jump to navigation

How to use popularity lists to influence your users behavior May 20, 2009

Posted by jeremyliew in Consumer internet, product management, social media, user generated content.

Two years ago I asked if crowds generated wisdom, or simply crowdiness. Today’s WSJ has a really interesting article on the same topic, concluding that popularity is self fulfilling and that arbitrary top 10 lists can meaningfully increase the popularity of the items on that list:

A more-recent study demonstrates that popularity in the music world, even unearned, breeds more popularity. Researchers enlisted more than 12,000 volunteers to rate and download songs from among 48 chosen for their relative obscurity. Some of these volunteers were lied to: At a certain stage in the experiment, popularity rankings for this group were reversed, so the least-downloaded songs were made to appear most-downloaded.

Suddenly, everything changed. The prior No. 1 began making a comeback on the new top dog, but the former No. 47 maintained its comfortable lead on the old No. 2, buoyed by its apparent popularity. Overall, the study showed that popularity is both unstable and malleable.

I think music and other entertainment sources are an interesting case study because in these industries the problem of discovery is quite difficult to solve for many bands/movies/writers etc, as well as for consumers, and these top 10 lists can help solve that problem. However, it isn’t just improving discoverability that is important, but also the perception of popularity of the discovered items as another study found:

Another group of researchers demonstrated this with restaurant diners in Beijing. Table cards at Mei Zhou Dong Po, a Szechuan restaurant chain, touting the five most popular items boosted ordering of these items by 13% to 20%, according to a forthcoming paper by a team from Peking University and Duke University. “Part of it is reassurance that something is good and worth buying,” says Bill Paul, a restaurant-menu designer.

Calling these items popular is crucial, the researchers found, because other table cards that highlighted five sample items but made no claim on their popularity had little effect on sales. And the diners liked following the pack: “Diners who were exposed to the popularity information treatment are more satisfied,” says co-author Hanming Fang, a Duke economist.

These findings are consistent with one of Cialdini’s principles of persuasion, social proof.

The WSJ article mentions another study where a hotel tried to get customers to reuse towels. Claiming that 75% of people who stayed in the same room as the customer reused their towels increase towel reuse rates by 300% over the control message as you can see in the left hand column of the chart below.

Like the hotel, social media sites, e-tailers and other companies that are trying to influence their users’ click paths can use claimed or actual popularity to get their users to do more of what they want.

Crowdsourcing missions for MMOGs April 21, 2009

Posted by jeremyliew in game design, games, games 2.0, gaming, mmorpg, user generated content.

Really interesting post at Kotaku about City of Heroes experience with crowdsourcing story arcs.

In a letter to the community posted on the official City of Heroes website, Matt “Positron” Miller revealed that within the first 24 hours of the new updates’ existence, players in both hero and villain factions had created more than 3800 story arcs, each consisting of five missions a piece – more content than the development team had created during the game’s entire existence.

Players have been busy trying out missions and critiquing them in the forums as well. Out of the more that 20,000 arcs now available in game, 2,860 of them have been rated 5-stars by players, with only 582 rated at 1-star. Popular themes include the 5th Column, featured in 794 arcs; the super-heroic Statesman, starring in 134; and time travel, which is the subject of 112 arcs.

As an indication of volume, this is more story arcs that have been created by the game developers in five years!

One popular element was creating custom opponents notes the City of Heroes blog

70% of the arcs that are published use Custom Enemy groups. These are enemies created using our fantastic costume editor, coupled with a large sampling of the powersets that the game already uses. These unique enemies have proven to be extremely popular and sparked new life into the game. Players absolutely love fighting custom enemies for the simple fact that they no longer know what to expect. One of the biggest problems with MMOs is you eventually learn what all the critters you are fighting do, and the game can get pretty rote. Developers make new critters, but there can be months before you get new ones. Now players have the opportunity to be constantly making new enemies with new, interesting capabilities that can challenge and vex themselves and their friends, any time they want.:

I don’t play City Of Heroes, so I don’t know how directly applicable this idea is to web based social games. However, any of the social games currently available have very similar structures (e.g. the “wars” genre) which can get old over time. Perhaps this approach of crowdsourcing missions might add some interesting eldergame elements to these games.

