Google Analytics is the most famous tool to measure how your players behave on your website, mobile app, or video game. Even if you think you have released the best product ever, you still need to know the pros and cons from your users’ point of view. Google Analytics offers an enormous amount of metrics that can be applied almost anywhere. In this article, I’d like to focus on some indicators that apply to the video games world so beginners can take advantage of this technology without going crazy deep into its complexity.
Getting Started
To start using Google Analytics, you need to follow three basic steps:
- Sign up in Google Analytics with any Google account.
- Register your video game in the Google Analytics console and get a unique tracking identifier.
- Insert the tracking identifier in your video game’s code.
- Configure Google Analytics interface to filter the information you actually want to consult.
This looks simple on paper and I’m sorry to be this pessimistic, but . . . it’s absolutely not! There are different ways of achieving that last step depending on the platform you have developed your video game for.
Google Analytics was first created to analyze web traffic, so HTML5 video games are supported by default by including specific Javascript code. Android and iOS platforms can also include Google Analytics. Keep in mind that in the case of mobile devices, the statistical variation of data may be higher because some people still play with their phones not connected to the network. There is even a plug-in to integrate Analytics in Unity that is worth a try even if your game is still in beta.
What to Track?
Beginners tend to think the more events they track the better, but that will end up with an inoperable set of information. The most difficult part in Analytics is deciding and planning what is useful to track and what is not. Best practices recommend not to capture events that are too repetitive. For example, every time a player presses the HUD. In fact, the Analytics engine restricts the tracking to one event per second which should be more than enough. Once you have decided what important events take place inside your video game, list them and choose their names wisely so the interface configuration process is simpler afterwards.
Behavior Flow
Some information comes by default just by integrating Google Analytics in your game’s code. The screen flow is a visual representation of the path of screens your users follows and is something very useful available without any extra piece of code. However, if you want to improve your reports’ readability try to manually set a name in each screen. Using accurate names is also important if your video game has been implemented in different platforms and you want to see the most visited flow for all of them at glance. Follow this guide to name your screens properly.
You can access the Behavior Flow section by pressing Behavior > Behavior Flow in the left panel. You will get used to the fact that almost any information you see in Google’s interface can be filtered, and this view is no exception. Play around with segments, Operating Systems, Versions, etc. to understand all the possibilities you have in front of you. By default, you will only see the first four screens of the users’ path, but you can extend it easily.
Knowing the most visited screens can help you decide where to put the most important information or where to place the hated advertisements without bothering the player too much. For example, if you have two ways of accessing the main video game screen (where you actually play), you can focus your future work on the most popular one.
Know Your Players
Knowing your target audience can help you decide what features include in your game’s road map and which ones you should discard. For example, if you detect your avatar-based video game is played mostly by young people, it’s a good idea to include avatar customization with clothes young people usually wear.
The logic can also work the other way around. Every time you include a new feature, you can check in further releases if users find it really useful and thus keep improving in that direction. For example, if you support Italian in your game, you can easily find out if that attracts and retains Italian players or if they are outside your market.
Subsections under the Audience section can give you this kind of information about your users:
- Location
- Language
- Version
- Active users in a particular moment
- Mobile devices /Browsers/Operating systems
Level Complexity
Most of the time, especially in small developments, programmers become too proficient in the game they are building and make the mistake of thinking a particular level is easier than it actually is. That’s why it’s so important to measure which stages players have more problems with. To achieve that, you can measure the number of hits of the following events:
- Starting a level.
- Dying in a level.
- Finishing a level successfully.
Events are any action that can take place in your game, and they have four levels of granularity:
- Category
- Action
- Label (Optional)
- Value (Optional)
To start tracking this type of information, you can use the following snippet once you have defined your App Tracker:
// Get tracker. Tracker t = ((AnalyticsSampleApp) getActivity().getApplication()).getTracker(TrackerName.APP_TRACKER); // Place the code where level is started t.send(new HitBuilders.EventBuilder() .setCategory("LevelStarted") .setAction(levelID) .build());// Build and send an Event. t.send(new HitBuilders.EventBuilder() .setCategory("LevelFailed") .setAction(levelID) .build());// Build and send an Event. t.send(new HitBuilders.EventBuilder() .setCategory("LevelFinished") .setAction(levelID) .build());
Then, you need to go to Behavior > Events > Top Events in Analytics’s console and filter by Event Category to see “LevelStarted,” “LevelFailed,” and “LevelFinished” events. You will need to review the difficulty of those levels where “LevelStarted” is similar to “LevelFailed” and “LevelFinished” is too low. The time graph will help you see if the fix has really made your players’ lives easier after you launch a new version.
Retention
Statistics say only 35% of players play a free-to-play game a second time. That’s why retention is so important to detect as soon as possible. You need to make a long-term relationship with your players if you want to get a successful game.
Analytics tracks retention in Audience > Cohort, where one of the metrics is indeed User Retention.
What information does this graph give? In the first column, it displays the number of players that started playing your video game that particular day (you can change the visualization to weeks or months) The following columns display the percentage of users that played every day, week, or month after the installation.
They say it’s easier to find new customer than to keep them, and in free-to-play games, the same principle applies. Do your best to give your players a good reason to come back and keep enjoying your game.
Crashes and Exceptions
When you publish your video game to the world, you are not only making public your work and effort, but also a set of bugs that will keep popping up in every player’s device. Even if you think you and your team are the perfect developers, I warn you: There will be bugs. That’s why you should use mitigation tools to detect the exceptions and crashes you will surely have as soon as possible. Crashlytics is a very good tool for that, but if you just want to use Google Analytics, it can give you a lot of information as well.
To access this view, you have to click on Behavior > Crashes and Exceptions. You will see the number of crashes your video game has experienced over time and in every version. If you click on an item from the versions list you can see the class where it happened and how many times it took place. Every time you launch a new version, Google Analytics will automatically add a new row and you will feel how rewarding is to see how the number of crashes decreases every time.
Just the Tip of the Analytics-berg
This article is really just a beginner’s guide to all the possibilities Google Analytics offers, but you can get a lot of benefit just applying some of the previous tips. You may get lost at first with the interface, so try to play around and keep experimenting to get the best of it.
Knowing your audience is the most important part of any product and it will help you keep building your masterpiece in the right direction . . . and in the meantime attract more and more happy players!