Download PDF: The Game PM Field Manual.pdf

Google Docs: The Game PM Field Manual (Google Docs)

Preface

What happens when the world of business analysts, management consultants, and SaaS vets invade the game development floor, you get the modern video game product manager. Product management went from a relatively unknown role to a key discipline at major video game studios around the world. They are responsible for running and optimizing the dominant games-as-a-service business model employed by the highest grossing studios. Known as the masters of revenue to some, the people who have ruined the game industry to others, there is no denying this has become the defining role in the modern game industry.

Product managers matter more now than ever before to ferry video games through development to evergreen success. If you love video games and want to shape the winning strategy for the next big game, product management is the role you want to play in.

Introduction

Product management is the most misunderstood role in the modern video game industry. While development-focused roles such as animators or programmers have clear skill sets and responsibilities, product managers don’t directly craft any specific portion of the game. To increase the confusion, responsibilities and core skillsets can vastly differ at studios, publishers, and platforms within the game industry.

The Game PM Field Manual will define what a product manager is, the skillsets they must possess, how they operate during each stage of development, and where product managers can take their careers.

This field manual is intended for everyone from aspiring PMs who want to break into video game product management to principled C-suite executives that want to better understand how to differentiate good PMs from great PMs.

• Existing Game PMs Supplement existing skills, have access to a quick reference, or need a reference to share with interns and other non-PMs
• Students and Aspiring PMs Develop a base PM framework and basic playbooks of game PMs
• Tech PMs lateraling into games Learn how the more creative corner of product management operates and the differences between game PMs and non-game PMs
• Execs or Hiring Managers Understand how a great PM operates and how to evaluate PMs job performances

Think of The Game PM Field Manual as a practical, actionable playbook you can use to build your own product strategy, navigate cross-functional teams, navigate your PM career, and solve real-world problems across every stage of game development.

You can read it front to back if you want the full perspective on the lifecycle of a PM, or you can jump to specific chapters for specific guidance or reference.

Each section of the manual covers a different topic:

• What is a Game PM?
  The role of Product Manager, how it fits into studios, what are PMs responsible for
• Core PM Skills
  Mastering metrics, running experiments, modeling player behavior
• The Playbooks
  Actionable guides for each stage of game development product
• Owning Your Career
  Everything from landing your first PM role to future proofing your career

Who am I? My name is Eric McConnell, and with 15+ years in the video game industry as a product manager at companies like Amazon Games, Google, EA SPORTS, Zynga, and WB Games.

From an early age I knew I wanted to work in the video game industry, turning any school project I could into an excuse to create video games. I’ve spent over a decade building and managing games in many different roles on everything from games with 97 Metacritic scores to games that are so bad the entire development team was laid off days after launch. As a PM, I’ve turned 9-figure games around and I’ve also launched features that caused our most loyal players to lose every single match they played.

The Role of a Video Game Product Manager

Ultimately I like to define what a role is responsible for by defining what that role will be yelled at for. Game PMs are accountable for the business outcomes of video games, which almost always translates to gross revenue. Remember this: Video Game Product Managers are responsible for translating game design into business outcomes and business outcomes into game design. On Madden, I once had to Irish exit a Sunday date to fix a sudden drop in revenue and draft a communications email to all the executives at EA. A game PM’s career will live or die by revenue performance.

Let me paint an example:

You are the product manager of a mobile puzzle game about to start development on a big new feature. Not only do you need to take the proposed game design and forecast the business outcomes. You also need to build a model working backwards from the target business outcome to reflect what an ideal game design needs to perform number wise. Everything a product manager does needs to be data-backed to ensure you aren’t explaining why there is low ROI on the flagship game’s brand new feature.

To shoulder the weight of owning financial performance, the modern game product manager needs to have strong analytics, business acumen, and be able to lead cross-functional teams at game studios to marshal game teams to revenue riches. Product managers need to master both hard and soft skills, being able to explain complex data trends to developers that don’t speak analytics while also defining measurable game KPIs from the business goals; product managers help connect the executive level with the development teams.

Great product managers influence all aspects of their games without direct authority by quantifying parts of the game and measuring their performance. Who needs Jedi mind tricks when you are armed with data and market research?

Game PMs vs Traditional Tech PMs

Game PMs and more traditional tech PMs need to acquire customers, create features to meet customer needs, differentiate against competitors, develop competitive advantages, and loads of other things found in an endless array of business strategy books. Where video game product management separates itself from traditional tech product management is the translation layer of art between the business outcomes and the customer needs.

In traditional B2B SaaS product management, the product manager needs to define the product requirements based on customer needs. Then, they manage the product strategy to ensure it meets the business outcomes required from the product. In SaaS, customers ask for features and product managers gather the requirements for developers.

In video game product management, the product manager needs to measure and communicate business outcomes into game design. There is no simple metric for fun and you can not ask players what they want and build it. In games, players can’t tell if a game meets their needs before playing it.

Game PM vs Tech PM Example:

	Netflix	Fortnite
Engagement	Surface high likelihood to watch shows for users on both the home and up next surfaces	Continually create and optimize features, game modes, and content that is attractive to the player
Monetization	Retain subscribers; promote ad supported to standard and standard to premium	Create new appealing virtual items that drive spend depth from a crafted high perceived value
Retention	Continue to invest in shows that keep current subscribers from leaving	Craft short, medium, and long-term progression systems to give players layered goals to work towards
Cross-functional teams	Engineers, data scientists, consumer insights, product marketing, internal content teams	Game designers, engineers, producers, artist, animators, QA, and data scientists

Who Do We Impress? PMs in the Modern Game Studio Structure

The video game studio has changed a lot from the days where development teams consisted of a handful of engineers to now, where the modern development team has thousands of developers with as many job titles as characters in the Marvel vs Capcom 2 roster. In the modern game industry, product managers are responsible for the business performance of games and have to represent business outcomes while working in cross-functional teams with game designers, engineers, marketing, and producers.

Video game studios structure can be described in three tiers:

• External Stakeholders Executives, board members, shareholders, IP holders, and other entities or people that have influence over a studio or game
• Studio Leadership Head of roles, Director level roles, General Managers, Presidents, and Executive Producers
• Development Teams Junior, Mid, Senior, and Lead level studio employees

	External Stakeholders	Studio Leadership	Development Teams
Responsibility	Influence direction, approve timelines & budgets, set business goals	Define game & properties, make big decisions, approve work	Build the game and make day-to-day decisions
Product Roles	Chief Product Officer (CPO), Senior Vice President (SVP) of Product, Vice President (VP) of Product	Head of Product, Director of Product	Lead Product Manager, Senior Product Manager, Product Manager, Associate Product Manager
Game PM Interactions	Give weekly/monthly business presentations Be on call to answer metric related questions	Receive feedback, reviews, and promotions Implement frameworks, processes, and strategies set by leadership Give presentations on metrics and performance Identify opportunities for growth or optimization	Define goals for design and game features Translate game designs and decisions into forecasted outcomes Lead cross-functional teams on feature implementation
How to Win Over	Identify problems and solutions Present facts and metrics, not opinions Make information digestible for executives	Always be on top of everything Identify problems before leadership Provide data-backed decision making	Give data-backed feedback on designs Identify clear KPIs and goals for development teams to hit Give actionable competitive analysis

The Business Player Class: Video Game PM Responsibilities

The traditional disciplines within a video game studio are engineering, art, design and production. Engineers are responsible for the programming and tech ops; artists create 3D and 2D assets, designers craft meaningful experiences for gameplay and UX, and producers handle the scheduling and project management. So what are you going to be doing?

Product managers are the newest kid on the block, we are responsible for:

• Validating creativity by forecasting business outcomes
• Translating revenue targets to game requirements
• Identifying target audiences and where to play
• Designing monetization strategy
• Creating in-game store offerings
• Monitoring and measuring metrics
• Identifying KPIs and designing metric funnels
• Identifying, designing, and running experiments

Product managers are the voice of the player for balance, monetization, and content engagement in aggregate or through player segmentation. Through measuring metrics, running experiments, and generating player insights from data, product managers can identify and pursue opportunities to improve the game.

These responsibilities can further vary between AAA, mobile, and indie. Here are some generalized difference:

AAA	Mobile	Indie
Define monetization strategyIdentify KPIs and targetsCreate business reports to stakeholdersCompetitive analysis across genre	Manage game economyDesign and run experimentsMonitor metrics dashboardsForecast LTV of players from UA campaigns	Wear multiple hatsIdentify target audienceManage content and feature ROIBalance creative vision with business needsEnsure game is setup for market success

Core PM Skills

Product managers don’t directly develop components of the game but that doesn’t mean they don’t have core competencies that need to be masters. To lead games into the mythical lands of the evergreen cash cow, there is a skillset expert product managers need to master:

• Monitor Metrics and Identify KPIs
• Model Outcomes and Forecast Performance
• Run Experiments to Determine the Solution

Excelling at these skills and identifying when and how to deploy each is crucial to becoming a master product manager.

The Alphabet Soup of Metrics

You’re sitting in the studio all hands wondering why you didn’t spurge for that second cup of coffee when the lead product manager gets up to present your game’s performance. “Blah blah blah users, blah blah blah revenue, and our ARPDAU increased 15% this month.” ARPDAU? I’ve heard them say DAU before but now PM’s are just making up metrics.

Starting out in the alphabet soup of metrics can be intimidating. What are all these acronyms? Which metric should I care about the most? Is it good that a metric is up or down? Even seasoned product managers can struggle to understand how and when to use different classes of metrics effectively.

I’lll cover metric basics, common metric categories and their uses, and how to create bespoke metrics to your specific game. Using metrics to measure game performance, monitor player habits, and analyze issues in the game are probably the most synonymous skills with the video game product management role. Mastering the alphabet soup of metrics is the first step in your product manager journey.

What exactly is a metric?

At their core, metrics are a measurement and a time period. Daily Active Users (DAU) has active users as its measurement and daily as its time period.

Metrics give product managers a snapshot of performance of a game. With that snapshot, you can compare performance against other points in time, other games, and against your game’s goals.

Metrics give us standardized language to talk about business performance. We can’t just say “Our game has more active users” or “Wow, active users went up by 253”. Product managers need a defined language to formulate how active users have increased and over what time period.

How do we use these metrics?

The best way to think about how to use metrics is to answer questions. How does this week compare to last week? Why is revenue down? Where are players stuck in the game?

There are different terms for investigating different types of questions. Determining the underlying cause of an issue or sudden change in performance is referred to as Root Cause Analysis. This is looking backwards to explain changes in performance. On the other hand, Opportunity Discovery is referred to when performing forward looking analysis to identify upside and positive outcomes to pursue.

To aid in accelerating answering questions, we use structures such as Metric Funnels. This can turn finding a needle in a haystack into shooting fish in a barrel. We’ll cover building and using metric funnels later during the playbook section.

