LP Login

Think Big. Move Fast.

Bokardo has an interesting interview with Bryce Glass of Yahoo, about Yahoo’s social design pattern for reputation. Building reputation systems can really help drive high quality engagement on a social media site, but is also fraught with danger and unintended consequences if not done thoughtfully:

What are the biggest hurdles in designing for reputation?

I think it’s probably the number and variety of unintended consequences that little design decisions can have further down the line. I’m fond of the article—so I cite it a lot—but Ben Brown, who founded the dating site Consumating, has a great blog-post about the ‘ill-fated points system’ that they used for that site, and the variety of… um… less-than-ideal behaviors that those incentives gave rise to. Early on, Slashdot struggled with many of these same issues, and they’ve re-jiggered their comment karma system several times through the years.

A big hurdle—and if you can solve this, you’re halfway there to having a well-designed and effective reputation system—is appropriately marrying the incentives that you offer your users to the appropriate set of goals that you have for your community. You want to be sure that you’re rewarding folks for behaving like good citizens, and not just rewarding them for no good reason. (Or for vague and misguided reasons like “to keep them engaged” or “so we can have a leaderboard.”)

Earlier this year, Glass gave a presentation on designing your reputation system at the IA summit outlining eleven different reputation systems:

    > Named Levels
    > Numbered Levels
    > Identifying Labels
    > Points
    > Collectible Achievements
    > Leaderboard
    > Top X
    > Temporal Awards
    > Statistical Evidence
    > Peer Testimonials
    > User to User Awards

and how to select between them depending on questions including:

    > What are your business goals? [Engagement? Promote a specific feature? Acknowledge top contributors? Increase content quality? User Retention?]
    > What community spirit do you want to encourage? [Caring? Collaborative? Cordial? Competitive? Combative?] More detail on the competitive spectrum in Yahoo’s Design Pattern Library.
    > What motivates your community members?
    > Which entities will accrue reputation? [People? Things? Collections of Things?]
    > Which inputs will you pay attention to?
    > How transparent should the rules be? [More transparency is more likely to affect behavior]

He also notes that reputations should always decay over time to prevent a log jam at the top that can discourage new members and make a community appear stagnant.

Yahoo has done a nice job of categorizing some of the various reputation systems available to social architects and how to think through choosing one. I would highly recommend reading the interview and presentation and reviewing the material in the design library.

  • http://t-machine.org adam

    Thanks for the post, I hadn’t seen the Y! slides on this before.

    As an MMO developer, I’d say that the answer to “which form of reputation system for your social game?” is simply and clearly “all of them”.

    I’ve worked on games that have had a heavy social/web element, and adding additional parallel rewards/reputation systems has only ever helped both the community and the game. Nowadays, everything I see reinforces this, at least for games.

    For instance, easy example – look at Kongregate. Kong has 5 independent, parallel rating systems for each game, and 7 (!) reputation systems for each user/player/developer on the site.

    Looking at how those interact with each other, I would argue that a lot of the site’s success is precisely because it has these multiple *independent* forms of valuing user content; it allows you as a member of the community to say “this is nothing special in many ways, but in one aspect it’s the best thing I’ve ever seen” – essentially allowing for a pareto measure of “goodness”.

    So, although I understand and by default agree with Y!’s suggestion that you should look at your community, assess your target market and your product aims, and pick one reputation system, I feel that you really shouldn’t do that with games.

    And that leads me to wonder whether the suggestion itself is a nice idea in theory, but perhaps not appropriate in the modern web world: perhaps communities now are sufficiently savvy, picky, and accustomed to being the ones to control success (e.g. youtube, where the community makes a video successful, not the site owners), that single-value measures of reputation are no longer what your communtiy wants and needs.


  • Pingback: T=Machine » Reputation and ranking systems for online games and web games()

  • http://t-machine.org adam

    …now with more detail on my blog:


    I’ve got a followup post coming that looks at the 12+ ranking systems currently in place on Kongregate, a subject that’s come up a couple of times in meetings and planning, so I decided it was time to write it down instead of just repeating myself to different audiences all the time :).

  • Pingback: T=Machine » Kongregate’s Rating/Reputation systems()

  • Natasha

    “(e.g. youtube, where the community makes a video successful, not the site owners)”

    While online social communities do seem to have a rising desire and comfort in applying reputations to its members (per community-determined values as the Glass presentation stated), let’s not forget that members are only able to express those opinions to the extent of the tools they are given within the property.

    The youtube community makes a video successful largely by number of views, but that’s information the site owners chose to display. That’s part of what successful reputation system toolboxes are– a correct identification of which information transparencies the community will find valuable.

  • Pingback: Inside Social Games » Blog Archive » Reputation Systems Can Build Community Inside Social Games - Tracking the Convergence of Games and Social Networks()

  • Pingback: Quora()