This game design topic is usually completely invisible to players, yet it often has the largest impact on updates applied to a game after release. Once a game is released, subsequent updates often tackle balancing issues. Balancing without metrics is based only on intuition and best guesses. Metrics are collected from your players and typically stored in a database. If you don’t wish to set up your own server and database, there are many services available, and many game engines come with metric tracking built-in. With good metrics, you will be able to make informed decisions for your updates and learn what is effective and what doesn’t work for designing your next game.
Metrics are simply the tracking of statistical information about your game. What are good metrics? This is a questions that can be difficult to answer as they vary with every game. After you release your game, you will realize that there are countless metrics that you wish you were tracking. Tracking playtime spent in different sections of the game is one of the most useful metrics. This includes splitting up time spent in menus, time spent talking to npcs, time spent active, time spent traveling, etc. The more precise you can get your metrics, the more information you can glean from them later. Tracking game crashes, player deaths, skipped dialogue, and locations when the player quits the game are all good metrics to keep track of as well. Collecting a rich set of metrics will provide a source for many insights.
Finding balancing issues and game breaking bugs can be made infinitely easier with access to metrics. Multiplayer games are often the most difficult to balance, you have to make sure that nothing is neither too overpowered that it becomes the only viable choice, or too underpowered that it isn’t viable in any situation. Knowing the percentages for what loadouts players are using, and correlating them with wins and losses, you can often easily find problematic areas to address. Metrics can also be useful for pinpointing game crashes. For example, knowing the location and states of the game leading up to a crash can provide insight into which systems are causing the problems. Another use for metrics is finding areas that are not intuitive. It’s common to see players skip dialogue, then spend long time spent in an area, and eventually quitting the game afterwards. These areas can often be improved by adding visual cues and other hints to improve the flow of your game. Applying what you glean from the collected metrics will allow you to make better design decisions in the future.
Making games is a constant learning process and metrics can provide a wealth of knowledge. You can often find out when experimental concepts paid off by tracking player engagement when they interact with them. For example, you could add in an optional mini-game testing a new mechanic and track how many players played it more than one time. However, you should beware of making all judgements on metrics alone as they can be a double-edged sword, as not everything’s value can be conveyed by in-game statistics alone. Some publishers tell developers not to invest time into hidden areas in a game that will only be seen by a small percentage of players as it’s a bad return on investment. If you look at a game like Dark Souls that has a large zone hidden behind a painting and another hidden by two illusionary walls, you can see that the draw of finding such things can be extremely rewarding to the select players. Those players might have an impact on sales based on their reviews, videos, and social media that can’t be tracked with in-game metrics. Conversely, if you are developing a casual mobile game and devote a majority of your time creating hidden content that is only ever unlocked by 0.01% of your audience, then you know that your time might have been better spent polishing other areas of the game. Take what knowledge you can find from your metrics and weigh them against your intuition on what will make a better game on your next endeavour.
Future design decisions and balancing choices will be made easier with adequate metric tracking. While not all sources of knowledge should be blindly trusted, metrics can help guide you in your development adventures. At the very least, you will know that your choices will be based on more than just a best guess. Improving game balance, making smart design decisions, and fixing game breaking bugs will make your players happy, even if they don’t know how you came to your decisions.