Scott Kim
www.scottkim.com
Games on the internet are not just for dedicated gaming sites. Businesses that have nothing to do with gaming, like Yahoo!, Sony and Pepsi, are adding familiar games like hearts and crossword puzzles to their web sites as a way to keep customers coming back. These simple quick games, played in a web browser, are aimed at a general audience, not at the dedicated gamers traditionally served by computer game publishers.
Successful web games need to be tuned to the needs of the companies they serve. This paper shows you how to analyze the business needs for a web game, what sorts of games work best for which purposes, and how to fine tune a game for different situations. This paper is addressed both to people at web sites that need games, and to game developers who want to build web games. The examples are based on my experiences as an independent game designer developing games for the web.
Web + Games = Opportunity
For game developers, the web is an opportunity to expand their audience to include the vast world of casual gamers. The rise of casual gaming is fueled by the spread of cheap computers and the internet into mainstream society. Computers are no longer just for young males; the demographics of WebTV owners, for instance, match those of the United States at large.
Casual gamers want much simpler games than the epic battles favored by traditional gamers. Many of the core features of traditional games, like realistic 3-d graphics, are irrelevant in casual games. For instance, revivals of classic arcade games from the 80s have been a runaway success on Shockwave.com.
|
|
Traditional Gamers are |
Casual Gamers are |
|
Audience |
Technologically savvy young males |
Nontechnical males and females of all ages |
|
Technology |
With the latest high-end computers |
With older low-end computers |
|
Price |
Who are willing to pay for |
What want free |
|
Games |
Violent action games |
Familiar nonviolent turn-based games |
|
Complexity |
That have depth and complexity |
That are convenient and easy to learn |
|
Play time |
Played over many long sessions |
Played intermittently in short sessions |
The shift toward causal games is causing major upheavals in the game industry. For instance, Total Entertainment Network (TEN) started off as a high-performance network for multiplayer 3-d games, but quickly morphed into the classic gaming site Pogo.com when the original business model didn’t pan out. Traditional game companies like Electronic Arts are struggling to make the transition, while small new companies are springing up everywhere.
For businesses with web sites, games are an opportunity to increase customer loyalty. Using games to improve business is nothing new. Crossword puzzles in newspapers increase reader loyalty by providing a dependable daily amusement. Contests under bottle caps boost sales by promising a chance to win prizes. Trivia quizzes in movie theaters keep customers entertained while waiting for a movie.
On the web, games are actively engaging experiences, increasing their ability to hold people’s attention. But be aware that games are only one of many ways to make a web site stickier. Before you decide to put a game on your web site, consider the alternatives.
|
Advantage of games |
Disadvantages |
Alternatives |
|
Single player games are easy to support |
But they aren’t viral, and don’t persist beyond the play period |
Greeting cards, screen saver, desktop, email games |
|
Single player games avoid problems with troublemakers |
But they lack social interaction, meeting other people |
Message boards, chat, email, instant messaging |
|
Multi player games build community |
But they are harder to build and support |
Asynchronous multiplayer interaction, e.g. high score boards |
|
Intellectually challenging |
May be intimidating |
Jokes, quotes, noninteractive fun |
|
Controlled experience |
Uncreative |
Open-ended software toys and tools |
|
Compelling, engrossing |
Distracts from rest of site |
Syndicated content relevant to site |
|
Fanciful, escapist |
Too frivolous for serious sites |
Timely relevant information |
|
Small size, big effect |
Requires programming |
Noninteractive puzzles, just HTML |
|
Cheap to build |
Expensive to maintain |
Automate or eliminate maintenance |
|
The possibility of winning a prize in a contest is a great way to attract attention |
But lotteries are illegal many places, and skill-based contests invite cheating |
Creative contests judged by humans, e.g. write a jingle, which cannot be faked |
Analyzing Your Needs
Suppose you are a business that wants to put a game on your web site. What sort of game should you choose? The first step is to be clear about your business goals and practical constraints. Here are eight questions to ask. Some may seem obvious, but all too often companies put games on their site without being clear about why they are doing it, or whether they can actually support the game.
1. Purpose. How does the game serve the goals of your business? The purpose of a game can be as vague as to keep visitors amused, or as specific as selling a particular product. In either case a game should serve your business goals.
2. Bottom line. What is the business model of your site? How does the game feed the bottom line? Sometimes the answer to this question will be the same as for the previous question, but people who control web site budgets often have their eyes on different goals.
3. Audience. Who is your audience? What types of games do they like? Most people strongly prefer either word, logic or visual puzzles. Older people shy away from action games. Do they speak different languages? How do games fit into their lives? Characterize four typical players: who they are, why they play, and a detailed scenario chronicling their play experience.
4. Judging success. How will you judge the success of the game? How will you collect the relevant information? Web games can be built to automatically collect detailed information about players, but only if these mechanisms are specified up front. To track player-specific information over time you need players to register.
Incidentally, the best time to ask a player to register is at the moment they want to submit a high score. This delays the tedious registration process to the last possible moment, lowering the barrier to entry. When players are focused on winning a game, they are unusually willing to divulge their email address and other personal information.