Using Prosper and Mechanical Turk to figure out if people who are shifty look shifty March 19, 2009

Posted by jeremyliew in user generated content, web 2.0.

The Economist has another fascinating article about face. Specifically, about physiognomy – the idea that the way you look is a reflection of your character.

In particular, it describes research done by at Rice University to see if people could identify people who were bad credit risks by the way they look. They looked at 6,821 loan applications on Prosper. They asked 25 Mechanical Turk workers to assess each of the potential borrowers’s likelihood to repay a $100 loan. Here is what they found:

Their first finding was that the assessments of trustworthiness, and of likelihood to repay a loan, that were made by Mechanical Turk workers did indeed correlate with potential borrowers’ credit ratings based on their credit history. That continued to be so when the other variables, from beauty to race to obesity, were controlled for statistically. Shifty physiognomy, it seems, is independent of these things.

That shiftiness was also recognised by those whose money was actually at stake. People flagged as untrustworthy by the Mechanical Turks were less likely than others to be offered a loan at all. To have the same chance of getting one as those deemed most trustworthy they were required to pay an interest rate that was, on average, 1.82 percentage points higher, even when the effects of historical creditworthiness were statistically eliminated.

So it takes two web 2.0 services to tell you that many people who look shifty are indeed shifty.

How many user reviews is enough, and how many are too many? March 16, 2009

Posted by jeremyliew in Ecommerce, user generated content.

The Economist in their latest technology quarterly review look at how user reviews stimulate ecommerce.

They find that once you have about 20 reviews of a product, you start to see increases in sales conversion rates:

The sheer volume of reviews makes far more difference, according to Google’s analysis of clicks and sales referrals. “Single digits didn’t seem to move the needle at all,” says Mr McAteer. “It wasn’t enough to get people comfortable with making that purchase decision.” But after about 20 reviews of a product are posted, “We start to see more reviews—it starts to accelerate,” says Sam Decker, the chief marketing officer of Bazaarvoice, a firm that powers review systems for online retailers.

His company’s research shows that visitors are more reluctant to buy until a product attracts a reasonable number of reviews and picks up momentum. In a test with Kingston, a maker of computer memory, Bazaarvoice collected reviews of Kingston products from the firm’s website and syndicated them to the website of Office Depot, a retailer. As a result there were more than ten reviews per product, compared with one or two for competitors’ offerings. The result was a “drastically” higher conversion rate, which extended even to other Kingston products that lacked the additional reviews.

Even if some reviews were negative, sales still increase:

Online retailers have generally been reluctant to allow users to leave comments, says John McAteer, Google’s retail industry director, who runs shopping.google.com, the internet giant’s comparison-shopping site. But a handful of bad reviews, it seems, are worth having. “No one trusts all positive reviews,” he says. So a small proportion of negative comments—“just enough to acknowledge that the product couldn’t be perfect”—can actually make an item more attractive to prospective buyers.

However, some books on Amazon now have thousands of reviews, more than enough for a potential buyer to draw an overall conclusion. So why do people continue to write new reviews for these products, even years afterwards?

Mr Shirky suggests that in many cases, writing a review is more like writing fan mail (or hate mail) for a product, and the people who post them do not really expect it to be read.

Whereas new people continue to write reviews long after a book is published, blog comments have quite a different set of behaviors.

“You can probably have a decent discussion until you get to about 350 comments,” says Markos Moulitsas, the founder of Daily Kos, a popular left-leaning political site. But after that, he says, “most outside people will stay away from the thread, and further growth will come from people already inside that thread carrying forth a discussion, debate, or argument.” Such discussion threads are more of a conversation, and the page they inhabit usually has a limited lifespan during which people continue to post—unlike the Amazon pages for the “Harry Potter” books, which continue to attract reviews even today, years after the books’ publication.

