Managing by values

A lot of traditional companies have posters on the wall of the office with written value statements (like honesty and innovation) that are often vague and ignored. Most people in those companies don’t even know about these values or don’t understand why they exist. Usually the executive team review and writes the values once a year and think that that’s enough for the company.

Values are the things you do and live by. It’s also important that the values communicate with the purpose and mission of the company. At Cupcake, our values influence every decision we make and how we manage the company on a daily basis. Because of that, it’s important for everybody to know, understand and live the values of the company.

The Cupcake values are:

  • Focus on the player
  • Be goal oriented
  • Keep it simple
  • Use data to learn
  • Be a team player
  • Self managed

Here’s how we use them to manage the company.

Focus on the player

First of all, it’s important to know who your users are. If you have more than one game, it’s very helpful for all of them to have the same users, so you can learn with all your games. If you have a game for young men and another for older women, it’s complicated to balance the learnings. This is very important specially for gaming companies that are starting.

Focusing on the player means having a great game where they have fun or having great customer service. Every interaction with the players must make them feel special. You should focus to not make something that will hurt their experience, even if that means making more money. Revenue is the result of a great experience.

At Cupcake, most of our users are women over 45 years and most of them play all of our games. In the end, it’s our customer that is our boss. If they are not happy with us, they are going to spend their money somewhere else.

Be goal oriented

The best way to get to somewhere is to have a goal in mind. If you don’t know where you want to go, anywhere is enough. It’s important to know what you want with your game or even a feature inside a game. It’s not good to build something just because it is cool.

If you want to make a living or have a impact in the world with games, you have to take a step back and think about what you want to achieve. Having a goal in mind, now you got to plan how you are going to achieve it. A goal without a plan is just a wish. Failing to plan is planning to fail.

Everything we do at Cupcake is related to a goal. Our goals help us achieve our mission of being the #1 casual brain puzzle games company in the world. The bigger the goal, the more challenging it will be. Continuous improvement is key to achieving those goals.

Keep it simple

It’s very common for a game developer to want to start building something that is going to take years to be ready. It’s also very tempting to be polishing the game for a long time instead of just launching it. A game is never perfect, that’s why done is better than perfect.

We are very careful with this at Cupcake. We like to use the Pareto principle (80/20 rule) that says that 80% of a outcome comes from 20% of the actions. If you are building a prototype, it has to be fast. If you are building something for the final user, you have to focus on what is going to make the most impact for her.

What you decide not to do is as important as what you do. If your goal is to build great puzzle games and you have an idea for a shooter, are you going to build it? I hope not. Focusing is about saying no. Time is a finite resource so it is necessary to focus on the long term objective. Keeping things simple allow you to learn faster and delivery a better experience to your players.

Use data to learn

This is something very important specially for F2P games. If you don’t have data to measure, you can’t learn and improve your game. When you build something, you got to measure it and learn from it. It’s the build-measure-learn feedback loop. Just gut feeling is not enough.

You can also use data to see if there is any technical problem instead of hoping everything is ok. You can see how your goals are doing and decide what needs more attention. You can even measure the happiness of your team and find ways to have an happier environment at work.

There is no guessing at Cupcake. Decisions are made over data. If you want to learn something, you have to check the numbers instead of making assumptions.

Be a team player

Game development is a multidisciplinary activity. It’s hard for someone to be good at more than one thing and that’s ok. It’s better to have a very good programmer instead of a so so programmer/artist. You have to build on strengths instead of weakness.

In order to make great games, it’s important to have people willing to help one another, the team and the company. Communication is very important. There can not have space for selfishness.

Everybody at Cupcake do home office and we have people all around Brazil. For us, it’s very important to find smart ways to interact with each other in order to have things done.

Self managed

It’s better to work smart instead of work hard. If hard work was all that was needed to achieve success, construction workers would all be millionaires. There is a lot about working long hours in the game industry but that’s not healthy in the long term. That’s why it’s very important to be organized and have ownership of the work.

People need to know when they are more productive and how to accommodate time for each of the tasks. We expect people to figure out the best way to get things done, as well as to step up when they need help from the team.

We also don’t like the type of environment where it is necessary to have a boss saying to everybody every little detail of what has to be done. Great people are self managed. They don’t need to be managed. Once they know what to do they will figure out how to do.


Everything we do at Cupcake is focused to improve the experience of our users (Focus on the player). For that, we need to understand how their experience is doing (Use data to learn) and work on goals on top of what to improve (Be goal oriented). To reach those goals, we need to focus on what really matters (Keep it simple). We also need to work as a team to reach the outcome (Be a team player) and know how to be productive in order to finish what needs to be done (Self manage).


You must have seen a lot of different acronyms in game analysis and were not sure what each one meant. Not everyone should be a specialist on them, but understanding what they mean is important to improve team communication. Let’s go:

  • LTV (lifetime value)
  • CPI (cost per install)
  • KPI (key performance indicator)
  • ROI (return on investment)
  • DAU/WAU/MAU (daily/weekly/monthly active users)
  • ARPU (average revenue per user)
  • ARPPU (average revenue per paying user)

Support material:

5 lessons from Flappy Bird

Flappy Bird is a game that became popular with help from a PweDiePie video. The game was live for more than one year. In a certain moment the game was making US$ 50 thousand per day with ads. Many believe that the success of the was pure luck, but the game shows some important lessons as well:

  • Simple control: only one finger necessary to play;
  • Easy to learn: very clear tutorial and very easy to understand what to do;
  • Fast matches: it is very easy to lose on the game, making matches very fast with very high replay;
  • Progress: the progress of the player it is very clear and listening to the sound every time you pass a barrier becomes enjoyable;
  • Points: the game always incentivize the player to beat his own record.