5. Technology. What is the intended platform? What is lowest end browser you support? Are plug-ins required? What is the game built in? Common answers: Java, Shockwave, Flash, or JavaScript. What is the minimum screen size? What connection speed do you assume? How large can the download be? What server side technology do you have?
6. Staff. Who will you need to build and maintain the game? Will you produce the game In-house, hire outside developers, or use a mixture of both? Will you build or license? If you license a game, do you want to customize it some way?
7. Budget. How much money are you willing to spend over what period of time? What are your timing constraints? As with all software, allow plenty of time for design iteration and testing. As with all web content, allow time to integrate the game into your site.
8. Production pipeline. How will you maintain the game over time? Will it require ongoing support staff? Is defending against cheating an issue? If the game requires periodic updates, be sure to plan the production pipeline. TV and newspaper companies tend to be comfortable producing daily and weekly content, whereas software companies accustomed to releasing products are often surprised by the need for ongoing support.
Choosing a Game
Here are the most common reasons for adding games to a web site, and the sorts of games that support each reason.
Attract interest. Contests are the big draw here. For instance, the portal iWin is pulling within striking distance of Yahoo! through the simple gimmick of giving away daily cash prizes. The key is to choose a contest that will appeal to your audience. For instance, 3play on ESPN.com is a lottery based on randomly chosen sports figure cards. But beware: contests have legal issues, and are vulnerable to hacking. Gary Rosenzweig, head of Clevermedia.com, often talks clients out of contests by warning them that they will spend most of their time defending against hackers. Top three defenses against game hacking: host the game on the server side to avoid client-side hacking, keep prizes small, and inject randomness into gameplay.
Draw repeat customers. Television, newspapers and magazines encourage loyalty by delivering content that changes every day, week, or month. Syndicated periodic content on the web, available from companies like iSyndicate.com and uclick.com, does the same thing. Ways to increase the incentive to keep coming back include: preview of the next installment, answers posted next time, high score boards, skill ratings, continuing story, and cumulative redeemable credits for playing often. Games that work best have a fixed form with changing content, e.g. crossword puzzles and jigsaw puzzles. Fixed form makes the game both easier for players to understand, and easier for developers to build.
Brand marketing. The goal here is to enhance the image of a brand or company, not to directly sell a product. One way to market a brand is to choose a game that can use your product or logo as a playing piece. For instance, Nabisco’s Candystand.com has Shockwave games that use pieces of candy as playing pieces. Another way to market a brand is to design a game that teaches a concept. Finally sites like Pepsi.com place games on their site that contribute to an overall lifestyle image.
Product marketing. The goal here is to sell a product. If the product is itself a game — computer or otherwise — the web game can give you a sample of the gameplay experience. Many games translate poorly when compressed into a tiny web game; in these cases a movie may be better. Games can explain complicated product features. For instance, many automobile web sites effectively turn choosing a car into a game by making it fun to try out different combinations of features. Games can also give you a surrogate experience of using a product. For instance the jet ski maker Seadoo has a jetski driving game.
Viral marketing. Games can aid word of mouth by encouraging customers to spread the word to their friends. Many games include a button to send an email to a friend telling them about the game. Online greeting cards can incorporate games. Games played by email are naturally viral. You can drive people from email to the web by putting answer on a site.
Extend a brand into a different medium. Sony created online versions of its popular TV game shows Jeopardy! and Wheel of Fortune for its web site www.station.sony.com in order to reach younger viewers who prefer the web over TV. The issue here is to translate the essence of the show while fudging the details based on the limitation of the medium. For instance, on TV Jeopardy! contestants respond by talking; on the web they respond by choosing from among four canned responses. Brand extensions can also create connections across media. MTV’s Web Riot, using technology from spiderdance.com, was the first game show to include web play as an integral part of the TV show.
Distraction. Games are good way to keep customers entertained while they are waiting for something to happen. For instance, MondoMedia.com uses small light games like the memory game Concentration to prevent viewers from giving up while waiting for weekly animated Mondo Mini Shows to download. Such download games should be tiny downloads (50K is good), extremely familiar so they require no explanation, and reward the player frequently. The best download games are customized to match the content being downloaded, and allow the player to keep playing even when the download is complete.
Alternate entry point. Games can be a good way to get people to your site that might not otherwise find it. For instance a flower shop could put a flower game on their web site in order to capture people who are looking for games. Games that serve as alternate entry points need to be easy to find, cross-listed on relevant sites, and thematically consistent with the host site.
One stop shopping. Portals like Yahoo! and AOL are like malls: they provide a comprehensive experience designed to keep you in one place for a long time. Rather than let you leave the site when you want to play games, portal sites offer their own game areas, similar to game arcades in malls. Since portals try to appeal broadly, the game areas should include a representative sample of the most common games, e.g. card games, board games, and classic arcade games. For legal reasons some games cannot be copied exactly. Note that you can customize generic games by changing the graphics without changing the rules. For instance, Las Vegas casinos spice up generic gambling games by placing them in exotic themed environments.
