Big data analytics for video, mobile, and social game monetization

Understand and influence profitable consumer behavior

The game industry is experiencing tremendous change as the traditional gaming landscape expands to include new types of games, platforms, and players. Game developers and brands have an opportunity to apply these big data analytics techniques to capture rich and varied behavioral and multi-structured game and player data. You can store this data in noSQL databases and integrate it with relational transactional databases to gain keen competitive advantages through deeper and more actionable insights.

Share:

Kimberly Chulis, CEO and Co-founder, Core Analytics, LLC

Photograph of Kimberly ChulisKimberly Chulis is one of the original founders of Core Analytics, LLC. With over 18 years of professional advanced analytics experience, she's demonstrated analytic expertise on projects at several companies and industries, including WellPoint, HCSC, UHG, Great West, Accenture, Ogilvy, Microsoft, Sprint/Nextel, Commonwealth Edison, TXU, Eloyalty, SPSS, Allstate, Cendant, and others in the financial, telecommunications, healthcare, energy, nonprofit, retail, and educational sectors. Kimberly has conducted PhD research at Purdue University's Health and Human Services Consumer Behavior program, and has a Masters degree in economics with a focus on health economics and econometrics from the University of Illinois at Chicago. You can reach Kimberly at kim@coreanalytics.com



August 2012 (First published 17 July 2012)

Also available in Chinese Russian Portuguese Spanish

There is a revolution happening in the game industry. The sheer size of the global game industry is staggering. According to industry analyst Colin Sebastian for RW Baird, video games generated US$60 billion in revenues in 2011 and are expected to reach US$80 billion by 2014. While exact projections and expected growth patterns by game type diverge depending on sources, some herald the video game industry as the anticipated mass media channel of the decade (see Resources). Computer games are forecasted at US$20 billion, and social games are anticipated to bring in US$2 billion this year (2012). While these numbers indicate that the current largest share of market belongs to the video game industry, an important preferential shift and player demographic change is occurring that is expected to result in a partial decline in popularity of core video games and a rapid decline in the console variety.

Social games are now offered at a fraction of the price through social platforms, such as Facebook, and are played on various mobile platforms, such as Android and iOS. These games differ from traditional video and massively multiplayer online (MMO) games. Traditional hardcore gamers are predominantly male, between the ages of 18 and 34, who buy boxed console and computer games that they pay for with cash and credit cards and play solo or within a limited interactive environment. The rapid adoption of mobile games represents a fundamental change in the gaming landscape, expanding who, how, and why gamers are playing. Social gamers tend to be 54 percent female, use mobile payments and PayPal for purchases, and play across platforms and devices (see Resources). Of further interest are stark differences that are emerging in game revenue patterns across devices. Recent surveys conducted by Newzoo point to iOS games generating 85 percent of in-game revenues over Android and other platforms.

Industry analysis points to an expected decline in console games and related hardware and apparatus sales, with web-based games becoming the go-to gaming platform. This trend will be parallel to continued growth in video games played on the computer and to explosive growth in mobile games, with the latter expected to hit US$16 billion in revenues by 2016, according to ABI Research. This translates into big opportunity for analytics providers.


Game analytics

Although the global game industry is now larger than the music industry and on par with the size of the film industry, game developers are only starting to adopt advanced analytics to support game development, product design, targeted marketing efforts, and data-driven in-game monetization optimization. The corresponding game analytics industry is still greatly underserved. A list of niche analytics vendors for social and mobile games continues to expand, with representation by Kontagent, Flurry, Mixpanel, Totango, Claritics, and Google Analytics. There are far fewer vendors focusing on the computer and MMO games, and no single analytics provider appears to focus on delivering cross-game platform analytics.

