-portraits_websize-(47-of-100).jpg",{"type":12,"text":156},"\u003Cp>If game jam participants had to choose one thing that they both love and hate about game jams, that would be… time. If only there were a bit more time, one more day - everybody had these thoughts, yet the tight time limit is exactly what makes game jams so special.\u003C/p>\u003Cp>So, the final second has passed and all the teams take their hands off their keyboards, drawing tablets, and phones. It’s time to reveal all those amazing games, and let the people of Outfit7 decide which one will be named the best this time. Of course, it’s not really about winning, but a bit of recognition is a nice bonus on top of the amazing journey that you and your teammates have been on. \u003C/p>\u003Cp>The games are then evaluated across multiple topic categories, and on our latest game jam, there were four of these: \u003Cstrong>Out of the Box, Most Engaging, Prettiest Design, Winner Winner\u003C/strong>. A game is a game, and no matter the skill, age, or personal preferences, it must be fun to play for everyone — that's the ultimate criterion!\u003C/p>",{"type":15,"items":158},[159],{"caption":18,"asset":160},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/outfit_gamejam-portraits_websize-(14-of-100)-1728547665.jpg",{"type":12,"text":162},"\u003Cp>Our aim is to get as many people to try out the games as possible. On the latest game jam there were eight final games and all of them were really good! But as I said already, it’s not about winning. The main thing is that everyone who participated had an amazing time and created something truly unique.\u003C/p>\u003Cp>Game jams are a mirror of Outfit7’s culture. It’s not just about making games, it’s about creating memorable experiences! And not just for our players, but also for the people behind the games. \u003C/p>\u003Cp>Our next game jam is coming up soon! If you want to be a part of it, then check out our careers page for open positions. Maybe next time, \u003Cem>you’ll\u003C/em> be on the winning team!\u003C/p>","https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/outfit_gamejam-portraits_websize-(60-of-100).jpg","Written on behalf of Outfit7 by Vitaliy Vlasevych, Software Engineer",[166],{"id":49,"title":50,"slug":51},{"title":168,"copy":18,"backgroundImage":169,"cta":170},"Building a Game in Just Two Days!?","https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_card_oct24_11_gamejam.jpg",{"title":56,"type":57,"href":171},"/blog/stories/building-a-game-in-just-two-days","building-a-game-in-just-two-days",{"title":174,"description":7,"canonical":175,"robots":7},"Building a Game in Just Two Days!? Why There’s Nothing Quite Like a Game Jam | Outfit7","http://localhost/blog/stories/building-a-game-in-just-two-days",{"open_graph":65,"twitter":65,"site_name":66,"title":168,"description":177,"image":178},"Why there’s nothing quite like a Game Jam.","https://cdn-o7.o7web.com/img/asset/YXNzZXRzLzRfYmxvZ19uZXdzL2xpZmVfYXRfb3V0Zml0Ny9zb2NpYWxfaW1hZ2VfMTEuMTAuMjAyNC5qcGc=?w=1200&h=630&q=70&fit=crop&s=70c01f7b1443ebe9bc18edb965d86f9e","content:stories:6d8856a2-a369-4fed-839a-eb33be748477.json","stories/6d8856a2-a369-4fed-839a-eb33be748477.json","stories/6d8856a2-a369-4fed-839a-eb33be748477",{"_path":183,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"date":184,"id":185,"collection":5,"content":186,"coverImage":7,"author":211,"categories":212,"card":214,"order":18,"slug":219,"title":220,"uri":218,"url":218,"meta":221,"canonical":18,"social":225,"_id":228,"_type":71,"_source":72,"_file":229,"_stem":230,"_extension":71},"/stories/afdb0369-2aa8-4a4e-bdcd-32671396b299","2024-08-20T22:00:00.000000Z","afdb0369-2aa8-4a4e-bdcd-32671396b299",[187,191,193,197,199,203,205,209],{"type":15,"items":188},[189],{"caption":18,"asset":190},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_ibi_aug24_23_zjeram_3.jpg",{"type":12,"text":192},"\u003Cp>When it comes to junior roles at Outfit7, expertise isn’t the main thing we’re looking for. We’re much more interested in drive and a willingness to learn. And that’s how, back in 2017, I managed to land a job as a Junior Quality Assurance Engineer at Outfit7 (Ekipa2 subsidiary). Before that I’d spent about five years working in the psychology field, and that experience has actually proven pretty useful. And, of course, I’m an avid, lifelong gamer. I even used to sketch out my own game ideas on paper when I was a kid. I think all of this must have been taken into account when I was hired, because I certainly didn’t have any previous games industry experience.