Or, The Barkeep and the Bouncer. Call 'em social networks, discussion forums, blogs, whatever. Some businesses have never learned to distinguish between participating in a conversation and trying to sell an online audience. Here's a free clue: the latter doesn't work. User-contributed content is big business nowadays. The whole notion of community is far more common than it was when I started posting on BBSes and when I was a CompuServe sysop. There are now several online community and social network business models and justifications, including online advertising (as it is here at CIO.com), review\/rankings that can encourage product sales (Amazon is the prime example), membership upselling (as at Flickr or LinkedIn), brand-building corporate blogs, open source projects that want to encourage more developer involvement, product support sites and probably a few that slipped my mind. This is a major change. At an online community conference I attended in 2000 (where Craig Newmark discussed his intention to maybe take Craigslist beyond San Francisco), most of us community leaders were fretting about how to demonstrate community ROI. While that's still a concern, I think Management is already sold on the necessity for online customer interaction, even if they aren't sure what they're going to do with it. But "doing community" doesn't mean "doing community well," and I'm sad to say that plenty of businesses have screwed it up (and often pointed to the wrong source of failure). In the 25 years I've shared myself with the world online (I live more in TCP\/IP packets than in a physical community), I've seen entirely too many dumb mistakes made, over and over, usually by businesses that should know better. Here's some of the stupidest errors. We'll pretend that your company hasn't made any of them.1. Trying to control the conversation. Actually, the problem here is that the company generally doesn't recognize that they're in a conversation. They think of the online community as part of a "marketing message," which has a defined target audience, on-point theme, and so on. All of which can and should be controlled by a marketing department. And which completely misses the point of community. As my friends Rick Levine, Christopher Locke, Doc Searls and David Weinberger wrote so many years ago in the Cluetrain Manifesto, the Internet has helped to make the world an ongoing, always-on conversation. Conversations do not happen from a podium. They rarely happen from a panel discussion, even if audience members can ask questions of the Authorities. Conversations happen around a table, with everyone on more-or-less equal footing (though some people will always know more than others... and that's one reason people show up). Still, far too many businesses set up an online community and then insist that they have to control the conversation. Criticisms are removed; questionable language rejected; conflict discouraged. All very Happy-Happy-Joy-Joy. And always a complete flop. People don't come to a company's community because they want to see only good news. They participate because they want to share successes and failures. Because they want to solve a problem, or to help someone else avoid one. (And for several other reasons that I won't go into.)If avoiding the problem of "I wasted 50 bucks" means they'll write, "I hate this product and I'm sorry I bought it," they're going to share that passionate response. They might write it on your site, or they'll write it on Amazon, but they will write it. Your only option, on a company-sponsored site, is to conduct the conversation in a venue where you can respond. And by "respond," I mean, "You have an open and public conversation." Simply by permitting (and encouraging) the transparency, you admit to the world that your company pays attention to what its users and customers think. That is never a bad thing. Bottom line: If you want a community to be a success, approach its design and purpose from what the participants will get out of it\u2014not what your company aims to achieve.2. Forgetting that volunteer participants are part of your company's community. The people who take the time to post messages in your community on a regular basis are, in some way, committed to your business model. If they write product reviews for your site, it's because they like your site and what it offers, and they choose it over the competition. With luck, that also means they prefer to give their money to you rather than to a competitor. Cherish that. That means you have to treat your committed community members (say, your top reviewers, or forum members who go out of their way to share tech support tips) as though they are members of your team. Minor members, perhaps, with minor recognition. (Borland had Team Borland t-shirts for its active CompuServe forum leaders; those were great T-shirts. I was jealous.)If you are going to make a big change in the way the system works, you should involve those committed members in the decision as soon as possible. Listen to them\u2014or feel their wrath. Don't "announce" a big change in the community system; invite these people to participate in the decision-making. They probably spend more time interacting with the community system than you do. 3. "Managing" a community by parachuting in, then parachuting out. Companies sometimes imagine they're using social networks or online communities to engage their audience. But the scope of their community involvement is to put the software in place and then, on an irregular basis, send in a well-meaning but powerless marketing person to make some sort of pronouncement. (Such as, "We're changing the way we rank product reviewers.") If you really want to reach your community, and you truly mean to have a conversation that benefits both your firm and the people you intend to serve, then your executives should participate at least occasionally. I don't mean write a PR-prepared "holiday message from the president." I mean a sincere public response from the boss to an upset customer, promising to get to the bottom of the problem (and then making sure she does). The truth is: Any community that works has at least one person in the role of community moderator. This may be a volunteer role or it may be paid, but it has two important components which you'll observe in any successful bar: the barkeep and the bouncer. It is the community moderator's job to welcome new members to a community and encourage them to participate (the way a bartender makes a customer feel welcome); it is also her job to throw out those who do not conform to community guidelines (the bar's bouncer). When the community moderator also serves a corporate master (as opposed to someone who helps to run a peer forum), she also needs the power to report issues and expect swift action. ("We're getting a lot of complaints about the new WhizBang model. What shall I tell them?") These people should be cheerful, knowledgeable and well-paid. 4. Failing to see the community as volunteered market research. I can understand how companies can miss the first three points, because some of these lessons come from experience. But I cannot fathom how businesses miss the to-me-obvious fact that anyone who plays in your sandbox is a free focus group participant. This is particularly evident in technical support forums where the marketing people dare not tread (and where, I suspect, the tech support community moderators are just as happy they don't). However, if you see a lot of messages posted by customers who cannot figure out a product feature, you do not need to do a major research study to determine that the feature should be fixed. In fact, that's a perfect place for the boss to step in, the way that, in CompuServe days, Pete Petersen did during the development process for WordPerfect 5.1 ("Here's a screen shot of the way it'll look; does this work for you?"), and the way Philippe Kahn responded to Borland's customers (including starting a thread that, bless the man, he titled, "Bull merde"). The benefit to the boss is that he, too, can be in-tune with the customers' mindset.My favorite example for this market research point is ancient, from the early days of the Web. A hosiery company slapped up a forum and was surprised to see how many people had things to say about pantyhose. But they were even more surprised to see a discussion thread in which several men explained what they wanted from the company: pantyhose designed for guys. (Nothing effiminate about it, they said; they just found pantyhose more comfortable than socks. Except women's legs are differently shaped, so could the company create a new line for them?). As far as I know, the company never complied (and I note that they no longer have a forum\u2014Ha! Cowards!); but this is market research no one would have gained otherwise, because nobody would have thought to ask the question. One warning, here: Do not overgeneralize. One visible complaint or comment does not mean that every user feels that way. One complaint is the beginning of a conversation, and an opportunity to ask the community members, "Is this true?" 5. Imagining that employees will contribute to an internal community without special attention.Internal communities have their own unique circumstances, because they reflect the corporate culture. If yours is a company in which everyone trusts one another, where "dumb ideas" are encouraged in the name of brainstorming, and where the politics is under control, then you're in fine shape. (You are also tops on the list of companies where people want to work.) But whether you call your online community a wiki on the intranet or a corporate knowledge base, participants have to trust one another; they must feel safe, appreciated, recognized. All the components of the previous four points apply... except they apply more, because there's always a sense of competition. If I share this knowledgebase article explaining how to do something, does it make me more valuable to the company ("Gosh, look how much she knows!") or less ("We have now extracted all her wisdom and transferred it to others; her position can be eliminated. Perhaps she can be replaced with Folger's Crystals")? There are plenty of ways to address these concerns, I'm sure; but the first mistake is to fail to recognize their existence. 6. Putting Technology in the Way of the CommunityI've seen a bunch of companies get their shorts in knot about technology choices. I don't mean only "Should we have a corporate blog?" but agonizing about their software choice or creating a custom app with bells and whistles. The truth is, the best tools are those that are simple and fast (or which can be automated into simple-and-fast) because they reduce barriers to participation. (That's probably true for just about anything.) The easier you make it for people to participate, the more they will do so. I wrote all the above without reference to anyone else's opinions, referencing only my own time spent enduring frustrating user experiences (including--transparently, you notice--some right here in CIO's own blogging system). But I am far from the only authority on the subject. If you want to learn more about how make an online community a success, here's a few links you should probably follow. Just as a start. Kathy Sierra wrote a few brilliant blog posts about How to Build a User Community. For a current picture of the state of online communities, see a report from the Online Community Summit (which, dammit, I couldn't attend but wish I had). Patrick O'Keefe has a new book, Managing Online Forums: Everything You Need to Know to Create and Run Successful Community Discussion Boards, which I confess has been sitting on my To Be Read list for a few months now. I've only flipped through it, but it seems to be a very good guide for those who feel they might benefit from clue-acquisition.