Many of these off-the-shelf products adequately provide the expected standard list of metrics for social games (see Resources), including:

  • Daily active users (DAU)
  • Monthly active users (MAU)
  • A combined DAU/MAU ratio
  • Engagement, which measures time spent playing a game
  • K-factor, which is an infection rate of viral game growth as the core and casual player base expands
  • Average revenue per user (ARPU)
  • Lifetime value (LTV), which captures a player's value to the game based on in-game purchases and other monetization-related behaviors, player influence on virality, and net game evangelism

Game monetization

Traditionally, video and social games have had distinctly different business models. Now there is more of a convergence of business monetization approaches between game types. Historically, video games relied on a subscription-based model requiring gamers to make an upfront game purchase and subscribe to an ongoing monthly fee. This business model is being replaced by some games with the testing of free-to-play models that closely follows the emerging freemium model typically offered by social games (see Resources). Currently the terms mobile and social are often used interchangeably; however, they will likely become distinct game genres as game distribution patterns develop. They will expand game access and greatly increase the corresponding potential player universe. Whereas some games today are available only on certain sites such as Facebook, on single platforms such as computer or Xbox, or on specific devices such as iPhone or iPad, eventually many game titles will be released across more platforms in a more site- and device-agnostic fashion.

Social and mobile games make money (monetize) in a few ways, and related business and monetization models are changing dynamically. Some games, such as the iOS-based W.E.L.D.E.R. word game, charge an initial purchase subscription fee but will likely eventually move to a freemium model. Other games, such as those on Facebook, rely on the sale of virtual goods. Games such as Zynga’s CityVille, for example, fit the mold where players use actual money or Facebook Credits to purchase in-game virtual goods to help them perform better, gain advantages, get premium access, and move to higher levels (level-up) faster. A third popular form of game monetization is in-game advertising. Companies such as TapJoy offer in-game banner ads, video offers, and full-page interstitial advertisements. There is a growing list of companies that offer hybrid models aimed at both raising awareness of new games and monetization optimization. As games and associated monetization models evolve, the relevance of underlying analytics to identify segments of players, how they play, propensity to click on an in-game ad, or pay cash for virtual goods becomes exponentially greater. Current providers offer standard metrics, and some of these vendors, such as Flurry and Kiip, offer virtual good or ad-recommendation optimization engines. While the game analytics industry is making big strides towards catching up, it is still arguably in its infancy. Stores of virtually untapped social data reserves at the player ID level exist as potential predictors that can be tied to segmentation and propensity to purchase models that can drive value-based game development, localization, targeted in-game offers, and ads. This social data can also measure the success of net promoter scores, identify player evangelists, and track redemption of in-game offers.

In addition to the mainstream game monetization approaches detailed above, another major marketing trend is emerging that is worth mention, and is also relevant from an advanced game analytics focus. Gamification is one of the hottest enterprise trends for 2012, involving the use of game mechanics and design to motivate people and drive specific behaviors (see Resources). In a consumer behavior sense, gamification involves the introduction of game elements such as leaderboards, badges, trophies, points, virtual currency and credits, and more to reward desired behaviors. In one such example, Bunchball and IBM® have teamed up to introduce gamification to drive user adoption and engagement in the IBM Connections environment (see Resources). Healthcare companies such as UnitedHealth Group have integrated gamification and video game strategies to promote the self-management of wellness and healthy behaviors. Gamification is forecasted to receive US$2 billion in direct spending by 2016. As these programs gain popularity and are more widely implemented, game analytics solutions to measure impacts of gamification programs on customer engagement and loyalty will be in demand.

With all of the expected growth around video, social, and mobile games and the increasing enterprise adoption of gamification, one might expect the associated game and gamification analytics solutions to be more mature. A large barrier exists to nimble game-and player-related analytics that has hampered rapid advancements in this space. The sheer mass of in-play, leveling up, skill achievement, in-game purchase, and peripheral game data at the individual player level presents a challenge for traditional database structures. Legacy relational database management systems (RDBMS) were not built to manage, store, and process the petabytes of data generated by modern MMO and mobile or social games. New big data solutions are based on NoSQL technology (see Resources) and are far better suited to manage the rapidly changing data volume, sources, and format of structured, semi-structured, and unstructured data, and filter datasets to a manageable level as an input to on-the-fly analytics solutions.