Suport material:

What is soft-launch?

You don’t need to have your game completely done to know how it is going to behave. After validating the idea of the game through the MPV, you can also do a soft-launch.

Soft-launch means launching your games to a limited audience to gather information before doing the official and/or global launch. Usually game developers choose a specific country. Many choose Canada for having a similar profile than United States or choose their own country because of the proximity.

The most important metric in a soft-launch is retention. Some games stayed in soft-launch for more than a year before reaching the numbers desired by the team. It’s also important to measure how the monetization of the game is doing to make sure that is worth the investment with the official launch.

Cupcake games are always launched first on Facebook as a soft-launch, because measuring the information on web is much faster than mobile. In mobile you need the user to update the game in his device, which won’t always happen, as well as a longer review time. Updates on web are immediate and it is possible to have numbers the next day. We develop for mobile when the game reach the quality that the team desires. Nowadays we are doing this with Numbers of Gold, that was launched in the end of March on Facebook.

No guesses, use metrics

We talked before about the importance of validating your game through MVP. You can lose a lot of time discussing ideas that can be tested directly with final users and validated through metrics. If you are not sure if something is gonna work, it’s not up to you to guess what the user is gonna think. Put the game in front of your users and use metrics to be able to measure if things are working or not. This also works for improvements to your game. You’ll notice that what you think is gonna work doesn’t, and what you think it is not gonna work actually does. Seeing the impact in retention of the changes made is a good way to check how things are going.

Before launching Numbers of Gold, we had a lot of doubts about things that could work or not. We decided not to waste time with philosophic discussions and launched the game with these doubts. After a few days we had enough data to see what was working and not working. This saved us a lot of time and allowed us to work in other important stuff.

Support articles:

Why Candy Crush has lives?

If you think Candy Crush has lives to push you into spend money on the game, you are wrong. Having lives in a game is a strategy that influences the retetion of a game.

Imagine that you have a whole cake to eat and you can eat as much as you want. If you eat half of the cake right away today, you are not gonna want eat more tomorrow. However, if you eat just a slice today, you will want to eat more tomorrow. After a few days you ate the whole cake.

This metaphor is a good example of the use of lives/energy system in a game. Running out of lives makes you stop playing making you want to play more. This has a great positive impact on retention. Work balancing your game so that the player don’t play continuously without stopping. That’s right! Make the player go away, but with a good reason to come back.

Having lives is something that works very well on Cupcake games. Spending money to recover lives is a very small piece of our revenue. And it shouldn’t be different.

Support articles:

GDD = bullshit

We already passed the time where one of the first things to be done was to write the GDD (Game Design Document). In this document, the game designer writes everything that the game is gonna have. From that on, the rest of the team (programmers, artists, etc) did the rest. However, how the game designer can be so sure that the game will be a success?

Nowadays, GDDs, especially in F2P games, are used to document information (at least it should) of everything that worked and didn’t work. The GDD is something that evolves with the game. We’ll talk more about how to measure what it’s working and not working in a future post.

At Cupcake we always start with a prototype of an idea that can come from anyone. The game designer has an important role evaluating if the idea should go ahead or not. If the idea doesn’t go on, it’s necessary to understand why. Same if it does. Do an MPV to help. Killing an idea is not something bad, on the opposite, it’s a learning opportunity. You should have a lot of ideas and learn with all of them, just don’t write a GDD before having a game.

The importance of First Time User Experience (FTUX)

A lof of people talk about UX but few talk about FTUX. Products care a lot about user experience inside the game, but what is the point of having a good overall experience if the first one is not good? Hence term First Time User Experience (FTUX), the first experience of the user with your product/game.

Nowadays with F2P players give up on a game in seconds and move to the next. You have a few seconds to convince the player that your game is worth to be played. Don’t waste time showing cutscenes, menus and maps before the game really starts. Make the player go right to the gameplay. If the player doesn’t understand how to play they will leave. That’s it.

Make the first experience with the game easy and fast. If the first interaction with the game takes too much time to end, the user will get bored and leave. If the game is level-based, make sure the first level is a fast winning. It will make the player liberate dopamine fast. The time spent by each player can be different but the experience needs to be good for everyone.

On Cupcake games, like Numbers of Gold, we do exactly that. The first levels are fast and give a feeling of fast progression. Play them to understand how it is done. If you already played them, play them again, but this time with an analytic view.

Support articles:

MVP for games

MVP (Minimum Viable Product) is a very popular term around startups. It is a way to test a product with the miminum amount of features. Instead of waiting a year to launch a product full of features and find out what is wrong, you can launch the product in a few months with a few core features and find out what needs to be improved. How to do the same with games?

First, it’s important for you to understand what are the key parts of your game. What is the main mechanic of your game? That’s what you need to validate. If the main mechanic is not fun, it doesn’t matter how many extra features you put in your games. After validating the main mechanic, you need to validate every new feature. Looks like something that takes work, but in the end you are saving time by not developing things you don’t need.

Cupcake launched its first two games (Letters of Gold and Words of Gold) with 20 levels each. The mechanic used on Letters of Gold was validated in games in English, but not in Portuguese. That was our validation goal. In Words of Gold the goal was to validate if the mechanic from scrabble would also work in single player mode.

Support articles: