As part of the project requirements, and as a way to improve our game’s design we will be implementing unity analytics. Analytics will provide us valuable data and insight from a play-tester’s session allowing us to identify any problem areas.

For our project, we will be using analytics to check:

  • Which objects players interact with the most – This will allow us to see if they are spamming a particular object in particular (which may require us to ‘disable’ it after it has been activated once) or if players miss an interactive object within our scene which would require us to find a way to make it more visible or appealing to the player
  • How long they take to read a letter or letters – If they take too long then the letters would need to be edited to have a shorter length, enabling us to keep the player’s attention as some players would lose interest in having to read large chunks of text
  • Total playing time – We would like to know how long the players are playing the game for, as ideally we would like them to play the game’s entire length to get the full experience. If they are playing it only mid-way, then we know that the game may not be appealing enough to the player if they are losing interest
  • How long they spend playing each snapshot – Ideally, I would like to know how long it takes for a player to play each ‘snapshot’ or levels, if players are taking a longer time in certain levels then it would suggest they either are taking a long time reading the letters, searching for interactive objects or simply just exploring the environment

To display the stats of the information gathered will be shown in the form of graphs. For the first bullet point, I can see it being presented visually as a pie-chart, showing which object had the most interactions and which had the least. The second could be a bar or pie chart to show the length of time taken to read letters and so forth.

Besides gathering data from the analytics, we also plan to observe the players during the play-testing. This will allow us to see in-person any areas that the players struggle with such as finding an interact-able object for example. Although we would need to decide how to design or ‘structure’ our tests and find the one that work best for us.

We could do it as a one-on-one test, where we watch a single person test our build while we observe, taking notes in how they interact with the environment and asking them questions during or after they finished the game. There is also group testing, similar to the one-on-one test but with a group of people instead.

Providing feedback forms is a possibility as well, although this would take in the form of a questionnaire. Alternatively, we can conduct an ‘interview’ where we ask them a series of questions after they play a game, a verbal form of the questionnaire if you will.

Lastly, we go with an open discussion which can be done with individuals or as a group, where everyone discuss after the play-test session. By having an open discussion, we can hear different people’s opinions of the game from their perspective in a more lax or structured situation. Not only would we hear what we did well versus areas of improvement, it would allow us to receive suggestions that may or may not improve our game, or at least help it guide us in the right way to achieve what we want to achieve with our project. There is also the option of combining some of these options, such as having an open group discussion then asking them to fill in an online questionnaire individually.


Methods of Playtesting (source)

Playtesting – Game User Research Methods (source)