Metric Categories

Metrics can be divided into a number of categories based on what they measure and the questions they are answering. These categories cover the most common measurements across all games and will be useful whether you are running a traditional fighting game, battle royale shooter, or match-3 adventure.

Active Users

Measures Unique Users

Active User metrics measure the number of unique users that played the game during a time period. How many players played your game during a span of time? User can mean all users or it can mean a specific user like someone who has made a purchase in the last week. Active can mean boot up the game or it can mean performing a specific activation event like log-in to an online account.

This is the most common metric to be measured and is useful to measure for any game regardless of genre or monetization strategy. Every product manager should be tracking a number of different Active User metrics related to their game: total active users, users by game mode, users by IAP status, users by level, etc. to get a wide view of the different distributions of players across their game’s population.

Examples

Daily Active Users (DAU), Weekly Active Users (WAU), and Monthly Active Users (MAU) are the most common and widely used Active User metrics. They answer the question: how many unique users played your game during a day, week, or month. Every game would benefit from tracking these 3 metrics. Common ways to measure are calendar aligned, meaning they line up with the days, weeks, and months of a calendar, or rolling-basis meaning last 24-hours, last 7-days, and last 30-days.

Spender and Non-Spender are useful metrics for games with In-App-Purchases (IAPs). A spender is someone that has bought in-game purchases with fiat currency and non-spender is someone that has not. Common ways to measure are life-to-date (LTD), meaning has the user ever made a purchase in the game, or by daily/weekly/monthly, meaning did the user make a purchase today, this week, or this month.

New and Returning is useful to distinguish whether a user played the game for the first time during this time period or are they a returning player who has played before.

Retained, Reactivated, and Lapsed are further distinctions of returning users. Retained is a user that played during the previous time period and during this time period. Reactivated is a user that played before the previous time period, did not play the previous time period, but played this time period. Lapsed is a user that played during the previous time period, or before, but did not play during this time period.

	Before Last Week	Last Week	This Week
New	Not Played	Not Played	Played
Returning	Not Played	Played	Played
Retained	Played	Played	Played
Reactivated	Played	Not Played	Played
Lapsed	Played	Not Played	Not Played

Engagement

Measures Interactions

Engagement metrics measure how much a user interacts with your game. How much do people play your game when they are an active user? These interactions can be measured in time periods, such as hours, days, or weeks, as well as units of gameplay, such as matches, missions, or seasons.

Engagement metrics are great for measuring how your users spend time in your game. As a product manager, you should have a thesis on what each user segment’s engagement patterns should be. Measuring engagement for each type of content will unveil how valuable users perceive that content.

Examples

Session is the most common time-based engagement metrics. A session is a continuous period of play, marked by a clear start and end such as login and logout. A session is a great engagement metric because it ties game interactions to time measurements. Sessions per day or sessions per week measures how often your game is played in a single day or week. Hours per session measures the length of play during each time the game is booted up.

Gameplay containers are the most common non-time-based engagement metrics. A gameplay container is a consistent gameplay event within a game. For example, an American Football game would have downs, games, and seasons as gameplay containers while a Match-3 game would have swaps, levels, and worlds. These measure gameplay in terms of the content and game systems that define it and are great for translating engagement into game-specific measurements. You can tie gameplay containers to sessions with metrics such as levels per session or games per session from the examples above.

Retention

Measure return rate

Retention metrics measure the percent of users that were active in one time period that are also active in another time period. What percentage of players played last week also played this week? Retention can be measured back-to-back or measured following a cohort of users over time such as what percent of users that bought the game in week 0 played in week 1, week 2, week 3, and so on.

Retention is the single most important metric in live-ops product management. A highly retentive game can snowball active users while optimizing monetization features. On the other hand, once retention drops and you are unable to retain or re-engage users, your game may enter a death spiral that few games can recover from.

Retention can be used as a proxy for fun. If users are consistently returning to your game, they are choosing to play your game over the multitudes of other entertainment options available to them. I once worked for a game company that regularly said “there is no metric for fun”, so until we find a universal way to measure fun, I use retention as a proxy for how fun my users find my game.

Day-over-Day (DoD), Week-over-Week (WoW) and Month-over-Month (MoM) are the most straightforward ways to measure retention. What percent of users that played your game in the previous week played your game this week? These retention metrics are useful for getting a baseline retention measurement for things like a marketing campaign or new seasonal content.

D1/D7/D14/D30, W1/W2/W4/W8, and M1/M3/M6/M12 are the most common retention metrics in the live service games. The D, W, and M stands for day, week, and month respectively. The number represents the day from the install date with 0 representing the date the user started your game. What is the average percentage of users that played on D1 or the day after the installation date, D7 or seven days after the installation date, and so on. These give you great checkpoints across different time periods.

Stickiness

Measure rate of engagement

Stickiness metrics measure the rate of engagement of users. Stickiness requires two time periods to measure the rate of engagement of the shorter time period within the larger one. How many days did users play this week? Stickiness metrics are great for measuring how engaging your gameplay is. If you have a game where players are expected to play at least once a week, Stickiness metrics can measure the rate they engage during each week.

Days per Week and Days per Month are the foremost stickiness metric for measuring daily rate of engagement. This measures for every user that played the game in the last week or month, on average how many days did they play. Days per Week will be a number between 1 and 7 while Days per Month will be between 1 and 30.

DAU/WAU and DAU/MAU are what I consider the poor-man’s stickiness metrics. By dividing the average DAU within a 7-day period by the WAU for that same 7-day period, you get the average percent of the WAU that engages each day. The same with DAU and MAU measuring the percentage of MAU that engages each day.

Revenue

Measures generated income

Revenue metrics measure gross revenue. How much money did your game make last week? Although I just said Retention metrics were the most important for live-ops success, revenue metrics will be the most important for your career. Product managers are judged by their ability to impact revenue. Understanding important numbers like how much revenue, on average, a new user generates over their lifetime is integral to making business decisions related to your game.

A word of warning on maximizing Revenue metrics such as LTV. Although it seems very obvious for a product manager to focus on driving revenue as high as possible, there is a graveyard of once healthy and successful games that were killed by product managers focusing development on driving up transactions and spend depth higher than the game or its audience could sustain. More on balancing growth with sustainability later in the playbooks section.

Gross Revenue measures the revenue generated within a time period. Gross Revenue can be segmented by user types, purchase type, or any other property. Ultimately, a game’s success is determined by topline gross revenue.

Lifetime Value (LTV) measures the lifetime value of a user. LTV is typically a forecasted metric that is calculated from a number of predictive properties: game progression over the first one or two weeks, stickiness, acquisition source, etc. LTV is used to determine the amount of marketing spend to acquire new users.

Conversion and Spender/DAU measure the percent of users that are spenders. Conversion measures the number of active users that spent in that time period divided by the unique active users during that same time period; effectively how many of your users spent during a time period. Spender/DAU is conversion for a single day, measuring the percent of active users who have spent at least once for that day.

ARPU and ARPPU both measure revenue per user. ARPU is average revenue per user and ARPPU is average revenue per paying user. ARPU is useful for measuring all encompassing monetization health per user while ARPPU is used to measure depth of monetization for converted users.

Transactions measure the number of purchase transactions made during a time period. Transactions bridges revenue to spenders. It can be used in conjunction with gross revenue to measure Gross Revenue/Transaction, tracking the average size of a purchase. Transactions/Spender can tell you the average number of transactions each spender made in a time period. Both are useful to see the full story of revenue as you’ll see later when we cover metric funnels.

Game Specific

Metrics unique to your game

The final category is any game specific metric that is not easily captured by generalized measurements. Does your game have subscription tiers to track membership to? Is your game supported by ad revenue? Do you need to track matchup data between characters in a fighting game to determine overall roster balance? Are there multiple currencies that you need to track the absolute value exchange rate between?

As a product manager you should think about what makes or breaks your game’s revenue, retention, engagement and measure that relentlessly. You should know the thresholds for when your game is healthy and when you need to sound the alarm.

Supercharging Root Cause Analysis with Metric Funnels

There is a single tool that can take you from “Red alert! Revenue is down 20%” to immediate action on the specific underlying issue that is tanking your KPI, the all mighty metric funnel. Metric funnels are models that break a single KPI into increasingly more specific indicators so that PMs can quickly hone in on causes during root cause analysis. Metric funnels break a higher up metric down into two or more metrics that are either added or multiplied together to equal the higher up metric. When complete, a metric funnel looks like a pyramid with the KPI in question at the top.

Installing metric funnels is the first thing I do when I am brought onto a live game. Their value is indispensable to being able to fully visualize and represent the different aspects of a game in metric form. Proper metric funnels help understand shifts and patterns in player sentiment across any metric type.

Building Metric Funnels

Building metric funnels is an art that takes time to master. The general process:

1. Identify the KPI for the funnel
2. Decide on key indicators you want to capture in the funnel
3. Decide on key questions you want to answer with the funnel
4. Start with the KPI at the top
5. Break the metric(s) from that layer into two or more metrics that, when added or multiplied together, create the KPI metric
6. Repeat Step 5 until the metric funnel is finished

There is no rule for how deep a metric funnel needs to be. Generally, you will have a good feeling when you’ve exhausted the different branches, captured the indicators you wanted to include, and are able to answer the questions you need from the metric funnel. Metric funnels should have equal length branches for every path, hence the pyramid imagery.

A common pitfall is using meaningless metrics or repeating metrics to satisfy the add or multiple rule for subsequent layers. This means you have either exhausted a funnel branch or you need to redesign the layer above to capture broader indicators.

Here is a generic Revenue metric funnel. Within the single funnel I captured:

• Gross Revenue
• ARPU
• Average Revenue per Transaction
• Average Revenue per Item
• Average Items per Transaction
• Average Transactions per User
• Average Transactions per Spender
• % of Users that are Spenders
• Active Users count
• Every count of Spenders and Non-Spenders

You can see one side utilizes multiplication between layers, because of its use of ratios, while the other side utilizes additions, because of its use of absolute counts.

Using Metric Funnels

Using metric funnels is incredibly simple. You will measure and label the delta between time periods on each metric in the funnel. So if you are measuring performance between weeks, it’ll be the percentage delta between week 1 and week 2. If there is a big delta for the KPI, positive or negative, you can follow that trend down the funnel like an artery leading you to the source.

Bonus Tip: Color code your delta by weight, making the root cause analysis visually obvious.

Here you can see that Revenue is down 10%, but instantly identify that Recurring Spenders is the culprit. Now you break the emergency glass on playbooks to reactivate Spenders or convert Non-Spenders. Also notice that all these metrics are symbiotic, so one decreasing might cause another to increase.

Developing Clairvoyance with Modeling & Forecasting

What if I told you there was a product management skill that the more you improved at, the closer you could get to predicting the future? This crystal ball tool is known in the product manager world as modeling.

