Gail Williams of Salon and The WELL invited people this week to share their thoughts on the subject of "influencers," and recaps the contributions here.   Following are some simple tips I share with companies launching online customer communities about how to cultivate "superusers" -- that group of active users from which influencers and advocates can emerge.

PRE-LAUNCH

Identify superuser candidates among known advocates, online or off.  Even before beginning their own online efforts, many companies already know people who are strong believers in their brand and their products.  They may know them through offline forums like user groups and user conferences. Today, they may also know them through groups they have created in Facebook or people who are following their Twitter feeds.  If you already have a relationship with these folks, reach out to them directly -- engaging them in your community can strengthen their ties to one another and to you.

Identify existing online communities (if any) to reach out to.  If you're like most companies, there are already online groups that aren't hosted by the company but provide a place where your customers are receiving peer help and support today.  I don't recommend that you reach out directly via posts in their forums or messages to those users.  Every forum that provides help to your customers is an asset to your organization, and you don't want to jeopardize that relationship by being disrespectful to the organizers of those groups.  You aren't in competition with them -- I always say, a healthy ecosystem is your best friend.   Instead, reach out directly to the organizers of those groups. Let them know of your plans.  Let them know that you appreciate what they contribute to the ecosystem around your products, and tell them you hope to your community can contribute to it as well.  In most cases they will appreciate the respect you show them, and they won't be shy about letting their superusers know about it.  Sometimes they often become superusers themselves.

Create "superuser-friendly" user guidelines and moderation policies. It probably goes without saying -- the best way to cultivate superusers is to create a great environment for them to do their work.  Sometimes companies say, "we need to be strict at the start in order to satisfy Legal, then we can relax once they realize everything's ok."   They then spend the first few weeks hitting users over the head for minor infractions, and guess what?  A lot of those early users are the superusers who, if you treat them right, will answer 30% or 40% of all questions.  Who answers those questions when you chase all these "answer people" away? That would be you. 

In this regard, most companies focus on user guidelines, but that's only half the story.  You also need good moderation polices. User guidelines describe what users should and shouldn't do.  You need a plan for what happens when they do what they shouldn't. It has to be well thought out, consistent, and respectful.  And anyone from the company who moderates or participates needs to be trained in those policies.

LAUNCH

Invite known advocates to preview community, create seed content, and provide feedback. Invite your advocates to take an early look at your efforts, a day or two before they are open to the open to the public.  They will appreciate being asked.  Not only will you get great feedback and buy-in from these users, but you'll solve a difficult as well -- they will help you develop a light layer of seed content which you'll need to launch successfully.   This is much better than creating phony accounts, which I don't recommend.   You don't need much content to launch successfully -- in the early days, users will look for freshness as much as depth, and will understand that a new community won't have deep content on day one. 

And those organizers of other forums?  Invite them to preview your efforts too.

Add elements to community structure that allow superusers to identify you (welcome forum) and to identify themselves (feedback forum). Among the forums and blogs and idea boards you'll see in Lithium communities, you'll notice two common forums -- a welcome forum, and a community feedback forum.  I can think of lots of reasons why you'd want these two forums, but here's why they matter to your superuser efforts.  The welcome forum tells potential superusers that someone is running the place, and identifies who that someone is.  Superusers expect that their efforts will be noticed.  A moderator can't do this alone.  Only the person in charge can.  

The feedback forum accomplishes a different purpose.  It's the place many superusers use to announce, "I'm here."  They do that by making suggestions for new boards, features, etc. 

By the way, this is a "feedback on the community" forum, not a "feedback on our products and services" forum.  You'll get the latter anyway, in the forums dedicated to product or service discussions.   And a forum designated generically for "feedback" is a magnet for rants.

And lest I forget: the welcome forum always goes at the top, and is always read-only.  That's your podium.  No one else gets the podium -- that's your privilege as the community manager.   The feedback forum goes at the bottom, so it doesn't get in the way for most users, who simply have a question they want answered.