Technology advancements in databases

Relational databases have been in widespread usage since their introduction onto the scene in the 1970s. While there has been explosive growth in user bases of online applications and resulting data generated from online and mobile systems during this time, new solutions better suited to managing data on such a big scale were not introduced and in wider-spread usage until recently. Different methods of extending the capacity of legacy systems were introduced.

  • Sharding is the practice of data partitioning across diverse servers, which requires knowledge of the server location of data and is limited by the fact that you can't perform joins across shards. You must maintain schemas for each server.
  • Denormalization is another method that involves grouping and indexing redundant data and often results in latency and issues with maintaining concurrency in relational database systems.
  • Distributed caching, which caches recent data in memory, is useful when data is needed. The application (web, game, social network, search engine, and so on) first checks a distributed caching system, such as memcached, for the needed data instead of going back to the relational database.

New NoSQL technologies avoid the common shortcomings of relational databases and the resulting need to employ the scope-extending methods described in the previous paragraph. They don't require a schema or joins, and are not relational. These databases can handle structured, semi-structured, and unstructured data. Additional data can be added to the stores at any time, irrespective of format, and be immediately available for analysis. These databases can easily handle hierarchical nested data structures and are elastic, automatically spreading data across servers as data expands and contracts, without impacting performance (see Resources).

There are a variety of NoSQL database types, including document stores, column stores, key-value stores, XML databases, and graph databases. There are over 122 NoSQL databases in use, and that number continues to expand. There are several popular open source NoSQL options including column stores, such as Apache CouchDB and MongoDB, and wide-column stores, such as Apache Cassandra, Apache Hadoop, and Hbase. There are also open source analytics tools that sit on top of systems such as Hadoop’s MapReduce paradigm. Apache Mahout is a machine learning engine that provides classification, clustering, and collaborative filtering. Open source R has been integrated to run massively parallel statistical processes directly in Hadoop nodes. Also popular are commercial NoSQL options that integrate with Hadoop and other open source tools and greatly extend these capabilities with analytics, text mining, in-application processing, map reduce functions, and graphing options. One such example is the recently released IBM InfoSphere BigInsights (see Resources) that is built on the Apache Hadoop platform and is available in both a basic no-charge option and a more powerful enterprise edition.


Video game analytics examples

Now it's time to take a look at a couple of real-world game analytics applications. Imagine a large MMO video game. This game genre includes powerhouse games such as World of Warcraft, the most profitable video game in history, and the recently released Star Wars the Old Republic. The video game subscription models fall into a few categories:

  • Pay-to-play is where players must pay a monthly subscription fee.
  • Free-to-play usually involves an upfront software cost but no additional payments.
  • Freemium allows players to access game content and play for free but offers options to pay for additional content and access.

The largest game titles in this genre use pay-to-play subscription models. Game studios are now facing stiff competition as a direct result of the growth of mobile and social games. So while the larger titles are following the subscription model now, there may be an eventual shift towards a freemium model.

For pay-to-play games, game analytics are focused on understanding who the most valuable players are, how they play, if they evangelize the game and influence others to play, and what their player personalities and motives look like. An important application of propensity modeling in this type of model is to identify those players with the highest propensity to do one of the following:

  • Continue a subscription
  • Return to play a game after a subscription pause
  • Encourage new players to subscribe
  • Become skilled and persuasive guild leaders

Data elements include traditional game-time dashboard key performance indicators (KPIs), time to complete levels, solo versus interactive behaviors, avatar selection, interaction style indicators, gender of avatar, game strategy behavior variables, game-related tweets, social network activity, language, and more.

