The “Gamification” movement continues to gain speed. And while I appreciate some of the careful thinking and discussions, the entire movement is at risk of getting it more wrong than right. Quite frankly, badges and points are, well, missing the point. Gamification should not be based on simplistic, Pavlovian responses to stimulus. Click on a button. Get points. Brush your teeth. Get points. No cavities? Wow. I get a sticker. If that’s the best that game designers can do, the world is going backward, not forward, and I don’t want any part of the Gamification movement.
At The Innovation Games® Company, we think that serious games are more about actual problem solving than racking up points and badges. To realize this vision, effective serious game designers need to create collaborative interaction models that naturally motivate high levels of engagement and participation. The key word is naturally. Not artificially induced and potentially farcical behavior, but games in which the “play” produces a result. The root of this is understanding that serious game design is actually quite a bit different entertainment game design. Not better than entertaining games (which we, of course, need). Different. In this post, I’ll elaborate on these concepts and show how they have informed the design of two of our games: Buy a Feature and our recently released iPad game, Knowsy.
Choice Modeling: A Multi-Billion Dollar Market Research Industry
Before discussing the design of our games, we need to take a step back and understand one of problems we’re solving through our games. Simply put, we’re trying to create a low-cost, high-quality alternative to traditional choice modeling market research.
Choice and Preference modeling is a multi-billion dollar industry, and encompasses such techniques as conjoint analysis, forced rank, and MaxDiff (to name a few). When done well, choice modeling is one of the most powerful forms of market research. Unfortunately, doing traditional choice modeling market research well has significant challenges. It requires specialized knowledge and skill to construct surveys, develop / select the sample, and post-process the results. It typically requires sophisticated tools and the ability to use these tools properly. As such, choice modeling tends to be expensive and take a fair amount of time to complete. For example, the last conjoint analysis study that I helped produce (I helped design the attributes and levels) took about 7 months from conception to result: 3 months to plan, ~1 month to acquire the data, and 2 months to post-process the results. Fortunately, the lead designer was a PhD in statistics and was able to create very powerful results.
The challenge occurs when you can’t wait that long, or don’t need as great a result. Perhaps more importantly, choice modeling market research techniques are designed to test things that are well defined. They don’t work so well when what you want to explore is not well defined. And they don’t provide the insights that you get when you explore a topic in a group, because these techniques are all based on solitary behavior.
And, they’re not very fun. This may might not seem that important. But it is, as I’ll explain shortly.
The Bad THING That Happened to Choice Modeling…
Every one of us engages in choice modeling research every day:
“Hey, Honey, what movie do you want me to put to the top of our NetFlix queue?”
“Hey, team – what kind of food do we want to eat today? Mexican? Indian? Sushi?”
“Hmmm… We’ve identified 46 potential new product concepts. We can probably fund around 12 or 13. Which ones make the cut?”
“What’s the best set of features to include in our next release?”
From our personal to our professional lives, we’re constantly engaging in choice modeling market research. Constantly. And since we do it so frequently, many of us haven’t noticed the bad thing that has happened to choice modeling: Too many people have been convinced that they can’t do it on their own without a fancy method. Not surprisingly, since there is a multi-billion dollar industry who needs you to believe that you can’t do good, actionable, and useful choice modeling research.
That’s wrong. More often than not, you don’t need an expensive choice modeling research project to generate the insights you need (and, if you do need the fancy methods – then no worries – there are a lot of people who are happy to help you – more on this later).
It is time to right this tragic wrong and give you some great tools so that you can engage in “Market Research 3.0”.
Buy a Feature: Preference Research Through Collaborative Virtual Markets
Overcoming the challenges of traditional market research and righting the wrongs of the traditional market research industry are, of course, an opportunity for entrepreneurs like us! Our first game to tackle this problem was Buy a Feature. Buy a Feature is a choice modeling game based on a scalable virtual market, where a group of 5 to 8 players collaborate to “purchase” any of 12 to 20 items. The trick? They don’t have enough money to purchase everything they want, so they must collaborate with other players to get the best items (full details of the game are here). The game is fun, fast, and deeply engaging. It produces results that you simply can’t get through any traditional market research technique. Let’s explore why this is so.
- It is collaborative. To purchase high-priced items, players must work together. To work together, players argue and debate the relative merits of each item. By reviewing these debates you gain deep insights into why items are being purchased. These debates also serve to shape the items that are being purchased.
- It is fun. Players enjoy working with each other to get the items within the group. In fact, many times people will change their bids so that more players can join in purchasing highly desired items. Suppose, for example, an item costs $120 and each player is given $70. Technically, only two players must collaborate to purchase the item. What we often see, however, is that groups of 5 or even 8 people will make bids that purchase the item.
- It is engaging. As players review the list, they see items they want. To get what they want, they have to find ways to convince other players to join them. Who says what to whom is market research gold.
- We don’t overwhelm the players. By presenting only 12 to 20 items, we keep the number of items completely understandable. By structuring the game with 5 to 8 players, we promote meaningful participation.
Buy a Feature overcomes many of the challenges of traditional choice modeling techniques.
- It is fast. Unlike traditional market research, you have have a Buy a Feature game up and running in less than an hour, more than likely leveraging existing information about the projects.
- It is easy to post-process the results. You don’t need a PhD in statistics to understand the purchase patterns. You don’t need a degree in linguistics to understand how players negotiated purchases with other players.
- You don’t need well-defined items to play the game. In fact, one of the most important benefits of the game is that the act of negotiating purchases actively shapes the items desired by the players.