Modeling is representing complex systems within spreadsheets for the purpose of forecasting outcomes. How much revenue are we going to make? How would a game need to perform to hit our KPI targets? Modeling allows product managers to hold a mirror to the game design and show how it is predicted to perform, or open an alternative dimension to show what a game would need to look like to hit specific goals.

What is Modeling?

At its most basic, modeling is using spreadsheets to build numerical representations of a game. You can use lots of paradigms and tools outside of spreadsheets, but for this guide we will focus on basic spreadsheet functionality found in Excel or Google Sheets.

When you model a game or part of a game, you are representing everything from the economy to content burn with metrics and numbers. Modeling connects the world of game design to metrics, allowing product managers to give guidance, guardrails, and feedback on game design in a role appropriate way.

Setting up a Model

Choosing the proper setup is half the battle to building accurate models; product managers need to determine whether to deploy a bottoms up or top down model, select the right variables from the data available, and decide on the fidelity or depth of the model to build.

Step 1: Identify the Question to Model

The first step is figuring out what you want to forecast or identify the outcome you want to achieve. It could be a question as broad as “How much will we make?” to something as narrow as “How many copies of a game do we need to sell for matchmaking to have adequate CCU?”.

This will determine what type of model you build, what metrics you need to gather, what variables you need to solve for, and what level of detail you need to answer it.

Step 2: Choosing the Correct Model Type

There are two main model types we are concerned with: bottom-up and top-down. Choosing which one to deploy will depend on what question you are trying to solve.

Bottom-up models are used to predict the performance of a game design.

• How would this new game mode perform?
• If we changed this system, how would it impact our economy?
• How long until players reach end-of-content with our current content cadence?

If you need to answer a question or predict the outcome about a specific design, you want to build a bottom-up model.

Top-down models are used to identify what must be true for an outcome to occur.

• What amount of users do I need to generate our target revenue?
• What amount does each game mode need to earn to reach our revenue goals?
• How much content do I need to make sure players don’t hit end-of-content?

If you are starting with an outcome and need to determine what must be true to reach that outcome, you want to build a top-down model.

Step 3: Selecting the Input Variables

Models are composed of two types of data: assumptions and input variables. Assumptions will come from game data and fill out most of the model. Input variables will serve as the values you change to test different scenarios.

Determining your input variables will depend on what you are trying to predict. What metrics directly impact the key outcome? What is unknown or able to change versus what isn’t changing or assumed to be true? The input variables are typically player-based in bottom-up models and game-based in top-down models.

Question: How much revenue are we going to generate in Q3?

Input Variables might be:

• New Users
• Active User churn
• Conversion Rate
• ARPPU

Step 4: Determine the Fidelity

Models can be as complex or simple as the situation requires. Sometimes a model can span multiple sheets and pivot tables, other times models can fit on the back of a napkin. Determining the model fidelity is a balance between level of effort and accuracy needed.

Low fidelity models can be thought of as fast math for quick decisions. These models are for directional guidance like doing quick opportunity sizing where the model only needs a few layers deep of calculations to arrive at a prediction. If speed is important and the outcomes will likely be obvious enough, a low fidelity model will suffice.

High fidelity models can be thought of as simulations that require massive complex data sets intertwining together to develop a prediction that is as accurate as possible. These models are for mission critical projections whose accuracy has company livelihood implications on the line. If accuracy is mission critical and ample data is available, you need to setup a high fidelity model.

Most models live somewhere between the low and high fidelity. Product managers will find themselves making predictions requiring more than quick math but not building a matrix-level simulation. Determining the balance between level-of-effort and accuracy is what separates the expert modelers from the MBA flunkies.

Step 5: Build the Model

Time to put it all together. If you want to become a PM, learn to embrace spreadsheets.

Model Example

You are working on a FPS that has 2 game modes. The team is thinking about launching a 3rd game mode but is afraid of cannibalizing one or both of the existing game modes. Model how a new gameplay mode will impact engagement of other game modes.

The first step is to identify the question:

How much will engagement change if this new game mode is introduced?

Next identify the type of model you need to build:

Are you trying to predict performance or understand what must be true for an outcome to occur? You want to predict how engagement will change when introducing a new game design for people to interact with. Predicting the performance of a game design requires a bottom-up model.

Then select the input variables:

The range of scenarios you want to explore is how appealing and consuming this new game mode is going to be, so you can then judge the potential cannibalization of the two other game modes. The player segment’s level of engagement with the new game is what you want to set as variables. You can also set the new game mode’s properties as variables, but for this example we are assuming the game mode is fully designed.

After that, select the depth of the model:

For this model, you want to create engagement profiles for different player segments, engagement profiles for game modes, and model active users for game modes based on the input variables. This is a fairly shallow model where the accuracy requirement is more about market sizing the interest in a new game mode than simulating the entire game’s engagement accurately. We’ll ignore retention rate, user churn, user acquisition, etc. and treat active user counts and player segment sizes as static.

Finally, build the model:

Modeling the player segments, each segment needs:

• Hours per Session
• Session per Day
• Days per Week
• Game Mode 1 Interest
• Game Mode 2 Interest
• Game Mode 3 Interest

* Where Game Mode 1, 2, and 3 Interest sums together to equal 1

Game Mode Interest is your Input Variable.

Model the player segment split:

• Segment 1 %
• Segment 2 %
• …
• Segment N %

* Where each segment added together equals 100%.

Model the game modes, each mode needs:

• Hours per Game
• Players per Game

The last step is to calculate:

1. Game modes played per session for each player segment then game modes played per day
2. Divide games by the players required per game mode game per day
3. Get the number of games per day played per game mode

Compare it to the existing data on game modes 1 and 2 to determine the impact of cannibalization.

The Dark Arts of Experimentation

Experimentation, also known as A/B testing, also known as split testing, also known as the evil that product managers do, is serving different players different experiences and measuring the performance difference between the two experiences. This is the most powerful tool in a product manager’s arsenal for optimizing games, giving PMs the ability to test hypotheses across similar player segments. The ability to set up and run impactful experiments is what separates the greats from the good in the game PM world.

Running successful experiments starts with identifying opportunities, then developing a hypothesis that pursues the opportunity, choosing success metrics, and identifying the player population to run the experiment on.

On the game side, experiments should be obfuscated so players can’t conclude they are being purposely served different experiences. Experiments can be applied to almost every aspect of a video game but S-tier product managers know where to focus their A/B testing for the largest impact.

If there is one PM technique that can draw the ire of forum dwellers and cause YouTubers to stage a boycott it is experimentation in live-service games. This is the most powerful tool for both making business performance gains as well as sinking your games reputation and your studio’s brand. Tread carefully and don’t wade into massive A/B testing without mastering the nuances of proper experimentation first.

Setting up an Experiment

The prerequisite to running any experiment is having an experiment manager service. This will either be an in-house or a 3rd party tool, and needs to be correctly integrated into the game, metrics database, and player accounts, while the gameplay content being experimented on needs to be data driven by the dynamic data served via the experiments manager segmentation.

Experimentation is an artform that requires deep knowledge of not only your game, but your players and competitor titles. There isn’t a greedy algorithm for decision making while setting up an experiment. With the ability to negatively impact the player population, it is better to air on the side of caution while exploring optimizations.

Setting up an experiment has 6 steps:

1. Identifying the Opportunity
2. Developing a Hypothesis
3. Choosing a Target KPI
4. Selecting a Target Population
5. Creating Segments & Control KPIs
6. Set Duration and Safeguard KPIs

Identifying the Opportunity

The first step is to monitor metrics and identify opportunities for optimization in the game. What do you want to improve about the game? This can be anything from declining performance, deficits compared to competitive benchmarks, or specific KPIs that can be driven further. This is again why metrics are the backbone of every other PM function.

Example You identify your game’s ARPU is much lower than competitor titles based on competitive analysis.

Developing a Hypothesis

Now that you know what aspect of the game you want to optimize, you need to develop a hypothesis on how this opportunity can be pursued. What do you think will improve this opportunity? Know what habits, game systems, reward systems, or gameplay content can be altered to address the opportunity.

Example You hypothesize that ARPU can be increased by targeting conversion rate for first time buyers with a new buyer offer.

Choosing a Target KPI

A target KPI is a metric that will determine which segment is the winner of the experiment. What is the one measurement that will determine the winning game experience from the different experiences you want to test to address the hypothesis? This is commonly the same metric identified during the opportunity section but can also be a downstream indicator metric for the hypothesis.

Example Although ARPU was the opportunity identified, First Time Conversions (FTC) will be the target KPI. Increasing FTC increases ARPU, and the experiment is designed around conversion rate for first time buyers.

Selecting a Target Population

Experiments need a player population to be the subjects of the test. Who is going to be served a variant experience to measure if it improves the target KPI? The population can be as wide as the entire game playerbase or as narrow as players that started the game on a specific calendar day.

Example For this FTC experiment, your target population is players who have joined after the start of the experiment and have yet to make an In-App-Purchase (IAP).

Creating Segments & Control KPIs

Now you have the hypothesis, target KPI, and target population, you need to determine segments and their test experiences. What are you changing in the game for the target population to improve the target KPI? You need to determine the number of tests (segments) you want to run, the segment & control weights, and the control KPI.

The first step is determining the number of tests you want to run. This will be influenced by the hypothesis and target population size. Is the hypothesis able to support multiple tests? Do you have enough players in the target population to support 2, 3, or 4+ segments? What is the minimal number of players for a segment to be valid? This will vary from studio to studio and experiment to experiment. These norms should be determined before a game is ready for A/B testing.

After you have the number of tests, now you need to set the segment and control weights. What percent of the target player population will be served each test? The control is a segment that receives no tests and their experience remains unchanged from the baseline game. A general rule of thumb is to evenly distribute the test segments across 90% of the target population and set the control to 10%. So if you have 2 segments, segment 1 receives 45% of the population, segment 2 receives 45% of the population, and the control receives 10% of the population.

Finally, determine the control KPI(s). What properties do you want to remain even across all the segments? A control KPI is a metric to evenly distribute the target population across. This is to ensure a group of outlier players don’t skew the results of the experiment. For this example, set the control KPI to the predictive LTV of players.

Example You are going to test the value proposition of two different FTC bundles.

Test Segments

Segment 1: Sell a FTC bundle that provides 25% extra value

Segment 2: Sell a FTC bundle that provides 50% extra value

Control: Continue the existing experience

Weights

Segment 1: 45%

Segment 2: 45%

Control: 10%

Control KPI

LTV

Set Duration and Safeguard KPIs

The last step is to set the start and end time of the experiment as well as the safeguard KPI(s), or what measurements should cause the automatic shut down of an experiment. How long should the experiment run? When should the experiment immediately shut down? The duration of the experiment will be heavily dependent on the experiment and the game. As a rule of thumb, general experiments should be run a minimum of 2 weeks and a maximum of 8 weeks.