Part of this is because the “pivot” of user engagement for a review is the product, whereas the “pivot” of user engagement for a blog is the conversation thread. Since the product is evergreen to new users, it will continue to attract reviews. But a stale conversation in the comments to an old post is unlikely to draw in new comment. It is usually clear that the other debaters have moved on from the conversation, and there is little incentive to speak to an empty room. Knowing what’s the right primary pivot for your social media drives a lot of design decisions.

This is reinforced by design; many blogs alert you to new comments if you’ve commented on a blog post; almost no ecommerce stores alert you to new reviews of products that you have reviewed. As a result, blog comments turn into conversations between engaged participants whereas product reviews. As always, behavior and culture are a function of UI.

How robust are communities? November 3, 2008

Posted by jeremyliew in communities, product management, usability, user generated content.

Wired has an article in its November issue about Urban Baby and You Be Mom. Urban Baby is an anonymous forum for Moms. Like 4chan, its anonymity makes for a mix of candid discussions, raw honesty and trolling, but with a mommy bent (think cheating, divorce and public schools). Says Wired:

Then in May, UrbanBaby, which was purchased by CNET in 2006, launched a redesign. All hell broke loose.

The changes weren’t huge, but each of them subtly altered the flow of conversation. CNET added a wide sidebar on the site to create space for ads. This reduced the reading area, a big problem on a board with hundreds of comments per hour. Discussions had been organized chronologically, but immediately after the relaunch, the default setting had “most popular” threads at the top, even if they had been started days earlier. Worse, you had to refresh your browser to see new posts. UrbanBaby users went nuts, demanding a return to the old design.

They soon got it. But not from UrbanBaby. A week after CNET rolled out the hated redesign, a couple of work-at-home computer programmers—longtime UrbanBaby users themselves—launched a rival site called YouBeMom.

They perfectly re-created the look and feel of the old boards. Better yet, they made improvements, including a souped-up search engine and privacy controls that make sure your spouse can’t use your computer to find out what you’ve been posting. They also set up a blog to capture users’ requests for site improvements and to outline what YouBeMom plans to do about them.

Within days, there was a mass exodus of users from UrbanBaby to the new site. CNET won’t give out traffic figures, and neither will the owners of YouBeMom. But I logged on to both sites recently and compared how often people posted. I’d estimate that YouBeMom has three times the traffic of UrbanBaby. That’s just how fragile a social application can be.

I found much higher comment volume and more vibrant conversations at YouBeMom as well when I looked at conversations on similar themes on both sites. The moral of the story according to Wired:

People have a very sophisticated sense for their online hangout—if you mess up the feel of it, or impede the ways they want to schmooze online, they’re gone.

What a terrific parable about the importance of community. What is strange though is that the traffic stats don’t appear to bear it out:

According to Compete, not only is Urban Baby far bigger that You Be Mom, but the redesign actually seems to have dramatically grown usage.

Sometimes communities are more robust than you think. Redesigns almost always create a lot of negative feedback when they first occur because all users hate change. You have to leave a little time to pass for users to get used to the changes before you can truly judge if the redesign has been a success or a failure.

There are three classes of user within social media, creators, curators and consumers. It may well be that many of the Urban Baby creators moved to YouBeMom, but the 90% of social media consumers, who read but don’t write, stayed at Urban Baby.

Do any readers have experiences of the impact of redesigns on a community?

When can paying people become counterproductive? August 11, 2008

Posted by jeremyliew in business models, game design, game mechanics, games, mmorpg, social media, user generated content.

I’ve posted in the past about how points can be used to drive user behavior.
Last week the Washington Post explored when play becomes work, and talked about some of the downsides of using rewards systems:

More than three decades ago, Edward Deci, a social and personality psychologist at the University of Rochester, found the first experimental evidence of a phenomenon with wide relevance to the way most Americans conduct their personal, professional and social lives.

Deci tracked a bunch of college students who were solving puzzles for fun. He divided them into two groups. One group was allowed to keep solving puzzles as before. People in the other were offered a small financial reward for each puzzle they solved.

The psychologist later evaluated the volunteers: He found that people given a financial incentive were now less interested in solving puzzles on their own time. Although these people had earlier been just as eager as those in the other group, offering an external incentive seemed to kill their internal drive.

