The datasets were provided in xlsx. format and were not chronologically sorted. Not to mention that the datasets included a lot of extra information that lies outside of the scope of research.

Step 1: Data Cleaning

In order to get a better understanding of the data provided, I immediately dove into cleaning the data and getting rid of junk information that lies outside of the scope of my analysis.

Since the Company is only interested in its user and billing performance over the 2016 to 2021 period, I utilized applicable functions such as SUMIF to extract the total number of MAU and billing on a month-to-month basis. This gives me a clear understanding of the total number of users that logged-in and the total amount of billing revenue accrued in a specific month.

Organizing the datasets in this chronological order also provides me with better visual clarity on the growth trends of the parameters under interest.

Step 2: Analyzing the Historical Trend

Next, I mapped month-to-month data and discovered some interesting things occured over the Observation Period.

First, looking at the line chart analysis of the MAU growth reveals a period of stagnation from 2016 to 2019, followed by a significant surge starting in 2020. This sharp increase aligns with the onset of the COVID-19 pandemic, as limited indoor activities drove an increase in screen time.

By the end of 2021, the platform had surpassed 27 million MAUs.

For billing growth, a drastic increase also began in early 2020, with the platform now surpassing $21 million in billing revenue by the end of 2021.

Interestingly, the line chart reveals spikes in December of each year, highlighting a seasonality trend in billing and monetization.

I hypothesize that the December surge in billing revenue is influenced by an increase in disposable income during the holiday season. For young users under 21, this may be driven by the redemption of gift cards received as holiday presents. For adult users over 21, year-end bonuses likely contribute to higher in-game purchases during this period.

To better understand the relationship between MAU growth and billing growth, I stacked the two line charts to observe their cohesion.

From this visualization, I noted a widening gap between MAU growth and billing growth through 2021, suggesting that user count was accelerating faster than monetization. This disparity might indicate a strategic focus on user acquisition while monetization efforts lagged, potentially due to a lack of emphasis on creating valuable in-game items that incentivize user purchases.

However, by the end of 2021, billing experienced a vertical acceleration and closed the gap with MAU growth. This shift highlights a positive value creation for players, reflecting the platform’s success in enhancing monetization strategies.

Step 3: Understanding Retentions

Next, I want to better understand user retention. For a gaming company, it is crucial to examine user retention by segmenting users based on their onboarding date, also known as a Cohort Analysis. Cohort Analysis is critical to understanding player behavior and lifecycle. It helps identify trends in user retention, gauge the effectiveness of updates or promotions, and evaluate how different player segments interact with the game. This insight enables companies to optimize user acquisition strategies, improve in-game features, and boost monetization, ensuring a better overall gaming experience and sustained revenue growth.

I conducted two seperate Cohort Analyses to assess the MAU retention and MUP retention independently.

Below is a heatmap-style cohort analysis visually tracks MAU retention over time:

Axes:

• The horizontal axis represents time in months (0–59) since onboarding (Month 0 being the start).

• The vertical axis represents cohorts grouped by onboarding date, with each row corresponding to users who started in a specific month.

Colors:

• The heatmap uses a color gradient (green to red) to represent retention rates, with green indicating higher retention and red indicating lower retention.

If we look at the average retention rate across cohorts from Month 0 to Month 59 after onboarding, we quickly see that most cohorts experience a sharp decline in retention within the first few months. This is typical in many user-based platforms where only a small fraction of users remain engaged long-term.

After Month 12 of joining the platform, MAU retention rate stabilizes between the range of 1.0% to 5.0%. These are the loyal players who stayed with the platform.

Similarly, for MUP retention, I leveraged a heatmap-style cohort analysis to understand the retention rate of paying users after initial payment on the platform.

For paying users, retention exhibits a distinct pattern compared to overall users. While there is a slight decline in the first few months after their initial payment, paying users demonstrate higher stickiness and loyalty. After 12 months, their retention rate stabilizes between 35% and 45%, indicating that many paying users find the platform worthwhile and continue to invest in it.

However, an interesting drop in retention occurs around Month 49 after the initial payment. I hypothesize that this decline may stem from a game design issue.