Safeguard KPIs are the combination of a metric and a threshold that if surpassed, shut down the experiment immediately. This is a protection mechanism against unpredicted behavior, unseen exploits, or other errant behaviors that adversely impact the intended experience for players.

Example The duration of the experiment is 4-weeks. The safeguard KPI(s) are New User Conversions, or first time purchases, drop below 5% for players within segment 1 or 2.

Running the Experiment

Hidden in Plain Sight: Obfuscation and Player Contract

One of the most underlooked aspects of running experiments is proper obfuscation. Players should never know they are being served an experiment instead of the standard game experience. One of the biggest reasons for players identifying they are in an experimentation is their ability to compare direct experiences. If players have disparate experiences, there needs to be a logical game reason for it such as the differences are a result of randomization (i.e. loot tables reward different items to players for killing the same monster).

Some common obfuscation techniques are:

• Hiding behind randomization
• Implying results were caused by other player actions
• Giving players the illusion of choice

However, obfuscation should never break the player contract. Player contracts are the norms and expectations that come along with a game’s perceived genre, platform, thematic wrapper, and previously established game systems. For example, some platforms and genres have the expectation of active experimentation (F2P mobile) while others would treat even the suspicion of experimentation as a complete betrayal of trust (MMORPG).

Knowing what experiments are appropriate for your game, how to properly obfuscate them, and how to identify the player contract and what not to break are vital to running successful experiments.

Analyzing the Results: MDE and Statistical Significance

It’s time to see if all your meticulous setup of the experiment has paid off. The final step is analyzing the results of the experiment. The target KPI(s) are your north star for determining a winning segment, but just because a segment performed better than the rest doesn’t mean you have a successful experiment.

The difference in performance between the winning segment and the control segment needs to surpass the MDE or Minimal Detectable Effect. MDE is a predetermined delta or percentage points threshold that a segment needs to perform above the control to be declared a successful experiment. If the winning segment is 3% higher than the control for the target KPI, and the MDE is 8%, the experiment can provide directional guidance but is not considered successful in addressing the hypothesis.

Another gate to pass for an experiment to be declared successful is statistical significance. Statistical significance calculates how likely a result is due to a true effect or simply random chance. Having a probability value, more commonly known as p-value, less than the alpha, a decimal such as 0.05, is considered a statistically significant experiment. There are also other calculations that can be used to determine the validity of the experiment, so please explore what the statistical requirements are for your studio.

What is and isn’t a good candidate for experiments

Good experiment candidates:

• Ask quantifiable and focused questions
  • Which price and quantity pairing drives the most purchases?
  • How short can the tutorial be?
• Have a measurable target KPI
• Is specific in its hypothesis and the measurements to determine the successful segment

Bad experiment candidates:

• Ask qualifiable or objectionable questions
  • “Does this art look nice?”
  • “Is this gameplay fun?”
• Require numerous target KPIs to determine the segment winner
• Have a small target population leading to underpowered samples
• Are susceptible to noise leading to false positives

Experiment Examples

Drop Rates
Optimizing the rarity and loot tables

Tutorials and FTUE
Optimizing the length and content of the first time user experience

Difficulty
Optimizing enemy properties and AI or gameplay obstacles and constraints

Store Offerings
Optimizing bundles, price/quantity pairings, store inventory, and item sales/discounts

Promotions
Optimizing giveaways and features such as daily login bonuses

The Playbooks

Now you know what product managers are, where they fit into the modern game industry, and you are equipped with the groundwork product manager skills, it’s time to put it all to work and make the money printer go brrrrrr.

Product managers are responsible for the business outcomes of a video game. No matter where a game is in its lifecycle, your job is to lead the game to measurable success. I am going to give you general playbooks to run and operate at each stage of development. After this section you’ll have a map to guide your adventures in game development.

For each development stage I will provide:

• A general timeline of that stage
• Product Manager goals during the stage
• Operations or functions you as a PM should be performing

These playbooks are recipes that you can alter and modify for your specific studio and game. Each stage is meant to provide a solid foundation for the next, building upon what was done in the previous stage.

Pre-Production

Pre-production is the first and most strategic stage of video game development. Each decision can have a reverberating impact later in the game development lifecycle.

Pre-production is when the game’s game design, art direction, tech stack, and resource requirements are being explored and defined. During pre-production the development team will produce everything from conceptual pitches to late stage vertical slices.

Common pre-production milestones:

1. IDEATION Generate, evaluate, and down-select ideas
2. VALIDATION Prove commercial and player viability
3. PROTOTYPE Prove the game is fun and achievable
4. VERTICAL SLICE Show the game’s vision fully realized
5. PRODUCTION READINESS Ensure the game is fully defined and scoped

The goal of a product manager during pre-production is to set up the game for success by validating its market competitiveness. For smaller scale indie games this can be understanding the potential audience for your art game. On high budget AAA games this can be gathering data to inform every design decision to maximize chances of success.

To get the product from conception to production readiness, you’ll want to perform these 5 steps:

Step 1: Defining Business Goals

Step 2: Player Analysis

Step 3: Market Sizing

Step 4: Competitive Analysis

Step 5: Business Model Exploration

Defining Business Goals

Set the stage for ideation by defining the business goals to the game team.

This first step to developing a successful product is to define why you are making the product in the first place. Why is someone paying us to develop a game and what do they want from the release of this game? This could be anything from winning an award from GDC/IGF/DICE/some other comforting acronym, to trying to fill the empty void in yourself with the equaling empty pursuit of self-therapy game development.

But for 95% of paid game development this will be gross revenue (the other 5% being serious games or games-as-marketing). This can be determined by a publishing contract, the needs of the studio, or cost analysis of what it takes to make the desired game. Other business goals might be metacritic ratings, platform charts, unique user counts, or any measurable outcome that is important to success.

These business goals will help identify constraints for ideation such as:

• Does the game need to support live operations or is it a traditional premium title?
• Does the game have a specific genre or audience to target?
• Are there technologies or features that need to be utilized in the game?
• Does the game need to be delivered by a specific date and on a specific platform?
• Is there a specific IP or theme that will wrap the game design?

Player Analysis

After ideation, work with marketing to define the target personas for the game.

The next step is to determine who will play your game. Who do we need to wow and excite with our game? This is done by identifying target player personas and their unique needs. As much as your business goals will scope ideation, player personas will help direct prototyping and vertical slice development.

Your goal is to create 3-4 target player personas and 2-5 secondary player personas. The target player personas are used to direct and validate design choices from art direction to moment-to-moment gameplay. The secondary player personas will help guide growth or expansionary decision making.

A player personas contains:

• Name, Age, Gender, Occupation, Location Information to paint a picture of this player
• Player Needs Player motivations, frustrations, and desired fantasies
• Currently Playing Games, platforms, and current
• Winning Hypothesis How do you pull this player away from their current games to a new one?

These personas will be used to drive decision making throughout the entire lifecycle of game development.

Market Sizing

Work with marketing to measure and validate the size of the target audience.

Now is the time to see if your business goals match the opportunity of the current market. Are there enough potential players for you to achieve your goals? Ultimately your goal is to either justify the budget and production scale or flag it as a known risk going into development.

There are a few ways to perform market sizing but for most games, the traditional top-down method of TAM, SAM, and SOM is the most straightforward and effective way. TAM, SAM, and SOM are three market sizing metrics where each gets more specific and narrow on the audience.

• TAM or Total Addressable Market The total market that could possibly buy your game
• SAM or Serviceable Available Market The target market for your game based on category or genre, thematic wrapper, and art direction
• SOM or Serviceable Obtainable Market The realistic achievable goal for the game to achieve based on production value and marketing reach

Let’s look at an example of a casual multiplayer FPS that will be released on Steam:

TAM	Total Addressable Market	Steam’s MAU
SAM	Serviceable Available Market	Casual MP FPS MAU
SOM	Serviceable Obtainable Market	Expected MAU of your game for launch

Using available information, you will want to start with TAM, which will likely be the total users across all your target platforms. SAM will be based on your competitive titles and market category. SOM will be your realistic business objective based on your own forecasting for your game’s appeal, production value, and marketing budget.

SAM and SOM will determine if the market shows the opportunity for the game justifies its budget and production scale. If these numbers don’t align it doesn’t mean your game is a bad investment. There are breakout hits, market making games, or games that don’t easily align to existing market definitions. The studio and publishing team need to work together to understand the risks and develop more sophisticated indicators that the game can achieve its needed market reach.

Competitive Analysis

Work with game design to break down the competition’s mechanics and systems.

For video games that compete against highly retentive games for players, competitive analysis is arguably the most important pre-production step. Competitive analysis is understanding a competitive or comparable game’s success, shortcomings, and weaknesses to make strategic decisions with your own game across gameplay, monetization, and aesthetic design.

The process for competitive analysis is pretty straight forward:

1. Identify your Competitors
2. Gather and Analyze Data
3. Determine your Competitive Advantage

Identify your Competitors

Your competitors are games that potential players will either choose between playing or be considered similar by marketplace categorization. This will be a combination of genre, difficulty, online/offline, single player/multiplayer, thematic wrapper, and production value. How will marketplaces and players ultimately group your game with others? For example, solo-dev indie games may be grouped together regardless of mechanics or narrative.

Gather and Analyze Data

You have your competitors, next you need to gather and analyze data about the game and its development. The data you want to gather is in two parts: game data and business data.

First, I highly recommend playing all of your competitive games to the point where you thoroughly understand all the mechanics and game systems. If it’s a competitive game, make sure you can hold your own against the design and QA departments (departments where studio GOATs are found), you are repping PMs everywhere. This is that mythical time you actually NEED to play a bunch of video games for work! Your 12 year old self would be so proud.

Game data means list out the game systems, mechanics, monetization, social features, and content types of the game. You want to understand everything about the game from a player’s point of view in terms of what makes players play and retain within the game, why do players spend money, how do they acquire new players, etc.

Business data means development team size, development time, budget, content cadence, live operations plan, and sales over time. These are the numbers behind the scene that measure the initial investment, growth post-launch, and current state of the game’s financials. This will help you further understand the viability of your budget, team size, and amount of content you’ll need to launch your game.

Determine your Competitive Advantage

Finally take the information from the previous analysis and determine what competitive advantages you have. Use your player personas as targets, apply the same analysis to your game, and determine whether your personas will or will not play your game given the competitive landscape for their attention and money. This will help you determine where your game’s strengths and weaknesses are.

Business Model Exploration

As part of production readiness, perform the initial pass on defining the revenue streams.

Before you leave pre-production you need to have a plan on what the revenue generating channels are for your video game. Premium price point, future content expansions, in-app-purchases, there are a number of decisions you need to make before entering productions. Time to go from “we need to make $200M” to “here’s how we make $200M”.