The implication for social media and user generated content businesses is that creators create for love, not money, and that paying them with money may in fact be counterproductive. Instead, creators want adulation.

One interesting counterpoint might be the gold farmers inside World of Warcraft. When people play MMOGs for money, do they still play for fun? Anecdotally, it appears that they do. So when do rewards work and when are they counter productive?

But rewards and punishments are not always counterproductive, Benabou said. He drew a distinction between mundane tasks and those that carry meaning for people. In the first case, Benabou argued, rewards and punishments work exactly the way economists predict: They get people to do things.

External rewards and punishments are counterproductive when it comes to activities that are meaningful — tasks that telegraph something about a person’s intellectual abilities, generosity, courage or values. People will voluntarily perform intellectually arduous work, for example, because it gives them pleasure to solve a puzzle or win a game of wits.

“If I pay my kids to do their homework, I am saying, ‘You will get this if you do your homework,’ but I am also saying, ‘Homework is not likely to have intrinsic rewards,’ ” Benabou said. To the extent that a child is doing homework because he or she enjoys the challenge, or wants to demonstrate intelligence and diligence, the homework has meaning beyond the task itself, and Benabou predicts that offering a reward will backfire.

In most cases when it comes to user generated content, the creators do consider their work to be meaningful. So pay attention to how you pay them attention.

Online video CPMs are north of $15 July 24, 2008

Posted by jeremyliew in advertising, user generated content, video.

TV Week reports on a new report from the Diffusion group on video advertising:

Professional Web programming yields very high CPMs, the report found. The CPMs for long-form online content are $40 today and will reach nearly $46 in 2013. Meanwhile, CPMs for short clips are clocking in at about $30 and will rise to a little over $34 in five years.

The CPMs for user-generated video will have the smallest rise, from only $15 today to about $17 in 2013.

While many have viewed this negatively because User gen video ad rates are lower than professional programing, I think that $15 CPMs with no content creation costs sound pretty good to me!

Notes, video and commentary on the Social Gaming Summit June 16, 2008

Posted by jeremyliew in asynchronous gaming, business models, casual games, game design, game mechanics, games, games 2.0, gaming, mmorpg, social games, social gaming, user generated content, virtual goods, virtual worlds.

The Social Gaming Summit was quite a success on Friday. Over 400 attendees seemed to enjoy the sessions based on the high proportion of people in sessions (vs in the lobby) and the fact that even the last session, that ended at 6pm on Friday evening, was very well attended.

The attendee list was a good mix of game developers and publishers, with people coming from both the gaming side and the social media side. Most of the attendees with gaming backgrounds came from casual gaming, web based gaming or MMOG backgrounds. With the notable exception of EA, there were few representatives from the giants of the console space.

Although each of the topics covered different topics, it was clear that monetization was top-of-mind for all panelists as the discussion on most panels eventually turned to this issue.

I (Jeremy Liew) moderated the first session, on What Makes Games Fun, featuring game design thought leaders Amy Jo Kim, Nicole Lazarro and Ian Bogost, plus John Welch of Playfirst, the company behind one of the most popular casual game ever, Diner Dash.

The discussion was wide ranging and covered Nicole’s framework for generating emotion in games and the four types of fun and Amy Jo Kim’s five game mechanics.

There was excellent discussion about how fun, addictiveness and business models can either collide or work together, with in depth discussion of two games in particular, Pack Rat and Parking Wars.

Pack Rat was lauded as an example of a game that did a masterful job of creating addictiveness through game mechanics, and a game that had a natural digital goods/service business model baked into it. But some panelists questioned whether the “grind” without real “payoffs” at different levels could burn players out. In contrast, Diner Dash had real changes in game dynamic and strategy as players level up (e.g. when Flo gets the coffee maker at level 4, it changes the winning strategy) that made leveling up more meangingful and rewarding.

Parking Wars was pointed to as a highly social game with a genre matching to the mass market that let players “play slight variations of themselves” where they could explore slightly nefarious behavior in a safe environment. But “winning” in Parking Wars forced activity to the edges of the social network, instead of to the core, so the “points” game mechanic ended up working against the “fun”.