Develop a rank and reputation structure to reward and incent superusers. My colleague Michael Wu has given you a graduate course on rank structures on his blog (parts 1, 2, and 3), but let me share a few more general pieces of superuser-specific advice here.

As with online community management in general, superuser management ain't about the software. That being said, as in every other area of community management, your platform can empower you or constrain you in your efforts to do what you need to grow your superuser group successfully.  One empowering element is a reputation system.   A reputation system works 24 hours a day, 7 days a week to recognize users for their participation.  You decide (often with input from your members) what activities matter in the community, and you write those activities into the rules that govern the assignment of ranks.   Users are automatically granted ranks when they have satisfied the requirements you define.  In the Lithium system, these ranks can also carry permissions, meaning that users can earn the privilege of doing things that other user's can't do.

Some companies, like FutureShop, make their rank structure completely transparent to users.  Others, like Nokia, reveal the rank structure but not the criteria.  If you read the entire thread at Nokia, you can see how the rank structure has evolved over the past 3-1/2 years, and user feedback and suggestions along the way. 

Should you reveal the criteria or not?   Most customer communities do not reveal the criteria to members. This is particularly true in consumer communities. Keeping the criteria secret prevents users from "gaming the system" to move up the ranks.  But as you can see from the Nokia example, it also adds an element of fun, as users speculate on what it takes to move up from level to level.

But there are advantages to transparency as well, particularly in professional or business-to-business setting.  In fact, many customer communities today use a blend of transparency and opacity, delivering the former through leaderboards (see Research In Motion's Support Forums for an example) and the latter through ranks.  

It's hard to generalize about rank names because it depends on the community, but I prefer rank structures in which no rank is disparaging (no "n00bs"), lower ranks don't suggest expertise (doing so can create cynicism), and upper ranks have titles which reflect the seriousness in which many users will regard them.  (Don't forget that some people treat them as a credential they have earned -- which they are -- and sometimes even put them on their resumes.) 

POST-LAUNCH, FIRST 30 TO 90 DAYS:

Identify emerging superusers as they move up the ranks and provide positive feedback. Within the first 30 to 90 days of a community, the community manager's task with regard to superusers is very simple:  you must make sure that your superusers know that you know they are there.   You don't need to give them anything, or promise them anything, or give them any privileges, or fulfill any of their requests.  You simply need send them the right signals.  If they send you a message, answer it promptly.  When you observe people rise up the ranks, or who otherwise stand out for the quality and quantity of their participation, send them a message of thanks.  Done.

Acknowledge their suggestions and ideas, without making commitments. I mentioned the value of the feedback forum, but there's an obligation there too.  A community manager must acknowledge those suggestions and thank the user publicly for offering them.  You don't need to agree to those changes, unless of course they are something overlooked and easy to fix, but you do need to acknowledge them as ideas you will consider as the community grows.  An offline message asking for more details is good too.

Tune reputation system based in real user data. Around the 60 day point you'll want to "sanity check" your rank structure to make sure the quantities you set for each ranked have proven to be reasonable and achievable.   Michael has some tips on this, but running a quick report that shows users by rank will show you whether users are moving up (good) or stagnating (bad).

POST-LAUNCH, 90 DAYS AND BEYOND

I say 90 days here, but the following steps should really occur no earlier than 90 days in most communities, and very often occur later.  The key question to ask are, a) has a group of superuser emerged, and if so, b) do I have enough history on each of those users to understand their habits and character?

Review participation history and refine superuser group down to supporters. Generally after 90 days, you can answer "yes" to (a).  It is generally a small group at that point, and the users are characterized by daily visits to the community, frequency of contribution, and quality of contribution.   All of these criteria can be measured and reported -- assuming that you provide a way for other users to rate the quality of a contribution.   

However, there's one other criterion that matters.  It's one that distinguishes supportive superusers from other who are not supportive.  By "supportive" I don't mean that they are advocates of your brand or product, although they often are.  I mean supportive of the community and its goals.   In general, this means they understand the mission of the community as you have defined it, and that their participation supports that mission.  At a minimum, it means they don't violate community guidelines.   More broadly, it means they provide a model for others and help ensure that other members participate successfully too.  This isn't true of every superuser, or even every helpful and knowledgeable superuser.  But it must be true of superusers who you bring closer to your organization as you formalize your superuser program.