Initial Launch	F2PPremiumIncluded with SubscriptionPlatform Exclusive
Content Updates	No UpdatesRegular CadenceMonthlySeasonalYearly
Paid Content	DLCSubscriptionIAPsLoot Boxes/GachaPremium Currency

These decisions will dictate the content runway you need at launch, content cadence post-launch, and expected cost of content for development. As the product manager you need to understand your monetization models and how to maximize their effectiveness. Are you playing a volume play or driving spend depth on a set of diehard niche consumers? Is the initial launch going to be the make or break financial test, or the beginning of a decade long marathon of live-ops revenue?

Don’t worry about nailing all the details right away, you will be making plenty of models and forecasts later in production. The goal right now is to make the big decisions that will be difficult to reverse mid-development.

Production

Production is the second stage of video game development and where the game is made. This is when developers roll their sleeves up and build entire worlds with rules, systems, and player interactions. Let’s just say the terms crunch or death march or development hell don’t get invented because making games is fun.

When game development enters the production phase, it’s the product manager’s job to ensure the game launches with the maximum chance for business success. While producers lead development with their various schedules and whatever a gantt chart is, product managers measure every aspect of development to verify its validity and give developers quantitative feedback on the game.

Here is where you will become best friends with the ever important spreadsheet by modeling:

• Content Runrates How long each player segment will take to reach “end of content”
• Balance & Difficulty How hard will it be for each player segment to complete content
• Monetization Strategy How much each player segment will generate for the game?

Player Segments

Before you get knee deep in INDEX(MATCH()), you’ll need to figure out who you are measuring the game’s content, difficulty, and monetization against. You need to create models of your target player segments.

The easiest way to start is with your player personas. For most games, this is all you need. For other games, a single player personas could manifest as a gradient of different engagement patterns. Think of player segments as “if I was going to look at player patterns in my game, how would I group different players together.”

For each player segment, you’ll need to define:

• Acquisition Patterns
  • New Users at Launch
  • New Users per Day or Week
• Engagement Patterns
  • Session Length (in minutes)
  • Sessions per Day
  • Days per Week
  • Weeks per Month/Year
• Retention Patterns
  • Daily or Weekly retention curves for churn
• Purchase Patterns
  • Conversion Rate
  • Revenue per Purchase

Content Runrate

Bill Gates once said “Content is King”, I can’t be 100% sure but he was probably talking about having sufficient content for your player segments at launch. Whether your game is a F2P title with a robust live-ops plan or a premium title that has zero future content updates planned, you should have a thesis of how much content is needed at launch. You will need to create a bottoms up model to predict when different player segments will reach the end of available content after launch.

Content Runrates describe the speed at which different player segments consume and reach the end of content. End of content is the state at which players run out of novel content and are at risk of churning or leaving the game.

Content runrates are useful for judging the different types of players that will play your game. On the extremes you have

• A player who slowly plods their way through your game
  You want to ensure they feel accomplishment and satisfaction in their choice to play your game despite putting limited time into playing it
• A hardcore super fan that will burn through content faster than logically makes sense
  You want them to not feel jipped or short changed by finishing the available content in too short an amount of time

Finally you have the Goldilocks players, who represent your average core gamer playing an expected average amount of time.

Next, model out your content in escalating gameplay containers. Let’s pretend you had a linear 2D platformer with levels that serve as the main gameplay, worlds that serve as containers for collections of levels, and navigation paths where players traverse from level to level.

Level Avg. Length (mins)

Level Avg. Success Rate

Avg. Traverse Time Between Levels (mins)

Levels per World

Notice the Level Avg. Success Rate metric. This represents failing a level and having to repeat it before moving along.

With models of player segments and gameplay containers, you can create a runrate content per session and work your way up.

• When does each player segment complete the game?
• How far does each segment get the first day, week, and month?

These can help inform everything from the amount of content needed at launch to the live-ops content cadence needed to keep the most engaged players from reaching the end of content.

The final piece to the puzzle is determining the cost of content. Here you need to work with those wannabe product managers amazing producers to determine the developers required to produce a content container. This cost of content will be used to help determine opportunity cost of new content post-launch.

Gameplay Balance

Balance has sunk more “next big things” as well as elevated more mediocre vanilla games than any other property of a video game. Although normally regulated to the forum squabbling of “fighting game character X is sooo000 unbalanced”, balance covers everything that touches player agency and player perceived value.

Balance is the art of measuring progression and experiences, and making sure they stay within intended thresholds. These measurements will be taken from either spreadsheet models, metrics from playtest, metrics from automated test, load test, or full on simulations. Some common areas that product managers find themselves responsible for balancing are:

• Difficulty Use playtest, bots, or scenario-based automations to measure the likelihood of success or requirements of success with content
• Progression Player’s acquisition of levels, skills, stats, items, or any other advancement
• Economy Model your game’s economy through currency drains and gains
• Asymmetric Equilibrium Using matchup charts to balance character roasters or abilities available to players

Economy Balance Example

Modern games-as-a-service use one or more currencies that need to be balanced so players don’t receive too little or too much of each. Create a bottom-up model of the different ways players can earn or lose the various currencies in the game. Tie these drains and gains back into the content runrate model and see what the different segments wallets look like. Does one type of player have way too much of a currency? Is another segment always starved for more? Product managers are generally responsible for overseeing the economy. Work with designers to keep these models up to date or have your model directly consume the actual game data.

Character Balance Example

A lot of competitive games allow players to choose between a number of characters with asymmetric abilities, think 2D fighting games. Work with design to define scenario-based design pillars such as how long it takes a player to recognize an ability and dodge it before the ability hits them. Work with QA to create automations that cycle through these scenarios, gather data, and create scenario charts. Work with designers and QA to digest and use data for decision making or to communicate how decisions impacted this scenario based balance.

Monetization Strategy

The greedy reputation of modern game PMs is built off of games with horrible monetization strategies. I’ve seen it, developers have seen it, players have experienced it, don’t add endless layers of monetization only to be a human shocked Pikachu face when you face player backlash and a dogpile of bad press.

During pre-production you explored the overall business model, during production you develop the detailed plan and forecast its performance. There is more to a monetization strategy than telling the studio heads to sell cosmetics items in a shop.

Monetization strategies WIDELY vary from genre, platform, studio, release country and many factors. The common process is to:

1. Identify what you are selling
2. How often you want players to convert
3. The rate at which you need to create content to support that conversion

What You Are Selling

Product managers have thoroughly explored monetizing different properties of video games. Focus on what your players will get excited about and keep this decision focused on their needs. Here is a greatest hits list options:

• Power (weapons, armor, characters, PvP advantages)
• Progression (advance timers, experience multipliers, game over saves)
• Cosmetics (peacocking visuals, limited time outfits, custom nameplates)
• Ads (your player’s attention, sponsored content)
• Content (DLC, expansion packs, content packs)
• Subscription (pay to play, battle passes, premium subs)
• Digital Exclusives (soundtracks, early access to content)
• Physical Merchandise (apparel, action figures, art books)

Expected Conversion Patterns

1. Determine your play segments that will participating in purchases
2. Add purchases to your content runrate model
3. Measure the impact of their purchase on their experience

Monetization Content Creation Rate

Determine the rate of content creation to satisfy the purchasing appetite of your players. What do you need to produce to provide enough content for players to continue their purchasing patterns? This varies from monetization strategy to monetization strategy as supplying ad-driven revenue is going to be different than providing new characters for purchase.

I’ll leave this section with an interesting exercise: when you finish your monetization strategy ask “what if a super fan wanted to spend $500 on your game?” Could they? Should they? Do you want to allow players to hand you that much money? What do you want to give them in return? What should a non-IAP player experience if other players are paying $500 a month? These are great questions to ask yourself as you develop your monetization strategy.

Validation Experiments: The Secret of the Pros

One of the beautiful things of the modern internet is the ability to purchase ads against a vast variety of tags. You want to target Gen Z cross-stitchers? Easy. New parents that love emo music and live in the midwest? Probably. Product managers can also wield this immense power to get amazing customer feedback for very little investment. This is one of the powerful tools that I am always shocked is not used and abused by every game studio on the planet. This is the art of experimenting with fake ads.

Experimenting with ads is very simple. You create a fake game studio to purchase ads from Google, Meta, and other self-serve ad providers. You create different ad campaigns, each with the different variant you want to test. Have them all point to the same landing page. Count the source specific identifier for which ad directed them there.

Some experimentation ideas:

• Art Direction
• Color Palettes
• Game Name
• Camera Perspective
• Story Hook
• Gameplay Hook

Go to Market (Launch)

Go to Market is the third and shortest stage of video game development. This stage is where marketing and publishing partners take center stage to drive purchase funnel.

If pre-production is where ideas are born and production is where those ideas are brought into reality, launch is where you determine whether those ideas will make everyone rich or will everyone need to polish their resumes. Launching a game can be the single most exciting and rewarding experience in a product manager’s professional career.

Up until now, you’ve been modeling in spreadsheets, arguing with designers about progression curves, and reassuring leadership that you will in-fact earn back that bloated development budget. But when you launch, there’s no more hiding behind models. Player behavior is now reality.

The goal of the product manager during launch is simple: get and retain as many players as possible into the game.

Product managers need to:

• Understand when to utilize soft launches and hard launches
• Create a launch plan
• Contribute to the go-to-market strategy (GTM)
• Help manage community feedback
• Preemptively create a crisis management plan with other game leadership
  …you know just in case you don’t hit that 99 metacritic out the gate

Soft Launch: Your Game’s Dress Rehearsal

Soft launching gives you the ability to launch your game, make and learn from mistakes, then use the sands of time to rewind and launch your game again as if the first launch didn’t happen.

Soft launching is when you launch the game where your core audience isn’t, gather real player data, tune the product, and fix your worst assumptions before the real show. If you’re building a global mobile game, this often means Canada, Australia, New Zealand, or the Philippines. For PC and console, it may mean a limited creator beta or cross-market early access.

During soft launch, the product manager needs to gather as much external data and validation as possible. You’re measuring:

• Retention curves (D1/D7/D30)
• Content pacing (Do players reach end-of-content immediately?)
• Economy survivability (Are spenders starved or swimming in currency?)
• Live-ops pressure testing (How are the services and load balancers handling traffic?)
• Player sentiment from people who aren’t your coworkers (Reviews and forums)

This is where you validate your forecasts. If you modeled your economy assuming a 3% conversion rate and then discovered soft launch comes in at 0.7%, congratulations, you just avoided a disaster. Fix the value prop, tune prices, and try again.

Soft launch ends when your game hits its KPI targets consistently.

PM’s Role in Go-to-Market Strategy

Go-to-market (GTM) is the bridge between building a game and players actually playing it. It is the strategy that answers:

• Who are you launching to?
• Where and when are we launching?
• How will players discover the game?
• What happens when they actually play it?

Think of GTM as the marketing funnel plan. Where the Launch Plan is the immediate events before and around launch, GTM is the first phase of user acquisition that includes launch but also includes pre-launch awareness marketing and post-launch user acquisition events.