UPDATE: Virtual worlds has an excellent writeup of the What Makes Games Fun panel.

The second session was focused on Casual MMOs and Immersive Worlds, with Joey Seiler from Virtual World News moderating representatives NeoPets, Nexon, K2 Networks and Gaia.

One of the key questions was how to get free to play users to open their wallet. Gamasutra covered this panel in detail and noted:

Added Kim (Nexon): “A lot of people think they can make money off of casual games where people play a couple of hours a week. I don’t believe that. When people get engaged with the social experience then they’ll buy items. You need to understand the psychology.”

Reppen (Neopets) continued: “For us, it’s all about a sense of ownership that our audience has. There’s a real sense that it’s their game… The identity component to virtual worlds is so important, but there’s so many other things going on in the meta games around earning points, acquiring wealth, shopping and customizing and creating your own experiences… It’s part of a mix.”

In other words, even for casual MMOGs, you monetize the hardcore players who tie their identity into the game. Erik Bethke (GoPetsLive) said the same thing at this years GDC previously in explaining why he applies game dynamics to make virtual worlds more addictive.

UPDATE: Massively writes up the panel in Q&A style.

After lunch Andrew Chung from Lightspeed moderated a panel on Asynchronous Games on Social Networks with the CEOs of the companies behind many of the top games on Facebook, including Friends for Sale, Zombies, Vampires, Warbook, JetMan and (fluff)Friends.

Inside Social games
took live notes from the panel. One interesting counterpoint in response to the question, “How do you move people down the spectrum to make them more engaged and hard core?”:

Blake (Zombies, Vampires etc)- There is always going to be some subset of your userbase that’s never going to play more than their 30 minute lunch break, because that’s all the time they have. Don’t inundate users with too much experience at the beginning, gamers hate to read, I’ve never read a game manual in my life.

Siqi (Friends For Sale)- I think there’s a lot to learn from traditional MMO design, things like levels. If you get to the next level, you get this new shiny thing. It makes the game more complicated, but it works. Our hardest core users use more synchronous features.

Shervin (Warbook, Jetman, etc) – The first generation of social games were incredibly simplistic, and the platform was so viral, that it was a lot easier for apps to grow. But it behooves all of us to invest in content. I’m staying up late at night building social games 2.0, games with richer content, deeper stories, much better user experiences. It’s going to become harder for independent developers. I can’t talk about the games we’re working on, but you can look at Playfish. Their engagement levels are high and they’re growing faster than those that have come before.

In other words, games need to be easy to learn, but hard to master.

Next up was Dean Takahashi of VentureBeat moderating a panel on User Generated Games in Social Networks and Virtual Worlds. The speakers were from IMVU, Dogster, Three Rings (Puzzle Pirates, Whirled and Bang Howdy) and Habbo.

Virtual Worlds News has coverage of the panel and noted that:

In IMVU, said Rosenzweig, creators “do what they do because it’s cool, but they like making credits” by selling the items in world. That can then be cashed out through IMVU, which leads to 90% of its revenue, taking a cut while transfering IMVU credits to real world dollars. That user attitude is true of Dogster and Catster as well–users don’t get a cut of the money generated by creating games around their items and boosting activity. They just enjoy creating and sharing.

In other words, social game players generate content for love, not for money. But if there is money there to be had, they don’t mind taking some of that too! Last month Chris Alden noted the same experience in the blog economy.

UPDATE: Worlds in Motion also has a writeup of this panel

After a short break for cookies, the attendees reconvened to hear Brandon Sheffield of Gamasutra moderate a panel about Building Communities and Social Interaction In and Around Games, featuring the leaders of Kongregate, Zynga and Addicting Games, along with noted social architect and game designer Amy Jo Kim.

The discussion centered on the desire that many users had to communicate with each other, and how games often served as an easy way to break the ice and provide topics that made it easy to start a conversation. I haven’t found any coverage of this panel online unfortunately.