Create a private forum accessible only to superusers. I like the idea of offering superusers a private forum where they can engage with you and with their superuser peers.  Because they are in your community every day, they notice things they want to bring to your attention -- for example, problem users -- or have questions they would prefer not to share on the community at large.   It's a privilege you can offer them, and it helps you as a community manager as well.   

Why wait 90 days?  Well again, unless you know these users via some other forum, you'll need some history in order to determine if they are supportive.  If you create a forum on Day 1, here's what will happen:  you'll invite your most active users.  Over the next few days or weeks, you'll realize that one or more of those users isn't particularly mindful of the guidelines, and in fact prompts complaints from other users.  However, when you try to remove that user from the private forum, other superusers will object -- you can't remove Bob, Bob is not a bad guy, give him a chance, etc.  At that point, mere days into your new community, you've created an "us and them" situation with your superusers. As a community manager, you'll have plenty of battles to fight -- you'll be thankful you didn't start by sparking this one.

Create a specification for the superuser program -- criteria for selection, term of membership, etc. Of course, superusers don't work for you, and you don't assign them tasks -- they do their work voluntarily because of the satisfaction they get from doing so.   However, if you do grant superusers privileges like access to a private forum, you do need to think about how you administer additions to, and exits from, the group.  Not everyone will want to participate forever, and you shouldn't want or expect them to.  Nor should you lead them to believe that they are "lifetime members."   In other words, you need to create a formal program.  

Microsoft's MVP program is larger and broader than a superuser program for the average online community, but I think its principles apply to almost any superuser effort.  Sean O'Driscoll describes the MVP program in this video.

Reward and empower superusers with additional permissions and privileges on the community. If your community platform permits, you may want to grant superuser privileges that other users don't have.  Access to a private forum, as mentioned previously, is one.  But you might also consider things like the ability to upload attachments or videos, to author blog posts, or to edit or delete messages.  There are more than 90 permissions on the Lithium platform that can be distributed in this way.

Explore other opportunities for rewarding and growing the superuser group (invitation to focus groups, participation in beta tests, previews of products, recognition at conferences, in-person meetings, etc.) Over time, you'll find many opportunities to reward and engage your superusers. Access to betas or early versions of products is particular prized by superusers -- not for any monetary value, but for the chance to keep their knowledge as current as possible. Don't neglect the chance to bring some of those virtual relationships into the physical world by inviting your active users to join you at industry conferences that may be of interest to them. I know a community manager who keep track of all the cities where his superusers reside, and simply calls them when business travel takes him to their area.  In our increasingly virtual world, there's still nothing like a having a cup of coffee with a friend.

If you've been to our home page this morning, you know that today we've introduced Lithium Social CRM, the next generation of our solution for customer communities. It's been many months in the making.  In many ways, it's the natural evolution of everything we've done and everything we've learned for the past 10 years. To mark the occasion, I thought I'd share some of my thoughts about how and why the worlds of "social" and "CRM" have come together, and what it means to the average business (i.e., our customers).

The topic of Social CRM actually goes to the heart of what this blog is about. Because I call this blog "Inside Enterprise Communities,"  I'm often asked, "How are enterprise communities different than any other web community?" The answer is that an enterprise community is created within an ecosystem of technologies, processes, and relationships. There's already a "there" there -- you are adding to it, harmonizing with it, integrating with it -- not creating it from scratch. This is fundamentally different than just you or me putting up our shingle on the web and trying to attract members.  

But integration with that ecosystem has been a long time coming. For many years, customer communities were nothing more than a message board at the far corner of the company website.   These communities were disconnected from other systems, often requiring a separate username and password. They were also  disconnected from other business processes. For example, if you called the customer support line and asked about an issue well-documented on the company's support forums, you'd often hear, "Sorry, that's not a known issue."