During GTM, marketing owns creative and distribution, PMs own the numbers that determine whether the marketing budget burns money or prints money. Before marketing starts scaling spend, they need your analysis on Net LTV, the difference between what a player is worth (LTV) and what they cost (CAC). If Net LTV is negative, you’re not running a business; you’re running a charity.

As a product manager, you will be responsible for:

• Forecasting LTV by acquisition cohort
• Stress-testing those forecasts against worst-case retention
• Working with UA teams to identify the cohorts that actually monetize
• Recommending when (and when not) to scale spend

Building the Launch Plan

Your launch plan is your studio’s map for navigating launch and all of its unexpected issues. It outlines what happens, when, and most importantly, what you’ll do when something goes wrong (which will 100% occur).

Generally, marketing will be responsible for putting together the launch plan, this is one of the few cases where we have to put our revenue generating fate in the hands of another studio discipline. As a PM, you will be responsible for the economy, monetization, metrics reporting, and player engagement patterns.

A proper launch plan includes:

1. Readiness Review: FTUE, progression, the economy, UX friction, content runway, server stability
2. KPI Thresholds: what numbers tell you “this is survivable” versus “sound the alarm”
3. Response Framework: who fixes what, and who you wake up at 4 AM
4. First 30 days of Live Content: events, offers, patches
5. “Break the glass” Changes: shortened FTUE, buffed rewards, nerfed difficulty spikes, emergency bundles

Launch is a time where the entire leadership team comes together and is on DEFCON 4 for the first week to month of the game being live.

Hard Launch: Welcome to the Thunderdome

Hard launch is when all your hard work finally bears fruit: marketing starts spending real dollars, creators start streaming, and players flood in with the collective power to destroy your servers, your economy, or your career.

During a hard launch, the product manager becomes the studio’s air-traffic controller. Everything passes through you:

• Metric abnormalities
• Player funnels
• Early economy shocks
• Unexpected player behaviors
• Sentiment bombs from Reddit
• “URGENT” Slacks from executives at 1:43 AM

Remember to use your metric funnels to root cause analysis issues, update your models with real player data, prioritize the biggest opportunities to tackle first, and communicate the business impact of different issues.

Launches are always chaos, learn to enjoy it and savor the moments, you’ll likely have a handful of these throughout your career.

Community Management & Player Feedback

The first week of a launch is when every player suddenly becomes a game designer, economy expert, and amateur psychologist. You’ll see feedback ranging from “10/10 best game I’ve ever played” to “this is a crime against humanity and you should be tried for it.” Remember, player sentiment is a compass not a KPI.

“But my studio has a community manager.” While community managers focus on nurturing online communities and bridging the studio with the players, you will be trying to harvest communities for insights on how to maximize player experiences. You are concerned about macro trends and sentiment as an actionable metric.

Put those insanely expensive LLMs to use to identify reviews, forums, Reddit, and Discord for:

• Trends in sentiment
• Game breaking bugs
• Balance issues
• Confusion around gameplay
• Surface player suggestions
• Identify new feature or content opportunities

Work with producers to prioritize and schedule updates to address these issues based on impact to game playability, reach and scope of issue, and level of effort to solve. Work with community managers on messaging for addressing issues. Use player sentiment to harvest the biggest metric moving opportunities from the community. Become the bridge between prioritization, outcomes, and communication to the actual players who are voicing their needs.

Crisis Management

Something will go wrong at launch, it is an irrefutable law of game development. A content bottleneck, a matchmaking failure, a broken store button, a bug that deletes inventory, a YouTuber who discovers an infinite currency exploit; launch is where Murphy’s Law goes speedrunning.

I’m stealing the best explanation for launch setbacks: during development you maybe had 100 QA and designers playtest your game across various versions and states. Once it’s launched 1,000,000+ players rain hours upon hours onto the same version of the game. Meta destroying strategies, game breaking bugs, and unintended design exploits will be discovered in a matter of days that would take the QA team months to years to uncover.

Your job as the product manager is to become a crisis manager.

If retention or engagement collapse? Start your root cause analysis at your metric funnels.

If revenue tanks? Check your economy, player progression, and store layouts to ensure the perceived value of premium items is as expected.

Here are the CYA steps of product management during a crisis:

1. Document everything (when did the issue get noticed)
2. Communicate with stakeholders (email is the paper trail of choice)
3. Root cause analysis (you built all those funnels for a reason)
4. Estimate impact (bottoms up model)
5. Build a plan of action (cross-functional leadership)
6. Communicate with stakeholders (yes, again)

Live Ops

Live Ops, or live operations, is the fourth and final stage of video game development. This is where games can live for decades and expand to dwarf the product that was initially launched.

Once the single most important revenue generating event, game launches have become only the starting line of a marathon in the modern video game industry. Live Ops has become the modern game industry’s business model, where video games continue development forever (or until players have moved onto different pastures).

Now we’ve reached the stage where product managers truly become dangerous, this is where PMs stop predicting what players will do and start shaping what players actually do. Here is where you will earn the infamous “product managers only care about revenue” title, all while hitting bonus milestones to get everyone in the studio paid. Here is where a product manager’s career is made by turning standard games in 10-figure ARR machines.

Rotating the Flywheel: Product Management During Live Ops

Pre-Production belongs to the design director, crafting and selling the vision of the game.

Production is the realm of producers, organizing the chaos of modern game development.

Launch is where marketing takes the starring roles, guiding the game to public availability.

Live Ops is where product managers finally take over. Time to make the money printer go brrr.

The goal of Live Ops is growth.

• Revenue growth
• Player growth
• Engagement growth

As the product manager, you will help guide the development team through data informed decisions while managing experiments, live events, business reviews, and strategic planning.

The virtuous cycle of live ops is:

1. Players generate new data
2. Analyze data
3. Identify opportunity
4. Build feature or content
5. Repeat

What Does a PM Actually Do

Here is a day in the life of a PM during live ops:

• Come in (early) and grab coffee
• Read industry business news (gamesindustry.biz)
• Check the KPI dashboards and metric funnels
  • Do light root cause analysis if needed
• Monitor live experiments
• Prioritize tasks with cross-functional leadership

This is what your weekly tasks might look like:

• Write weekly business report
• Write PRD/BRD/acronym for new feature document
• Set sprint direction
• Root cause analysis major issues
• Analyze player segment behavior
• Plan and execute new experiments
• Plan and schedule new events
• Play and report on competitive titles

Continuing the scale back, here is the outputs for a month:

• Give monthly business presentation to leadership
• Perform post experiment analysis
• Analyze player quality by acquisition source
• Plan major content updates
• Breakdown industry trends

Layered Events: How to Keep Player’s Attention

Novelty and goals are the backbone of retention. They give players a reason to come back tomorrow, get excited about releases next week, and plan out where their character is going to be next month. Events are the backbone of injecting novelty into a game, breaking up the core gameplay with new content, variant mechanics, and urgency generating rewards.

The general recipe for a great event:

is limited time, skinned with a unique thematic wrapper, offers exclusive rewards, has some form of community engagement, and is available to a wide segment of the player population.

Common events the biggest games deploy:

• Seasonal Events As generic as “winter” or specific as “cherry blossom”
• Holiday Events Specific holidays in-game or from target geographics
• Collaboration Events Crossover events feature outside IP
• Celebration Events Events for game milestones or anniversaries
• Real-time Events Live events that occur once
• Recurring Events Specific days or weekends where an event always happens
• Competitive Events Event-based tournaments or ranking competitions
• Community Events The entire game community works together on an event

Going hand-in-hand with events are well understood player goals. Players always need to know what they should be doing. What are their goals this session, this week, and on the long-term term horizon? Confusion is the enemy. Make sure at all times players know their:

• Short-term goals
  • What does the player want to accomplish this game session?
  • Complete a daily event, finish a questline, win X matches
• Medium-term goals
  • What is the player working towards and hopes to accomplish this week?
  • Acquire new gear, unlock a new character, reach X level
• Long-term goals
  • What are the player’s aspirational goals to work towards over months?
  • Achieve their full character build, conclude the main story, achieve x ranking

Experimentation in the Wild

Live Ops is where experimentation becomes the heartbeat of the product. Once real players are inside the game, your job is to constantly read the KPIs and metric funnels for opportunities to optimize the game experience for players. When you find those openings, you don’t fix them with opinions or meetings; you fix them with experiments.

A good experiment is a clean question: “If we adjust this reward cadence, do players stay longer?” or “If we shift the store layout, does conversion rise?” Every test should exist to prove or disprove a hypothesis about player behavior.

The work happens in the tight loop: observe → test → measure → ship or restart

Sometimes you’ll validate something you suspected, but more often you’ll discover that the thing you were certain about wasn’t true at all. Live Ops has a way of humbling everyone.

Experimentation in Live Ops is ultimately the discipline of continuous optimization. You’re tuning the product while it’s in motion, nudging behavior without disrupting the experience, and learning, over and over, what your players actually respond to. It’s the closest thing a product manager has to asking players for their unbiased opinions, wants, and desires.

Some of the highest grossing live ops games of all time do not beat the competition through superior gameplay, breathtaking art, or an unrivaled storyline. They beat the competition through hundreds of 1-2% improvements and optimizations over the course of years.

Sunsetting

Sunsetting is the final state of a video game. It is when a game is no longer profitable for the development team to continue working on, so the game is set to a final terminal state.

Sometimes sunsetting means offshoring the game to a studio in a lower cost of living part of the world. Sometimes it means switching the game off permanently. Sometimes sunsetting a game happens in phases, sometimes it happens overnight. Every studio or publisher should have a policy in place for sunsetting a game during live ops.

Regardless of the process, the product manager is integral in identifying when a game should be sunsetted. Some common reasons:

• Revenue is below cost of content
  • Business model preferences have shifted
  • Price war or race to the bottom pricing with competitors
• Player counts are too low
  • Market or genre zeitgeist has moved on from core mechanics
  • Thematic wrapper no longer en vogue
  • Audience has been captured by new game
• Opportunity cost is too high
  • Studio resources are better spent on higher potential games
  • Return on investment is too low for new game content development

As a product manager you should be most attuned to how much growth is left in a game or if a game is in a death spiral. Once games start losing significant revenue, they can’t spend as much on new content, which causes them to lose players, and in turn lose significant revenue; hence the name death spiral. Can you maneuver to recover revenue? Can you create excitement to re-acquire lapsed players? Can you identify a strategy game saving to deploy?

Sunsetting is a difficult event for developers and players alike. Communication and respect is the key to a successful sunset. Developers may have spent years and countless crunch hours bringing a shared vision into reality. Players may have spent large portions of their lives immersed in a virtual world and its community. Don’t suddenly shut the lights off on a server, give games the proper send off and celebration they deserve.