\u003C/p>\u003Cp>I don’t have a particularly technical mindset either, and quality assurance is usually more of a technical job. I could certainly find bugs — I literally opened thousands of Jira tickets when I was in QA — but when it came to really technical stuff like backend testing, that was not my jam at all. So, it was actually really cool that my managers didn’t push me in that direction, but instead encouraged me towards more product-centric QA tasks. It was always, “Žiga, you’re gonna check how this feature looks and feels, and what we need to improve from the player’s perspective,” which really suited my product-oriented mindset and gave me the opportunity to thrive.\u003C/p>\u003Cp>There is room for creativity in QA, but ultimately there’s a limit to how much. So, the product side of things and the design side of things always interested me more. That was what inspired me and what I always had a passion for. And if I'm going to be spending five days a week of my limited lifetime on something, it needs to really motivate me. So, over time I was pulled, quite organically, in the direction of the product department. Until one day Ante Odić, Senior VP of Product Development, came to me and said, “Hey dude, do you want to join our team?” And, of course, I did!\u003C/p>",{"type":15,"items":194},[195],{"caption":18,"asset":196},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_ibi_aug24_23_zjeram_1.jpg",{"type":12,"text":198},"\u003Ch3>Many moving parts\u003C/h3>\u003Cp>I came across the term “product manager” for the first time when I started working at Outfit7. I must admit, I didn’t initially have any idea what a product manager does, but I was immediately intrigued by it. There are many different aspects to the role, so I needed quite some time “on the field” to get all the essential parts that connect to make up product management. But by the time I moved over to product management, I had a pretty good idea of what was expected.\u003C/p>\u003Cp>My main responsibility as a product manager is to understand the vision of the game and, crucially, to make sure that everyone on the project is aligned on how we’re going to fulfill it. The amount of effort and brain power that goes into producing a game is insane if you think about it. There are all these moving parts that you need to tend to, and every one of these parts has the potential to ruin the final product. So it’s essential that you keep everybody involved aligned on exactly what we are making and why we are making it so that all these intertwined parts can come together as something solid and something whole.\u003C/p>\u003Cp>You have a lot of very smart people with a lot of different expertise that can actually influence those moving parts. And then on top of that, you have the games industry, which is very volatile and unpredictable. There are so many factors to take into account: trends, rules, platforms, consumer types, age categories… you name it. There’s a lot of stuff that needs to come together and align almost perfectly for a product to end up a success.\u003C/p>\u003Cp>But I also have to understand the broader market and the market opportunity that we’re aiming for — in other words, the business reason behind the idea. And then, of course, the follow-through. You produce something and then check what’s happening, and then pivot, or fix it, or make more of it.\u003C/p>",{"type":15,"items":200},[201],{"caption":18,"asset":202},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_ibi_aug24_23_zjeram_2.jpg",{"type":12,"text":204},"\u003Ch3>The human element\u003C/h3>\u003Cp>I got into psychology in the first place because I’ve always been really interested in people, so the human dynamics within a team are also a high priority for me. As a product manager, my day is filled with meetings involving diverse individuals, each bringing their unique background and expertise to the table. It's crucial to "keep the peace" in these meetings, ensuring everyone is aligned and has the opportunity to share their perspective.