Microsegmentation applications involve segmenting a player base to understand distinct segment preferences and behaviors to guide targeted game design, localization that reflects preferences of regional segments, and appealing targeted extension packages and additional content design (see Resources). When you segment a player base and assign players real-time propensity scores, product, design, and marketing have detailed individual player intelligence to guide strategy and back-end measurement. This approach to players of games is no different from the traditional customer view towards applying advanced analytics for player retention, churn, and marketing response efforts. The main difference in this channel is the new variety of data and tremendous volume and velocity at which it is generated. The video game industry stands to benefit as they begin to leverage the efficiencies of combining traditional relationship databases such as Netezza with Hadoop and other NoSQL data stores and apply data mining tools such as R or InfoSphere BigInsights, which effectively manage out-of-memory data processing and analytics.


Mobile and social game analytics examples

Mobile and social game developers have been early adopters of big data technology, cloud computing solutions, and associated data mining applications. Zynga, for example, is well known in the industry for strategy based on cutting-edge player analytics to drive actionable user analysis. Player analytics allows social game studios to understand in real-time why users are abandoning a game and identify other players at risk of leaving the game so they can develop player retention strategies before those players quit. The same analytics applications optimize ad-generated interaction and in-game virtual goods sales. If a game developer is able to identify viral users that spread the player base, they can perform an outreach on social media sites with rewards to ensure the continuation of desired behaviors. Analytics in a mobile setting identifies players who represent the highest value in terms of propensity to purchase virtual goods, evangelize a game, or generate ad revenue. Another major issue with all game genres is game fraud, and analytics uncovers fraudulent player behavior for removal. Mobile and social games employ analytics to understand content and campaigns that work best, often combined with embedded A/B content testing and refinement. Player segmentation based on variables related to device, platform, carrier, geolocation, and demographics facilitate more effective in-game and non-game brand and real-time partner-targeted offers.


Summary

Both the game industry and technology are evolving at a rapid pace. Technological advancements in both sectors provide exciting opportunities for game developers and media studios to apply advanced analytics to further enhance design and optimize game monetization efforts. If new state legislation passes to legalize online gambling, a new focus on advanced analytics and microsegmentation to drive player monetization based on gambling patterns will emerge. Game fraud and analytics to combat this across all game genres will continue to be stressed. As more and more users gravitate to smart phones and tablets, more game business models will move to freemium models, and new player monetization models will emerge that require propensity modeling and segmentation for better targeting. Gamification will continue to spread, and the need for metrics associated with these programs will grow. The next decade promises to be fertile ground for specialized game analytics solutions and analysts focused on incorporating big data technologies to uncover rich insights into both player and consumer behavior.

Resources

Learn

Get products and technologies

  • InfoSphere BigInsights is the starting point for learning and working with big data. Download the Basic Edition at no charge.
  • Evaluation software: Find more trial software. Download a trial version, work with product in an online, sandbox environment, or access it in the cloud.

Discuss

Comments

developerWorks: Sign in

Required fields are indicated with an asterisk (*).


Need an IBM ID?
Forgot your IBM ID?


Forgot your password?
Change your password

By clicking Submit, you agree to the developerWorks terms of use.

 


The first time you sign into developerWorks, a profile is created for you. Information in your profile (your name, country/region, and company name) is displayed to the public and will accompany any content you post, unless you opt to hide your company name. You may update your IBM account at any time.

All information submitted is secure.

Choose your display name



The first time you sign in to developerWorks, a profile is created for you, so you need to choose a display name. Your display name accompanies the content you post on developerWorks.

Please choose a display name between 3-31 characters. Your display name must be unique in the developerWorks community and should not be your email address for privacy reasons.

Required fields are indicated with an asterisk (*).

(Must be between 3 – 31 characters.)

By clicking Submit, you agree to the developerWorks terms of use.

 


All information submitted is secure.

Dig deeper into Big data and analytics on developerWorks


static.content.url=http://www.ibm.com/developerworks/js/artrating/
SITE_ID=1
Zone=Big data and analytics, Information Management, Mobile development
ArticleID=825736
ArticleTitle=Big data analytics for video, mobile, and social game monetization
publish-date=08172012