- You can see how collaborative purchases signal preferences. One of the most interesting results of a Buy a Feature game occurs when more players collaborate to purchase an item than necessary. This, of course, stands in stark contrast to traditional “rational man” theories of economics.
The concept of a scalable virtual market can be applied whenever you need to understand the preferences of a group relative to a set of items. For example, we recently helped the Cisco Collaboration Business Technologies team prioritize 46 projects through a specially designed set of tournaments. Think of a competitive sports tournament, like a swim meet, in which individual swimmers advance to the next round of the tournament by placing in their respective heat. In our tournament, the best projects advance based on the purchases made by Cisco employees. Each game within the took less than an hour. The results provided substantial insights into the highest impact projects.
There are NO points or badges within Buy a Feature. Because the Cisco employees are highly motivated to purchase the highest impact projects – the projects that enable Cisco to win and provide them the best opportunity for personal growth and achievement – there is no need to provide artificial motivation. The “motivation” needed for deep levels of engagement and participation is built into the structure of the system.
Knowsy: A Brand Engagement Choice Modeling Game
Knowsy is the first game that lets you determine how well you know your friends, your family, and your co-workers – while simultaneously testing how well you know them! The person who knows the other players the best wins the game. Originally designed for strategic sales account managers, who need to research the preferences of different constituents real-time, in a fast, fun, way, we’ve decided to release our first version of Knowsy as a consumer iPad app when our testers just kept playing the game for fun. Knowsy is even faster to set up and play than Buy a Feature, and the fact that it is a consumer game affords new kinds of choice modeling market research on entirely new kinds of data.
To understand why Knowsy is a serious game, you have to know how Knowsy is played.
- The game is played in rounds.
- At the start of the game, one player is selected as the first VIP (the Very Interesting Person). The VIP selects a topic (such as “Favorite Flavor of Ben and Jerry’s Ice Cream” or “Favorite iPad Game”) and announces this topic to the other players. As you can guess, players can also create their own topics.
- The VIP then privately selects and rank orders up to five items on the topic (Phish Food, Coffee Heath® Bar Crunch, and so forth) to create his or her Knowsy like-list. While the VIP is ordering their list, each of the other players are each of the other players are privately forming their own opinions on what the VIP likes based on their existing knowledge of that person’s personality and preferences.
- Once the VIP is finished the iPad is passed to the other players. Each player orders the list based on what they think the VIP likes.
- When everyone has had the chance to try their turn at guessing the preferences of the VIP, a score is computed based on how well each player knew the VIP, with more matches producing a higher score.
- Play continues until each person in the game has been the VIP. After the final round, the score of all the players are shared.
Knowsy is simple, fun to play, and very revealing. We’ve tested the game with adults, with children as young as five years old, and with people from different cultures. All enjoy playing the game. And we’re creating and will be soon releasing white label versions of Knowsy.
So, you may wonder, how is this a brand engagement and choice modeling game – and – why should I care? Let’s start with brand engagement. In Knowsy, players actually interact with the brand by dragging around beautiful images of the brand. The concept of choice modeling is built into the game structure. And you should care because people don’t find surveys fun – by they love playing Knowsy. Let’s consider these relative to our game design concepts.
- It is collaborative. Knowsy is played in a group, and players collaborate to discuss how “knowsy” they are about the other players.
- It is fun. It is quite a lot of fun to see how well you know the preferences of other players.
- It is engaging. The direct interaction model of Knowsy means that players directly interact with a brand in a really fun way.
One key advantage that Knowsy has over traditional choice modeling research is that it is a really cheap way for brands to conduct research. This is a really important point, as respondent recruitment and incentive fees can easily top $200/respondent. Knowsy acquires its market research data for free as a result of a natural gameplay. As Knowsy grows, we suspect that we’ll be hiring PhDs in statistics to help us process the massive data sets we’ll be generating through gameplay. Fortunately, our SaaS platform will make getting the insights you need from Knowsy easy.
What about points and badges? Admittedly, Knowsy does have points. We use them to keep track of the players status within the game. We don’t have badges. We added points to Knowsy because it is a natural way to capture a fun response. More thoughtfully, we are starting to realize that when you’re designing collaborative games for corporate problem solving things like points and badges may not be all that important. In the consumer realm, points, and yes, badges, are likely to be more important. We’ll be working on better understanding these design ideas over time.
But You Like Stickers, Right?
You bet we like stickers. They are part of our corporate values. And we’re not inherently opposed to points and badges. Indeed, we intend to explore the use of points and badges within our online games to see how they can motivate repeated play, mostly likely in longitudinal research. We’ll do this very carefully, as we want the players of our games to be engaged and highly motivated to participate through the structure of the game, and not because they are being paid $10 or because they get a players badge. Our points and badges will be based on engagement, participation, and the choice of the other players, much like players earning certain powers or rights for playing well in other games.
To illustrate, consider a few design ideas on how we could aware points in Buy a Feature.
- We could award points for playing the game. Uh, bad idea, because we’d motivate people to play about topics they don’t care about, invalidating the results. The game works because participants inherently care about the outcomes and strive to purchase the best ones.
- We could award points for certain behaviors – like more points for more chats in the chat log.Uh, another really bad idea. It isn’t the number of actions that a player takes in the game that matter, but the impact and effect of each action.
- We could award points for how skillfully a player negotiates with other players during the game. That’s a much better idea – provided the points were awarded by the other players, after the game was over, and these players had a limited number of points for distribution. Which means we might be better off having players award each other negotiation badges.
The take home point? Although we’re proud to be recognized as a leader in the serious gaming industry, we’re more than a bit concerned that the gamification movement is missing the point about points. So if you’re playing the gamification game, play it carefully. Because it is a serious game.