This post is also available in: Deutsch (German)
The world of gaming has grown from a simple virtual tennis game in 1958 to offering an overwhelmingly large number of genres and sub-genres of video games that captivate a diverse community of gamers. We have reached the point where even Big Data has arrived in the gaming scene.
Players now have infinite possibilities of things that they wouldn’t do in real life at their fingertips. Like winning their opponent’s Gallardo in an illegal pink slip race or building an empire or fighting intergalactic wars – wherever the imagination of game development companies takes them.
This $159.3 billion industry provides several opportunities for Big Data in gaming to bring it to the next level. Particularly, immense competition makes it necessary for developers to introduce the newest Big Data Science technology in gaming. In this article, we’re looking at use cases of Big Data in the gaming industry.
Leveraging user intelligence to improve game design
Similar to how retailers can collect customer data to personalize their products and services to the customers’ needs, Big Data Science in gaming can help game development companies find out their players’ demographics, their likes and dislikes, how long they play, the time of day they play, who they play with, which parts of the games are too difficult for them and other usable metrics.
From this, Big Data-generated user profiles are created to evaluate the types of players they attract, and what would cause players to stop playing the game, later informing decisions on how to make the playing experience better for them.
Big Data in gaming vs Level Difficulty
For example, if most players find that a level in a game is so difficult that they are tempted to throw their controllers at the screens, the data should show that there’s a bottleneck in winning this level as well as a higher number of players leaving the game.
Game development companies should then adjust the factors of difficulty for player satisfaction.
Ultimately, Big Data in gaming can help determine the best tradeoffs for player satisfaction.
Activision uses Big Data to stop cheaters
Activision, the main developer of the first-person shooter video game series Call of Duty (COD), is one company that uses big data to improve their games.
The Game Science Division (GSD), in charge of collecting and analyzing Big Gaming Data at Activision, had to deal with the issue of boosting among their players in COD.
In gamer speak, boosting refers to the attempt of increasing someone’s gaming scores through unfair methods like teaming with another player who is better than other players and throwing the game on both sides where one side intentionally loses to allow the other side to win.
It sounds like cheating if you think about it and boosting has its consequences. Not only is a boosting player getting prestige that wasn’t earned, but the boosting tactic can also affect the ratings of other players and the balance of reward systems.
That’s why boosting is highly frowned upon by most players and needs to be dealt with before it becomes game over for innocent players.
Activision’s GSD creates Machine Learning-based programmes to detect boosting, identifies key indicators of boosting in COD and tracks those indicators. Two of the indicators are players on friends lists that end up on different teams match after match and tactical spawn points where deaths occur after a very short period of time over and over again.
After analysing the boosting indicator data, GSD then engages with its users on Twitter about boosting and collected data on every COD tweet to aid decision making about the game features and playerbase.
Determining how game development companies can earn money
Another use of Big Data in the gaming industry is monetisation. Game development companies need to earn money for the work they put into making and updating their games.
Especially for mobile games, the two main monetisation methods used are in-app purchases and in-app advertisements. Big Data in gaming optimises both use cases.
The freemium model lets people use an app for free but charges them for additional features a.k.a. in-app purchases.
Since the free-to-play app encourages people to use it straightaway, developers can monitor and analyse and personalise game situations where players would get stuck, what would motivate them to make in-app purchases and where they’re most likely to pay higher prices for in-app purchases.
Developers can then decide which features can be added for a real-life price and adjust the prices accordingly.
Big Data-based optimisation of an interior design game
This reminds me of Design Home, a mobile interior design game I play, consisting of daily design challenges. When a player completes a design challenge, that player’s design will be randomly paired against another player’s design in the same challenge in a voting round.
Voting is done by other players to earn Keys (since a certain number of Keys are needed to enter each design challenge) and determines the Star ratings, out of five Stars, a design receives.
More votes gives higher Star ratings. This system sets the bar for the combinations and colour scheme of the furniture and decorations used in designing a given space.
This desire to get higher Star ratings through nicer designs allows Design Home to earn money through a premium store, which features seasonal matching furniture and decorations (as well as in-game currencies) that players can use to pull off awe-inspiring designs in the game’s design challenges.
Players who pay real money to buy premium furniture, decorations and in-game currencies (Cash and Diamonds) to buy more or higher-priced items will gain a competitive advantage over players who don’t make in-app purchases.
This plays a significant role in the number of votes designs with premium items attract, and ultimately, the Star ratings such designs receive.
Data-driven ads and waiting times
Additionally, to cater to players who don’t want to give their money to mobile games, advertisements can be added into the game. The usage of Big Data in gaming ensures the ads used are related to the players’ user profile.
Like in other areas, personalisation can do wonders here. Similar data gathered to personify players to retain them can also be used to personalise ads, so that players don’t get annoyed when they see ads that might not be relevant to them.
Let’s say I’m playing an interactive mobile game called Episode, where I help the main character of a story choose which course of action to do or sentence to say, which ultimately determines what happens later on in the story. A story would consist of a certain number of episodes and a Pass is required to enter each episode.
For the game’s developer Pocket Gems to earn money by default, they play two 30-second ads every time I enter an episode. And since I get four Passes every few hours, I would finish up the Passes by entering four episodes in a row and have to wait for a few hours to replenish the Passes.
The alternative to waiting is buying Passes with real money, which not everyone wants to do. It’s likely that Pocket Gems detected this as an opportunity to earn using more ads by giving players the option to play an ad to cut the waiting time by half an hour.
Based on the ads I’ve been shown, I can say that their personalisation is done on a segmented level. The ads are usually created by companies or personalities based in my country, so in my case, the game is showing me ads based on my location or nationality.
Developers can take personalised advertising further by incorporating Big Data-powered hyper-personalisation in gaming, where segments of user data are broken down into finer individually-scaled pieces of user data to target individuals rather than segments.
Keeping players hooked
The use cases above show that a game dev company can utilise Big Data Science to determine which features of a game can be improved to make the game more fun and fair.
Big Data Science in gaming can also be used to monetise the game with in-app shopping and determine how much to charge for premium items, how much in-app advertising to use as well as which in-app ads to use to target a certain user persona.
It’s all about attempting to read the minds of players who might be looking for avenues to escape the stressful reality, to host a virtual games night with their friends or to enhance creativity and problem-solving skills.
Luckily, game developers neither need a psychic nor a time machine to predict what their players want since Big Data Science in gaming can play a big role in this. The huge amounts of data collected about players can then give developers some ideas about what they can do to make games so engaging and addictive that it’s hard to switch off.
All in all, the usage of Big Data in gaming is a must for game developers to stay ahead of their competition. I believe that what we see today is just the beginning of Big Data in gaming and more advanced use cases are ahead of us.
Takeaway
What are the typical use cases of Big Data in the gaming industry?
– Collecting and analysing user data to find features of a game that require improvement.
– Monetising games through in-app purchases of premium items and personalised in-app advertising.
How can hyper-personalisation support games revenues?
– Hyper-personalisation helps games select and show ads that are more likely to be relevant to the user based on the user’s data.
– Making users watch ads is an alternative to in-app purchases since not every user is willing to make purchases with real money in a game app.
Two examples of improving game design.
– Adjustment of difficulty factors
– Tackling loopholes leveraged by cheaters.
What would motivate players to make in-app purchases?
– Bottlenecks that prevent them from advancing to the next round, so they’d want to use a premium item that removes that bottleneck.
– The premium item can be used to give a player a competitive advantage over other players, so that this player can get higher ratings.