Details into my Data Analytics, Product Management and Player Insights role at a VC funded gaming studio. Discussing the intricacies of my position, what I offered the organization and showcasing valuable Tableau dashboards that lead to the organization and game designers making data driven decisions based on player interactions and sentiment.
In 2023 I got a job as a player insights analyst at a VC funded gaming company that was creating a free-to-play cross-platform multiplayer MOBA arena-battler game. During my role here I was hired to do player research and analysis to derive insights from our players to help inform game design and business decisions. Upon arrival at the organization, there was no analytics infrastructure set up. My role quickly shifted to facilitating the process for implementing, maintaining and developing the analytics needed to gain insights from our players. During this process I wore multiple hats and got the opportunity to grow my knowledge of gaming analytics in multiple avenues. Below lists the details what I took on in this role, and also showcases 4 primary Tableau dashboards that came from this process.
I acted as the Product Manager and Data Analyst in charge of overseeing the entirety of analytics for a cross-platform multiplayer MOBA arena battler game. Acted as a one-person team to establish, maintain and develop the analytics infrastructure for the organization. This encompassed:
Additionally assisted in User Research and User Testing via onsite gaming cafes, Discord groups as well as using Playtest Cloud software.
Tools used: SQL, Tableau, Jira, Gitlab, Perforce, Playfab, Unity, Unreal, Google Suite, Discord
*Please note that none of these dashboards reflect an accurate representation of a particular game's real-life data. Data has been modified and names and dates have been hidden to protect the game's data privacy *
When creating dashboards we like to align the visualizations to answer a specific business question. The driving business question behind this visualization was "As players first encounter our game, how many players are we loosing at each step of the new player experience?" This dashboard allows for this question to be investigated as it showcases the player journey with the game from initial app open -> account creation -> first match completion-> fifth match completion. This funnel showcases how much player drop off happens at each step of a new player's journey into the game. The hover feature allows for the ability to see the % drop off at each step of the funnel (not visiblt in the image) and the filters on the side allow viewers to drill down into specific segments they might want to see such as the country the player opened the app from, the platform the players played on or see specific dates of time that these interactions occurred.
This dashboard contains 3 standard KPIs that gaming company stakeholders might want to see on a daily basis. These charts answer the questions:
The filter abilities allow for multiple ways to understand these KPIs more. For starters, the date filter lets you choose a period of time to view, so if you launch an update with a new feature and what to see how these numbers changed in that period, the date filter allows for easy visibility into any date section a viewer might need. The 'Last 15 days' filter ensures this daily monitor dashboard only shows the most recent 15 days worth of data. As the game progresses, having the ability to see every single day since the beginning of app launch becomes overwhelming and not visually appealing nor useful for the viewer. This filter automatically takes the most recent data refresh and the prior 15 days and showcases the corresponding line chart metrics appropriately. The 'Period' filter allows for the ability to no longer see these charts by day, but rather now view by aggregating and/or averaging by week or month. This subsequently now turns the DAU into WAU or MAU and allows for the viewer to see peak and average CCU by week or month as well as see new players and average time spent in the game per player per week or month. Refer to the image below to see this filter in action showcasing these graphs by week.
The Country and Platform filters work as described: filtering down to see data just for players in a specific country or on a certain platform. Considering this was a cross-platform game, this was very useful in showcasing the difference in interactions on PC vs mobile as well as drilling down into Android vs iOS or Mac vs Windows. As the game progressed, stakeholders wanted an ability to see into difference levels of engagement, and how more engaged players were experiencing the game. As for many mobile games, there might be players who download the game, play a match and then delete instantly, this can skew averages when trying to analyze the current active player base. To drill down into various levels of engagement, we implemented a First, Third and Fifth match complete filter. Allowing for the ability to see only the players who have successfully reached match completion on one, three, or five matches. This filter was quite useful, especially in analyzing the discrepancy between each of these levels of engagement. The image below showcases how the most engaged player base —fifth match completed — interacted with the game. One can see the total time spent per player per day increased drastically compared to the entire player base (tripled!) as well as see the new accounts created and DAP decreased drastically (around 90% of players lost) This helped inform design, product and executive teams on how to approach game changes to get these player's who fall off so quickly to make it to future match completions.