Owning Your Career

Product management careers can range from managing complex technical services on games that serve 100M players a month to being the sole product manager at a less than 30-person studio putting out small games on a quarterly basis. Regardless of your role, the same principles and strategies apply to starting, growing, and future proofing your career.

Think of your career as the ultimate RPG. You roam around town before your adventure kicks off (breaking into game PM), you find yourself killing rats with an oversized stick until you can afford good gear (growing your career), and finally you reach the end game of RPG character development where you are killing gods and staying relevant in an ever evolving meta (future proofing your PM career).

Breaking Into Game PM

There is no one-size-fits-all path to breaking into the video game industry as a product manager. I transferred to product management while at a mobile game studio from a Senior Gameplay Engineer role when a product manager suddenly gave their notice and the product team needed to fill the role fast. I will walk you through some common pathways but if you don’t fit into a cookie cutter pathway don’t be discouraged, some of the best product managers I have worked with come from untraditional backgrounds.

Although product managers have a wider variety of backgrounds than engineers or animators, there are still some patterns found on product manager resumes. The most common degree found is an MBA, particularly from a top 15 MBA program (use whatever rankings you prefer). Outside of an MBA, some common undergraduate degrees are computer science, marketing, economics, business, and engineering. Product managers are required to build metric models, understand large game systems, and be familiar with technical tradeoffs so attracting developers with more mathematical or business focused degrees is not surprising.

Switching to Product from Another Discipline

A lot of product managers joined the role through other functions such as data science, design, marketing, and engineering. The best way to make this transition is to:

1. Setup recurring meetings with the product leadership (Director of Product or Head of Product Management)
2. Let them know you want to transition to product management
3. Ask if there is any away team work you can perform or help with to get your feet wet moonlighting as a PM

Enthusiasm for product management and business, your reputation within the studio, and a willingness to grind it out on any available PM work while performing in your current role will go a long way. Your goal is to make the decision to transfer you to product management as easy as possible once a product role opens up.

Lateralling to Video Games as a Non-Game PM

Another common path is laterally as a product manager in tech or similar industries into video game product management. One of the easiest PM transitions are B2C (business-to-consumer) PMs because they already are equipped with a consumer product knowledge base. B2B SaaS PMs have easy transitions to technical product management roles overseeing services, tools, and automations. In reality any product manager can transition to the video game industry, your experience as a PM can be easily applied to video games by learning the video game industry dynamics:

• Studio, publisher, platform relationships
• Devices and ecosystems differences
• Different implementations of premium and F2P business models
• Trends in genres, gameplay, and content
• The millions of reasons one game succeeds where others fail

Tradeoffs Between the 2 Approaches

A common question is if you can’t get a video game product management role, should you take a non-PM role at a game studio or a PM role outside the game industry? In my opinion there is no hard answer, but instead a sliding scale.

Comparing a technical game designer role on GTA VI or Fortnite to a product manager role at an unknown medical device company makes the game industry role a no brainer. Comparing a QA analyst role at a small mobile game company to a product manager role at Google or Meta, well lets hope I don’t have to break this one down.

In a broad sense, I would recommend the product manager role over non-product role at a game studio because the transition is easier. Once you are an established product manager, you are likely 1-2 jobs away from your dream game PM role. But roles within art, UX, QA, HR, and even certain marketing and engineering roles are going to have a difficult time lateraling to product management. Think “I have a track record of delivering results as a product manager and now I want to do that in video games vs I am a really good server engineer and now I want to be a product manager.”

That Last Mile to Landing the Role

Sometimes applying to product manager roles can seem like trying to look through an opaque window, but aspiring product managers can stand out by taking action now:

• Get published on video game websites Break down monetization trends, retention and engagement driving game systems, or overall business strategy on sites such as gamesindustry.biz or gamedeveloper.com.
• Post case studies on LinkedIn Write about individual game or studio moves on a regular, recurring basis. Comment on posts by leaders in the game PM community.
• Present at industry conferences Getting a presenter slot at top industry events such as GDC, DICE, or PAX provides credibility and validation on you as a leading voice in the video game industry.

Your goal is to cover your weaknesses. If you are a product manager outside the game industry, you need to assure studios and publishers that you understand the nuances of the video game industry. If you work in a non-product role in the game industry, you need to assure product leads you have the business acumen and strategies to drive results. If you are neither a PM or work in the video game industry, get a highly sought after advanced degree (very difficult) or start making waves on industry publishers and at industry events .

Choose Your Character: Different types of PMs

To add to the confusion, there can be a dizzying array of product management roles from technical to marketing, at publishers or studios, or within large AAA studios or agile mobile game companies, each with different skill-sets and responsibilities. Here is a quick breakdown:

• Product Manager – a catchall title for all product roles, generally a game level role where you will be managing a mode, feature, or entire game
• Technical Product Manager – a product role that is more focused on technology, typically overseeing services, tools, automations, or other technical domains
• Product Marketing Manager – NOT a product role, this is a marketing role
• Growth Product Manager – a product role that focuses on user acquisition, driving a specific part of the product lifecycle and focusing on marketing funnels and activation
• Data Product Manager – a product role focused on gathering, sanitizing, and structuring data, generally working within data science or data analyst teams

Career Growth

You did it, you’ve broken into one of the most competitive roles in one of the most competitive industries. You are the envy of thousands of video game hopefuls everywhere. Being the business minded developer you are, about 3 weeks into your first product role you may find yourself wondering “how do I make more money”.

Product manager roles can be confusing as titles, pay scale, and responsibility do not match up from company to company. Some companies have people making mid-six figures with the simple title of “product manager”, other companies give “senior product manager” automatically to every inexperienced MBA graduate. Here is a general breakdown of product manager titles and their responsibilities:

• Associate Product Manager An entry level product role for those just starting their careers. They will be focusing on a single feature or part of a game with a PM mentor overseeing their work. Promotion comes when they can operate independently and demonstrate small wins.
• Product Manager The mid-level catchall title. They will own a feature, game mode, or possibly an entire game. Promotion comes when they can deliver larger impact and provide larger scope strategies.
• Senior Product Manager This is for seasoned product managers who need little to no direction from product leadership. They will own the most important parts of games or be responsible for entire metric categories such as revenue or engagement. Promotion comes when they either demonstrate impact beyond the scope of their single game or move into people management.
• Principal Product Manager This is generally the highest level individual contributor (IC) product role. They will own initiatives and strategies across a portfolio of games or products. There is no promotion as generally this is the end of the line (terminal role) for those who wish to stay as ICs in product management. Similar title is Staff Product Manager.
• Lead Product Manager This is the first people manager product role that comes after senior PM, parallel to principal PM. They will be responsible for managing a small team of PMs, setting strategies to reach goals, and spending roughly 20-30% of their time doing IC work. Promotion comes when they consistently deliver quarterly results exceeding goals, lead their reports to career growth themselves, and impact cross functional decision making across the company. Similar titles are Group Product Manager, Manager of Product Managers, and Associate Director of Product.
• Director of Product (DoP) This is generally the highest level product manager on the studio side. They will own the entire product strategy and product performance of a product line and are first on the responsibility line when revenue underperforms. Promotion to the executive level involves leadership and influence as this role and any role above it on the PM ladder will have no hands on execution. Similar title is Head of Product Management.
• Vice President (VP) of Product This is the first executive level product role. They will set overall product direction across a subset of the company while their day to day is meetings to approve timelines, discuss business performance, and help with stakeholder management.
• Chief Product Officer (CPO) The highest product role. They will set overall product direction across the entire company. This role works directly with the CEO, and serves as the product representative during senior executive and board meetings.

There it is, so simple. Just start as an associate and before you know it you’ll be a CPO getting paid 7-figures to fly in private jets and sit in meetings. Ok so maybe not. The truth is product management advancement can be extremely political. Think about it, it’s the only major role within the video game industry that focuses exclusively on revenue and where most people have prestigious graduate degrees from elite private universities (unfortunately I contribute to this stereotype). With talented professionals armed with a drive to succeed, trying to compete for limited leadership positions can become a full time job in itself.

The Real Level-Up: Promotions

The most common questions I get from direct reports or prospective PMs about promotions and career growth: How do I get promoted fast?

Promotions are more based upon reputation and visibility than impact or competency. Although having impact and being very good at your job makes getting promoted so much easier, how your leadership stack feels about you trumps everything. You need to be seen as a leader with the scope and influence of the role you are trying to reach before you get the promotion, seen being the key word. This is where fiefdom building comes in as leaders try to acquire more teams and scope to expand what they “own” in hopes of being viewed as the most viable promotion candidate.

Bonus tip: When it comes to promotions, your skip level’s opinion is more important than your direct supervisor’s opinion.

How can I reach VP of Product at company X in 5 years?

Surprisingly a common question of newly minted mid-level PMs (this lets you know how power hungry PMs can be). Here are the two paths I’ve seen PMs hit VP in a short amount of time:

• Path 1 – Have a leadership vacuum above you and own revenue on a premier product. This means that roles are missing in your leadership stack and there is a chance to ride the promo train up the ladder immediately. Owning revenue on a premier product of your company puts you in the most visible roles with the most responsibility. This path requires you to be the “hot shot” that is sucked up by the vacuum, receiving promotions every 1-2 years.
• Path 2 – Leave your company for a less competitive company where you are either one of a few product managers and “own” the product management discipline. Here you should have an easier time positioning yourself as a product leader with a bloated title. Next, you need to get industry notoriety and visibility by speaking at conferences, writing highly acclaimed articles, and delivering outsized returns on your products. From here you need to network your way back to company X to lateral to a VP of Product role. This path requires you to find a less competitive company that is either still respected enough for the eventual lateral back or has enough growth potential for you to take them to a level where you can lateral back.
• Bonus Path – Co-found or join a startup where you can get a VP of Product or Chief Product Office title and get acquired by your target company while fighting to keep your inflated title. This is extremely risky but yet, I’ve seen it done by someone with 0 product management experience.

I’ve hit a ceiling at company X, what do I do now?

I’m taking this question as if leaving company X is not possible, so what do you do if you’ve kinda just plateaued? When you’ve stalled out at your company there are generally two reasons:

1. Nobody above you is getting promoted or leaving the company, so there are no openings for you to position yourself for
2. You’ve hit your ceiling in terms of office politics or leadership capabilities

For the first issue, you either need to:

• Change orgs
• Change PM responsibilities
• Create your new product you can own and have the opportunity to lead

Sometimes companies have multiple product lines or organizations, so if your current position has stalled out, find the org that is growing and network your way there. It is much easier to get a promotion on a product that is growing and hiring than on a consistent blue chip product where leadership has been static for years.

Changing PM responsibilities means lateralling to another PM role. You can go from game side PM to service side technical PM. You can go from a PM that focuses on economies to a growth PM that focuses on user acquisition.

