Penguin Piste is a small action platformer created by Alex Gidge under the studio name Isolation Games. It’s available on Android devices and was developed with the Unity game engine.

Aim of the project

  • Make my first game
  • Transfer applications programming skills to the game dev world
  • Release a game to the public
  • Make something I find fun
  • Satisfy a craving for snow through summer!

Development Process

I wanted to build this game with a very structured development process from the start, and achieved this very well by:

  • Using Github projects to manage development lifecycle
  • Split development work down into adequately sized tickets
  • Ran weekly sprint process with backlog reviews, sprint planning & story points.
  • High commit standards to make dev lifecycle easier

Game Design

I wanted to make a game that I enjoyed, but was still within reach of my current ability whilst still pushing myself to learn throughout the build.

The mechanics have been designed to be simple, easy to learn, but still have a possibility for depth without adding on complexity to player inputs. Using one player input (tapping to make the character jump) I was able to make level design the focus of depth & complexity.


Alto’s Adventure was a big inspiration on my game’s setting, mechanics and game feel. Where I really differed from the developers, Snowman, was in wanting to increase intensity & difficulty. Because I found Alto’s adventure relied too much upon achievements to keep players interested in the long term. I wanted my gameplay to be the main driver for play sessions and so opted for levels with increasing difficulty instead of an endless scroller.

Angry Birds had a large impact on the framing of my game. Where Alto’s Adventure helped with the gameplay itself, Angry Birds was the inspiration for my level system. As a bit of a completionist gamer, I Rovio’s method of measuring player performance with stars. This style gives the player a clear concept of the win/lose state in the game. As well as giving replayability opportunities by attempting to reach a specific high score target on a previously completed level.

Firewatch was used as a minor artistic influence, sticking to a narrow colour palette, and using shades to show contrast between background and foreground. In the end, following that style has lead to a game that’s easy on the eye but didn’t require high level artistic skills to make.

Release of Version 4

Game Difficulty and Player Retention

Setting the right difficulty has been complicated and is still not right in the game as it currently stands. Where I’ve been able to satisfy a very small minority of hardcore players, I’ve struggled to set the skill entry point low enough for the majority of players.

Version 4 of the game saw massive decreases in difficulty on levels 1 and 2. With minor changes to all other levels.

This was done in an attempt to help players by providing a shallower difficulty curve. The theory being that the difficulty was putting off more casual players.


The *Data shows the levels were definitely easier, but it subsequently didn’t impact the win/loss ratio of any future level. suggesting the gameplay’s difficulty is based on the level design, not the player’s understanding of mechanics.

The bar chart here clearly shows a large increase in wins for the levels 1, and 2 of 27% and 50% respectively.

My target win/loss range would be between 20 and 40% win to loss ratio.

The difficulty of the unchanged levels is consistent between versions, showing a strong reliability in the data set.

*Data obtained from 197 anonymous players who played levels a total number of 1,419 times, averaging 7 level attempts per player.
Players were counted as users who pressed play. Users who are yet to open the app, or opened it but didn’t press play are not included in this data.


After the difficulty changes above, I saw a direct impact on player retention across the altered levels. In the line chart here, you can see the percentage of players who played each level, split by version.

100% of players played level 1 (obviously)
42% of players continued onto level 2
23% of players continued onto level 3
(Shown in blue)

The new version (Shown in orange) saw a 22% increase in player retention across level 1 & level 2 after the difficulty changes, suggesting this should be applied to all 9 levels as a high priority

Difficulty & Retention Conclusion

Analysing the data in this way has given me a really good insight into my players and how they’re playing my game. It’d be safe to assume difficulty is one of the biggest factors negatively impacting player’s enjoyment of the game.

Changes to lower the difficulty have now been prioritised as the next biggest changes that need to be made, and are planned for future versions.

Download stats

Downloads vs marketing

Downloads vs players

Future Development

  • Rebuild all levels, lowering the difficulty needed to pass, whilst trying to maintain the difficulty required to obtain high scores.
  • More Characters & Level themes
  • Unique mechanic changes into each level theme (e.g. double jump, slow down/speed up items)

Player Analysis

The data supporting this article has been collected using my own analytics tool Player Insights and manually analysed in MSSQL Management Studio