Building off of these daily monitoring KPIs, a very important indicator of a game's success relies in its retention rate. This retention dashboards allows for the ability to see both classic and rolling retention rates and the exact number of retained players. Offering both of these metrics side by side allows for the ability to see how the more lenient rolling retention compares to the stricter classic retention. The side panel shows the retention % for key dates, those being D1, D3, D7, D14 and D30 (upon screenshot, the game had not been out for 30 days so there is no D30 calculated yet).
The filter abilities in the other dashboards are also available here. The cohorts based off of various number of matches completed served to be useful in monitoring retention as well. Refer to the image below to see the fifth match complete filter applied to the Retention dashboard. One can see that those who completed five matches had a much higher retention compared to the overall player base (10% -> 68%). However the total number of players this is based off of is strikingly lower; only having around 1,400 compared to 64,000; this is a key indicator of players churning out of the game very quickly before even getting to experience the core gameplay loop. However, those who do stay seem to retain much higher. These discoveries lead to prioritizing improving the new player experience, and implementing measures to have players want to stay around to complete more matches, leading to a higher retention.
The final dashboard I'm choosing to showcase in this post is an engagement match level dashboard. The game in which my team was working on was a 5v5 multiplayer MOBA arena-battler game, in the game there are 10 minute 5v5 matches in which players are able to switch between a variety of characters and skins inside each match. The game ends when the 10 minutes are up, or when one of the team reaches 1000 points, where points are gained by contesting capture points around the map. Given this context, there was a need to understand which characters were picked the most, as well as which skins. In addition, understanding which condition matches were being completed under (10 minutes reached or max points) as well as the score discrepancies helped game designers establish an overview of base balancing mechanics of their game. The name of the characters have been blocked due to privacy concerns, but there were eight characters player's could choose from in each match as well as each character having 1-3 skins to choose from. This dashboard assisted our team in finding out pain points for our players — such as unintuitive UX in the ability to change skins for your character, seen in the default skin being by far the majority for all of these characters. IN addition, the filter abilities allowed for further in-depth discoveries of segments of the player base and how more engaged players interacted with the character and skin selections. The final score discrepancies and match completion were closely monitored by our game design team, and helped them analyze how these metrics changed with game mechanic changes pushed on update. For example, a seven minute match mode was tested at one point, but was quickly shut down due to the match completion condition leaning heavily towards time reached, and the score discrepancies becoming more drastic. This showcased the inability for comeback mechanisms for teams that fell behind, and lead to more 'washes' instead of closer more engaging matches, which was the initial goal behind the implementation of shorter match durations to begin with. Being able to have data driven insights behind player interactions in the game allows for faster iterations from the design and product team to cater the game to align with player wants and needs.
In my role here, I had the opportunity to merge my technical and creative sides while working on a game I'm passionate about. It allowed me to refine my SQL and Tableau abilities within the gaming industry's specific metrics and KPIs. One thing I didn't expect was the shift in my responsibilities. Initially hired for analysis, I found myself in a role where I was the sole person tasked with analytics and had to set up the infrastructure from scratch. As a result, my role quickly evolved into that of a product manager. Instead of focusing mainly on data analysis and SQL/Tableau, I spent the majority of my time facilitating the analytics process to create the necessary end dashboards for various departments. I discovered a passion for collaboration and organization inherent in being a product manager. Collaborating with game devs, engineers, marketing, game designers, and QA to create a roadmap for features, from conception to QA validation, was both exciting and essential for our analytic platform's success. Embracing agile methodologies, I created product backlogs, managed Jira sprints, and participated in retrospectives to refine our processes. I also enjoyed applying systems thinking to technical requirements and QA testing. With analytics, there are numerous edge cases to consider when engineers create events or data scientists write SQL queries. Thinking through every possible use-case and how player journeys impact the metrics we report was akin to solving a puzzle. Addressing QA logs of invalidated charts felt like being a detective, identifying issues in the analytics pipeline. Additionally, collaborating with the community and incorporating their feedback and data-driven discoveries into new product specifications created a valuable feedback loop. This allowed us to continually iterate on the analytics product, providing value throughout development stages.
I'm passionate about being a product manager in the gaming industry, where I can leverage my deep gaming expertise alongside my technical skills in data analysis and visualization. Combined with my proficiency in UX and user research, as well as my knack for organization and collaboration, I will be a catalyst for success, not just for myself but for my team and our products. I'm committed to further honing my product management skills in this field I am so passionate about and eagerly anticipate embracing new opportunities for my growth in product management ahead.