Creating your own opportunity means pitching new business lines, products, or games to upper leadership where you would be the edge PM, allowing you to secure promotions alongside the growth of the new product. This is especially attractive because it puts the locus of control firmly in your hands, letting you eat what you kill.

The second issue is much simpler, you need to grow as an employee and get excited about learning again. Turn to books, podcasts, YouTube channels, blogs, newsletters, and anything else you can get your hands on. Learn from history’s best business leaders and entrepreneurs, and experiment applying it to your current role.

Brownie points: Ask leaders for their favorite books and podcasts then talk to them about it.

One final note. Anyone who has worked with me knows I am not only terrible at politics and sometimes self-sabotage my career in the name of some form of integrity. This is 100% true and I would be a director of VP if I took my own advice. But my words are still true and this advice comes from my observations of the best professional climbers I have ever worked with.

The Future of Game PMs (Tech, AI, UGC, Platform Shifts)

The video game industry is constantly evolving with new technology, new business strategies, and a new generation of players with different tastes and needs. Product managers will be a part of that evolution in one way or another by connecting a game’s business outcomes to the game’s design. Whether the future is AI, blockchain, cloud gaming, or something else altogether, the product manager role needs to be there to lead game teams to their business objectives and to be held responsible when revenue drops twenty percent. Savvy product managers need to stay curious and understand the opportunities of every new trend.

Even as I write this section, the latest techniques in pre-training LLMs will be out of date by next week. However, there are some ways to guarantee you will be able to produce an opinion on whatever new hyped technology becomes a black hole for VC money in silicon valley.

1. Stay up to date on the other entertainment industries
   You should always be aware of the business strategies in our sister industries of music and movies. Music moves the fastest and you can see the impact of new technologies or business models here first. Movies provide parallel blockbusters to study and often feature new visual technology first, which is later approximated for real-time processing in video games. 
2. Stay up to date on foreign video game markets
   North America and Japan are not always the most innovative locals for the video game industry. Turkey, Israel, Korea, and China are leading the way in new business models, innovating in game genres, and creating new AAA production pipelines. 
3. gamesindustry.biz
   It’s this simple: if you are or want to be a video game product manager you need to be reading this website on a daily basis. gamesindustry.biz is the de facto business website for the video game industry.
4. Technology Stack
   The speed at which technology moves these days is neck breaking (either that or I’m finally getting old). You will need reliable sources to explain new technologies, the new and existing players in that technology, and the business strategies behind the moves being made with that technology. I get outsized returns in paying for news rather than using ad-supported news. I’ve been using Stratechary and The Information for over a decade at this point and cannot recommend them enough.
5. Business Stack
   There is an endless ocean of knowledge to gain from non-entertainment and non-technology business strategies. In fact, some of the innovations within entertainment were just adopted from more traditional business such as B2B SaaS. I recommend keeping a broader view of business, finance, and the arts, but not to the level you need for video games, entertainment, and technology. My absolute favorite is The Economist, the best periodical in the history of publications. One area to not discount is what I will call airport business books, these are the business books found on NYT Best Sellers lists and generally communicate business strategy or management in an easy to digest way.

Final Thoughts

I want to reinforce some learnings, if you take nothing away from this playbook you should take this away:

• Product managers are responsible for business outcomes.
• Product managers deal in data, revenue, and reality. Artists, designers, and other less quantitative roles can think about feelings and dreams and stuff. You need to create guardrails to raise the floor on business outcomes for your games.
• You should be able to see into the matrix and understand everything going on in the game, your players, and the potential target audience.

Product management can be an extremely rewarding and creative role. There is infinite growth ahead of the video game industry, it only needs great business leaders to guide it there.

Glossary

• A/B Test – A controlled experiment comparing two variants (A vs B) to measure impact on a target KPI
• AAA – Triple-A: high-budget, high-production-value games developed/marketed by major studios/publishers
• Active Users – A general category for DAU/WAU/MAU-unique players who performed a qualifying action in a time window
• ARPDAU – Average Revenue Per Daily Active User: Gross Revenue / DAU for a period
• ARPPU – Average Revenue Per Paying User: Gross Revenue / Paying Users for a period
• ARPU – Average Revenue Per User: Gross Revenue / Active Users for a period
• Business Model – How the game delivers value and captures revenue (premium, F2P, subscriptions, etc.)
• CAC – Customer Acquisition Cost: the average cost to acquire a new player
• CCU – Concurrent Users: players active at the same time; useful for capacity planning and multiplayer health
• Churn – Players stopping play (or pay) over a period; often expressed as a churn rate
• Cohort – A group of players who share a defining attribute (e.g., install date, source) tracked over time
• Competitive Analysis – Analyzing competitors to understand strengths, differentiation, and how your metrics/features compare
• Content Cadence – The scheduled rhythm of updates/events/offers meant to maintain engagement and revenue
• Control KPI – A KPI monitored during an experiment to ensure important non-target outcomes aren’t harmed (a “watch list” metric)
• Control Segment – The baseline segment in an experiment that receives the current experience for comparison
• CPO – Chief Product Officer: executive accountable for product strategy and outcomes across the org
• Conversion Rate – The % of players who complete a desired action (e.g., first purchase) out of those eligible
• D1 – Day-1 retention: % of new users who return one day after their first session
• D7 – Day-7 retention: % of new users who return seven days after their first session
• D30 – Day-30 retention: % of new users who return thirty days after their first session
• DAU – Daily Active Users: unique players active in a day (based on your definition of “active”)
• DAU/MAU – Stickiness ratio: DAU / MAU.
• DLC – Downloadable Content: additional content released after launch (free or paid)
• Director – A senior leadership level (often between manager and VP) responsible for a function/area
• Economy – The system of currencies, sources, and sinks that governs how players earn/spend and progress
• Engagement – How much and how often players interact with the game (sessions, time, depth, frequency)
• Experiment – A controlled test that changes something for a subset of players to measure causal impact on metrics
• First Time Conversions (FTC) – A metric measuring the first time a player buys
• Forecast – A prediction of a future outcome (users, revenue, retention) based on historical data and/or a model
• Friction – Points in a flow that create effort/confusion, reducing completion, engagement, or conversion
• FTUE – First Time User Experience: the onboarding/tutorial journey for new players
• Flywheel – A reinforcing loop where improved engagement/retention/revenue funds more content/quality, creating momentum
• Gacha – A randomized-reward purchase mechanic (often with rarity tiers) commonly used in F2P monetization
• Gross Revenue – Revenue before platform fees, refunds, taxes, and other deductions
• GTM – Go-to-Market: positioning, channel strategy, timing, and execution plan for launch
• Hard Launch – The full public release after validation-typically paired with major marketing and scaling
• Head of Product – A senior product leader role (title) responsible for product strategy and execution across a scope
• Hypothesis – A clear, testable statement predicting how a specific change will impact a metric for a defined population
• IAP – In-App Purchase: a transaction inside the game (currency, items, bundles, etc.)
• IC – Individual Contributor: non-manager track role level
• IP – Intellectual Property: owned or licensed creative assets/brand underlying a game
• KPI – Key Performance Indicator: a small set of the most important metrics that define success for a goal or time period
• Layered Events – Live Ops pattern using multiple overlapping events/loops to sustain attention and goals
• Launch Plan – A coordinated plan for milestones, marketing beats, readiness, and success criteria around launch
• Live Ops – Ongoing post-launch operation: events, content updates, tuning, and optimization loops
• LTV – Lifetime Value: expected total revenue per user over their lifetime in the game (often by cohort/source)
• MAU – Monthly Active Users: unique players active in a month
• Market Sizing – Estimating the opportunity size for a game/product (commonly via TAM/SAM/SOM)
• MDE – Minimum Detectable Effect: the smallest effect size you can reliably detect with your test setup
• Metric – A single quantitative measure of some aspect of the game, its economy, or its players’ behavior/outcomes
• Metric Funnels – A decomposition of a top KPI into component metrics (often multiplicative/additive) to reveal the real levers
• Monetization – How the game generates revenue and how revenue systems are designed/optimized
• North Star – A guiding “true north” outcome metric used to anchor decisions and interpret experiment results
• Obfuscation – Hiding or blurring information to influence behavior; discussed as a risk to the player contract
• Opportunity Discovery – Forward-looking analysis to find and size the highest-upside opportunities to pursue next
• p-value – A statistic used in hypothesis testing that estimates how likely the observed difference could happen by chance under the null
• Player Contract – The implicit agreement about fairness/value between player and game; breaking it erodes trust and metrics
• PM – Product Manager: owns outcomes by aligning stakeholders, defining problems/opportunities, and driving delivery
• Pricing – The price points and structure for purchases/offers and how they map to perceived value and segments
• PRD – Product Requirements Document: a written spec describing the problem, goals, scope, requirements, and success metrics
• Product Strategy – A coherent set of choices on where to play and how to win (target audience, positioning, bets)
• Progression – How players advance over time (levels, power, unlocks) and the pacing of that advancement
• QA – Quality Assurance: testing discipline that validates stability, correctness, and release readiness
• Randomization – Assigning players to variants randomly to reduce bias and support causal inference
• Retention – The rate at which players return over time (often tracked with D1/D7/D30 and cohort curves)
• ROI – Return on Investment: value generated relative to cost for a feature/system/campaign
• Root Cause Analysis – A structured method to identify the underlying cause of a metric shift or issue, beyond symptoms
• RPG – Role-Playing Game: genre featuring character progression and systems-driven advancement
• Safeguard KPIs – Guardrail metrics tracked to ensure experiments don’t damage core health while optimizing a target
• SAM – Serviceable Available Market: the portion of TAM you can realistically serve with your product constraints
• Segment – A subgroup of players defined by shared traits or behaviors (used for targeting, analysis, or tests)
• Segmentation – The process of splitting players into segments based on behavior, lifecycle, value, etc
• Session Length – The amount of time a player spends per session (often reported as average/median)
• Soft Launch – A limited release to validate retention/monetization/ops before scaling; used to de-risk hard launch
• SOM – Serviceable Obtainable Market: the realistic share of SAM you can capture in a timeframe
• Stakeholder – Any person/group who influences or is impacted by product decisions (leaders, partners, teams)
• Statistical Significance – A conclusion that an observed difference is unlikely due to chance, given a test and threshold (alpha)
• Stickiness – A type of metric that measures the rate of play for players
• TAM – Total Addressable Market: total potential demand if you had 100% share
• Target Population – The set of players eligible for an experiment (e.g., new users after a start date)
• UA – User Acquisition: marketing-driven growth-getting new players into the game
• UGC – User-Generated Content: content created by players rather than the dev team
• UX – User Experience: how intuitive, usable, and satisfying the game is across flows and systems
• Value Proposition – The ratio of cost to reward that players make for a given component of the game
• Variant – A modified experience served to one or more segments in an experiment
• VP – Vice President: senior leadership level overseeing major areas/teams
• WAU – Weekly Active Users: unique players active in a week