\u003C/p>\u003Cp>We all naturally pick up on subtle cues such as changes in someone's tone of voice, facial expression, or body language. But my background in psychology makes me more attuned and attentive to them. Understanding where someone is coming from is essential for me as a product manager. I strive to comprehend not just what someone is saying, but also how they are feeling about the topic being discussed. This keen insight is integral to my approach. It allows me to connect with my colleagues and find a common language, leading to better collaboration and alignment within the team.\u003C/p>\u003Cp>Another benefit of my keen interest in people is that it makes me mindful of the product’s end user. I was actually a real pain in the ass for a lot of developers and product teams back in the day, because I was constantly challenging — not just for the sake of it, but for the sake of producing a feature or product that users could enjoy. At every step of the way, you need to be mindful of how the user is going to perceive the game.\u003C/p>",{"type":15,"items":206},[207],{"caption":18,"asset":208},"https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_ibi_aug24_23_zjeram_4.jpg",{"type":12,"text":210},"\u003Ch3>People as a badge of quality\u003C/h3>\u003Cp>I believe my player-centric approach is a great fit here. Outfit7 games are all on mobile and are designed for a broad, diverse audience, so it’s important that they’re accessible and can be understood quickly and easily. Getting a game to the point where we’re confident that the end user — regardless of who they are and how much experience they have with games — will quickly understand it and start enjoying it is really quite difficult. But I think Outfit7 actually does this very well.\u003C/p>\u003Cp>And perhaps that’s because we’re so people-oriented. I’ve met a lot of really, really cool people during my years at Outfit7. People who have really influenced my career, my decision-making, my growth, and the development of my own expertise. It’s the people and culture that elevate Outfit7 above any of the companies I worked for in the past. Particularly, in my case, Mateja Vrhovnik, who was one of the managers who made sure I wasn’t burdened with technical tasks in QA engineering, and encouraged the enthusiasm and drive I have towards making games. And then there’s Jaka Robnik, Head of Product at the Ekipa2 subsidiary, who’s hands-down my favorite manager, and one of the smartest people I know.\u003C/p>\u003Cp>It’s truly a pleasure working alongside my co-workers. I think of Outfit7’s people as a kind of badge of quality for the whole Outfit7 organization. And it’s one of the main reasons I’m still here, for sure.\u003C/p>","Žiga Jeram, Senior Product Manager",[213],{"id":107,"title":108,"slug":109},{"title":215,"copy":18,"backgroundImage":216,"cta":217},"My Journey to  Senior Product Management","https://cdn-o7.o7web.com/assets/4_blog_news/life_at_outfit7/web_card_aug24_23_zjeram.jpg",{"title":56,"type":57,"href":218},"/blog/stories/my-journey-to-senior-product-management","my-journey-to-senior-product-management","With a Little Encouragement, I Was Able to Pursue My Passion",{"title":222,"description":223,"canonical":224,"robots":7},"With a Little Encouragement, I Was Able to Pursue My Passion | Outfit7","From sketching out game ideas on paper to helping bring them to life, Senior Product Manager Žiga Jeram shares his Outfit7 story.","http://localhost/blog/stories/my-journey-to-senior-product-management",{"open_graph":65,"twitter":65,"site_name":66,"title":220,"description":226,"image":227},"Senior Product Manager Žiga Jeram shares his Outfit7 story","https://cdn-o7.o7web.com/img/asset/YXNzZXRzLzRfYmxvZ19uZXdzL2xpZmVfYXRfb3V0Zml0Ny9mYi1saXBvc3RfYXVnMjRfMjNfemplcmFtLmpwZw==?w=1200&h=630&q=70&fit=crop&s=f29fb1f4840bc917da1acc76f1b2dbf2","content:stories:afdb0369-2aa8-4a4e-bdcd-32671396b299.json","stories/afdb0369-2aa8-4a4e-bdcd-32671396b299.json","stories/afdb0369-2aa8-4a4e-bdcd-32671396b299",1764606561175]