This isn't to say that these companies were wrong or stupid. As computer scientist Phil Agre says, technology changes rapidly, people change fairly rapidly, and institutions -- that is, that bundle of processes, policies, practices, roles, and responsibilities  we call "the organization" -- change very slowly. So it's taken about 15 years since the arrival of the commercial web to see the Internet fundamentally change the way companies work.  Social CRM is a milestone in that process of change.

So let's look at what Social CRM means. Last part first:  Everyone presumes you know what CRM means, but it's actually sort of a messy term. If you look up Customer Relationship Management on Wikipedia, you might think CRM is just another term for "enterprise systems," since it seems to include just about everything companies do with technology.  Perhaps not surprisingly, when I ask companies what they use for CRM, I get three or four or five answers rather than one.

Further down in Wikipedia's description you get to what I think of as CRM:

There are several different approaches to CRM, with different software packages focusing on different aspects. In general, Customer Service, Campaign Management and Sales Force Automation (SFA) form the core of the system (with SFA being the most popular).

Sometimes companies refer generically to CRM systems as "our customer database," but that leaves out the "RM" part.  The idea of CRM is not just to capture customer information, but rather to use it.

Ok, so what is "social"?  Unlike CRM, which predates the web, "social" is something new. Human beings have always been social, of course, but those social interactions never achieved a scale meaningful to business because they were limited by boundaries of time and distance. With the arrival of the web, social networks grew into the millions and hundreds of millions. Suddenly, what people are saying about your company matters in a way it never did before.

For companies, "social" has two pieces: on-site, and off-site. The on-site piece includes all the traditional elements of web communities, including forums, blogs, idea exchanges, and product reviews, with the additional ingredient of reputation, which allows communities to form. The off-site piece, often referred to as "the social web," is the continuously evolving ecosystem of blogs, social networks, and content-sharing sites where people are also talking about your products and services today. All three pieces -- CRM, communities, and social web -- are critical to any company that wants to compete effectively now and in the future.    

Some people wonder whether social elements on websites are still necessary in the age of the social web. The answer is that no social web rivals the concentration of interest you can achieve on your own site. Few people realize how large these on-site web communities have become today. The largest community on Lithium today has more than 6 million registered members.  The average community on the Lithium platform -- and remember that we work with middle-market companies as well as the Fortune 1000 -- has more than 100,000 members. Take a look at the Facebook group for almost any large company, and I think you'll see the difference. Will Facebook, Twitter, and other social networks continue to grow? You better believe it. But so too will web communities. More than 70% of Lithium communities are less than two years old. They are just getting started.

Why is concentration of interest on your site important?  It's because that's where you get your work done. That's where you sell to customers -- even if most of your transactions are still offline. That's also where you keep your customers, by providing the service and support they need to use your products successfully. It's your home field -- it's where you win. It's also where you gather the data that helps you keep winning.

But there's one more advantage, which our Social CRM whitepaper, released today, describes this better than I can. Bringing tens or hundreds of thousands of your customers together in one place amplifies your message, generating waves of influence that travel far beyond your site. On-site and off-site don't compete -- they work together. That's why our Twitter integration is so exciting. Bringing the conversation on Twitter to your site enables your advocates to join in.  It's the only way your activities on the social web can scale.  

Paul Greenberg sums it up nicely when he says:

You also have the integration of social media and community building tools with traditional CRM tools which are providing effective combinations which are leading toward SCRM. I want to emphasize. These are all good tools. They are worthy of any company's consideration. There is just no SCRM suite out there - as of yet or in the near future. Which doesn't matter one iota.

Where is this all going?  Paul see that too, when he says that "co-creation, mutually derived value, is at the core of SCRM."  

I hate to say "I told you so" -- except when I enjoy saying it.  Back in 2002, a customer asked me the same question -- "where is this all going?" -- and I put together the following timeline.  The time frame is a little longer than we're used to seeing in a business context, but I think it's important to think about timing in terms of when every company will be doing this, not just the enlightened few.   