Income from the game itself. Some advertising actually pays for itself. For instance, a branded game can also support advertising. Branded games work best when the character or slogan has become a part of popular culture, such as the Nike swoosh or the slogan “got milk?” But beware: advertising models on the web are changing fast, and pay for play is a difficult proposition. The key to pay for play is to offer something you can’t get for free elsewhere, such as cash prizes, persistent worlds, or community. It is also important to keep the billing model simple; pay for play is gaining ground in the cellular phone world, where people are accustomed to paying a monthly bill with many small charges.
Product placement. The flip side of placing a game on your web site is to place your product within in a game. Just as Nike pays athletes to wear their shoes, so many companies pay game publishers to have their products featured in games. Of course the games should be appropriate to the product and feature it in a positive light.
Case Studies
The Adobe Puzzle appeared for a year in 1998-1999 on the Adobe web site. The puzzle rotated among four templates, and featured Adobe fonts and clip art. Adobe users are graphically oriented, so naturally the puzzles were very visual.
In the example above each square design is a close-up excerpt of one of the letters in the Adobe typeface Flood. A font sample is shown at top. To play, you choose a letter from the menu beneath each square, then click “Submit Answer”. The number of guesses and correct letters is reported, and the menu below each correctly guessed letter is removed.
- Although the puzzle used templates, each puzzle still took much hand assembly, making the puzzle an expensive liability. Better tools could have streamlined production.
- Originally the puzzle was to be weekly, but it gradually slipped to monthly. Lesson: launch with minimal features and gradually ramp up.
- The first week each puzzle was posted it was mentioned on the front page of the site. Naturally the puzzle got most of its traffic those weeks. Lesson: get prominently featured.
- The Adobe site used plain HTML, so the puzzle was produced in HTML plus CGI scripts to check answers. Working within the limitations of HTML led me to the new idea of counting the number times you submit answers as your score. Lesson: constraints are your friends.
- Upon winning the puzzle the player was directed to a different web page. This made it easy to track how many people began the puzzle, and how many completed it.
- Many people played the puzzle while waiting for large downloads to complete.
- When the site was revamped to emphasize sales, the puzzle was dropped, even though it had been a popular feature. Lesson: if the purpose of a site changes, the game needs to change with it.
Mentalworkout.com was produced in 2000 by ad agency TMP Worldwide as part of a recruitment campaign for business consulting firm Cambridge Technology Partners. I act as consulting game designer. The campaign includes print ads that drive traffic to the web site.
The challenge in the puzzle above, called “Networking,” is to arrange the twelve tiles so all horizontal and vertical lines connect all the way across the board. In the example above the top two rows and the first and third columns are currently complete.
- Although I built the games in Flash, we chose to make the games a downloadable application, allowing the rich sound and graphics needed for sophisticated visual branding.
- Puzzles play several roles in the campaign: they attract attention in ads, communicate that Cambridge employees enjoy solving problems, provide a fun experience that people share with friends, and get people to give their names when they submit high scores.
- To meet a tight deadline, I reworked existing puzzles with new graphics and themes, rather than invent new ones. Here the theme is networking with other people.
- My main challenge was to add a scoring system to the puzzles so that there would be a wide range of scores on the high score board with few ties. I wanted it to be reasonably easy for players to complete the puzzles, but hard to get high scores, so players would be motivated to replay puzzles to improve their scores. My solution was to score puzzles based on number of moves. To score well on this puzzle, you must first find a solution, then figure out how to reach the solution in the fewest possible moves.
Conclusions
Games are a good way to spice up a web site. But they are not always the best choice. First be clear about your business goals. Then choose a game that helps you reach your goals. Choose a game appropriate to your audience, and don’t be afraid to customize or modify the game to better suit your needs.
To avoid headaches, use the simplest possible technology. Consider outsourcing the design and hosting of the games. If you do build your own games, don’t underestimate the cost of support. Allow time for testing, and launch with a minimal feature set. Be clear about how you will measure success, and build mechanisms for measuring performance into the original game design.
The rise of casual games is reshaping the computer game industry. If you are a traditional game developer wanting to move toward casual gaming, realize that you will have to change not only the types of products you build, but the way you do business. The web changes software from a product business into a service business. Give up bigger is better. Think small and familiar.
In the next few years I expect to see companies that offer turn key services that make it easy to add casual games to a web site. These companies will offer a continuous range of services, from free generic games hosted on another site, to completely custom games integrated into the fabric of your site. The games will scale to work on a range of platforms, from cell phones to high-end workstations.
As the web evolves, business models will keep changing. Old companies will die and new companies will take their place. But one thing will stay the same: successful game developers will build games that make business sense.
References
Gary Rosenzweig. Designing Web-based Games. Proceedings of the 2000 Game Developers Conference.
Gary Rosenzweig. Advanced Lingo for Games. Hayden McNeil Publishing, 2000. ISBN: 078972331X.
Scott Kim. Games for the Rest of Us. Proceedings of the 1998 Game Developers Conference.