The final session of the day was focused on Monetization and Business Models for Social Games. My partner Ravi Mhatre moderated the panelists, including the leaders of Mochi Media, Sparkplay Media, Stardoll and Acclaim. This was a fantastic panel. Virtual Worlds News has great coverage.

Although most of the discussion was focused on the four models of advertising, subscription, digital goods and retail, David Perry noted that there are by his count 29 business models for games.

On the mix between advertising and virtual goods, the panel mostly agreed that virtual goods was the primary revenue stream but that advertising was an important secondary stream:

“Microtransactions and advertising go perfectly togetehr,” said Miksche. “Microtransactions drive our business, but we will never have 100% of our users wanting to pay for that. Advertising is a good way to monetize that remaining X percent.”

There was some good discussion about the tension between game balance and letting players buy powerful items in the games. Several panelists noted that self expression was a key driver of virtual goods sales:

As for who’s paying, Perry (Acclaim) expected most microtransactions to come from hardcore MMORPG playerskitting out the avatars with fancy armor and such. Instead, it comes from Dance. The game is a simple dancing activity, but because users spend so much time looking at their avatars, the appearance and identity becomes even more important.

That works well for Stardoll, a fashion-themed site, especially with trends that match the real world…

“We’re One-Click Dressing,” said Miksche (Stardolls). “You come to the site and instantly start dressing. For our users, young girls, that’s very important–instant gratification.”

For those who couldn’t attend, UStream.tv hosts video from the social gaming summit.

Andrew Chen’s blog also has his takeaways from the social gaming summit.

I’ve pulled together all the coverage I could find, but if there were additional posts, please let me know in comments.

NY Times on why young women are driving the social web February 21, 2008

Posted by jeremyliew in social games, social gaming, social media, social networks, user generated content.
add a comment

The NY Times realizes that teen girls are driving social media:

Research by the Berkman Center for Internet & Society at Harvard Law School, the result of focus groups and interviews with young people 13 to 22, suggests that girls’ online practices tend to be about their desire to express themselves, particularly their originality.

“With young women it’s much more about expressing yourself to others in the way that wearing certain clothes to school does,” said John Palfrey, the executive director of the Berkman Center. “It ties into identity expression in the real world.”

That desire is never so evident as when girls criticize online copycats who essentially steal their Web page backgrounds and graphics by hotlinking (linking to someone else’s image so it appears on one’s own Web page). Aside from depleting bandwidth, it is the digital equivalent of arriving at a party wearing the same dress as another girl, Professor Palfrey said.

People building social media and social games need to think about their teen girl strategy; as I have noted in the past, it is no accident that Typhoid Mary was a woman.

Games 2.0: Lessons from Travian February 14, 2008

Posted by jeremyliew in asynchronous gaming, casual games, game mechanics, games, games 2.0, gaming, social gaming, user generated content.

Travian is a popular in-browser asynchronous massively multiplayer game; you build a village in “Roman” times and set off to expand your empire, raid others and form alliances. It has many of the games 2.0 characteristics that I’ve been blogging about and has been growing nicely, as Alexa shows:

alexa graph for travian

As a passive web game:

it’s persistent. it’s massively multiplayer. it’s competitive. it’s social. it’s portable. it’s passive.

passive web games are setup to permeate your life. they become habitual. they are inherently attractive to gamers with little time — whether that time is taken up with work or other games. they fit unobtrusively into the corners of your life, taking as much or as little time as you want to invest.

… it’s a game you’ve never heard of, but, it gets 225 million page views a day

Most game reviews are glowing, noting the appeal to both casual gamers and obsessive gamers. One review though notes that at heart this game is about negotiation and diplomacy:

Politics is the name of the game, as while there’s room for intelligence to make a difference in combat, in the end, if someone has enough troops (and the resources required to build them) they can crush anyone. So friends are important, or at least fellow wolves

The reviewer complains that the scale of the player base ultimately becomes a problem for a diplomacy based (ie social) game. With 20,000 or more players, it far exceeds Dunbar’s number.

I can’t help but think whether building such a social game on top of an existing social map might improve gameplay even further.