Today we're in a middle phase of this generation-long evolution, which I called "Learning how to listen." It's reflected in all the conversation you hear about "listening platforms," but it goes deeper than that -- it's really the same question of capturing, managing and using data that is the heart of CRM.  As we progress though this phase companies will be better prepared to have the relationship with customers that they will increasingly demand -- a partnership in which they speak, we listen, and every decision we make is informed (though not dictated) by their experience. I'm happy to be playing a small role in helping that partnership come about.

You may not know this, but there are six magic words that make phony social media experts disappear.  As with all such incantations, you have to get the words exactly right.  Here's how it goes:

"Show me five things you've done."

If your expert is still there, feel free to explain what you mean.   You can explain that you'd like to see examples of social media efforts they have completed for five different companies.   You can explain that the companies sponsoring these efforts don't need to be in your industry, but they should be comparable to yours or better in terms of size and brand recognition.  You can explain that these efforts should not be in development, or coming soon: they should be ongoing or completed.   You can explain that these efforts must be successful by at least one measure -- they engaged thousands or hundreds of thousands or (preferably) millions of users.   

If they are still there, and they still can't answer the question, you may need to ask them to leave.  Even magic spells don't work on some of these folks.

But seriously:  it's important to know whether an individual or firm has experience in what works, or just ideas about what might work. 

Just to be clear -- there's certainly nothing wrong with being a pioneer.  And in new disciplines like social media,  experimentation can be good, even necessary.  But you should approach a project very differently -- and invest very differently -- if the project is a wild guess or a sure thing. 

If you don't know the difference just say, "Show me!"

Today we introduced Lithium Insights, a set of analytic solutions focused on enterprise online communities.    The first two solutions in the set are the Community Health Index (or CHI), and Lithium Lifecycle Benchmark Services.    I think these solutions represent a significant leap forward in the way online communities are measured, and I thought I'd give you a little insight into what they are, why they are important, and the road we took to developing them.   I'll also give you a peek at what's coming down the road from our analytics group at Lithium.

The Lithium platform has always had a pretty rich metrics tool inside.   There are currently more than 150 unique metrics in the system - everything from common metrics like posts and page views to relatively more arcane ones like wireless views, or net anonymous blog comments approved.  You can see these for the community as a whole, or for a unique category or element within a category (blog, forum, chat, or ideas).  With a single click you can see them by quarter, month, week, day, or hour, or you can choose custom ranges like last Tuesday to midnight last night.

So - lots of data.  Typically the way we've approached this data is to choose four or five high-level metrics indicative of community health - such as those Heidi Cohen shared in her column this week - and then used the remaining 145 metrics diagnostically.   That is to say, we watched for trends in high-level metrics, and then used more detailed metrics to better understand the trends we saw.   Here's a simple example I like:   In a forum, if you see post count dropping over time, you know you have a problem but you don't really know why, and you don't know what to do about it.  If you then pull data on threads and replies - essentially breaking post count down into those two different kinds of posts - you can see whether the drop is attributable to fewer questions (threads) or fewer answers (replies).   If the former, you probably have a problem with how you are promoting the community, since questions typically come from new users.  If the latter, you probably have a problem with how you are managing superusers, since they often create 30-40% or more of the replies in your community.   Pretty simple, pretty effective. 

But there are a couple of reasons why that breaks down over time.   First, almost all high-level metrics in a community are retrospective in nature.  Registrations, posts, page views, searches, visits, etc., all tell you a lot about yesterday not much about tomorrow.    So you can address problems when you see them, but you can't really predict them or prevent them.    Past performance is no guarantee of future results, as they used to say on Wall Street when there was a Wall Street.

Second, looking at high-level metrics in isolation can be very misleading.   I know from experience, for example, that page views are the most lagging of indicators - they can and do continue to scale long after your community begins to slide into failure. 

Third, high-level metrics become harder to read as a community matures.   Think about it - in the early weeks and months, a steep climb in page views and posts makes it pretty obvious you're succeeding.   But those growth curves tend to flatten over time, and in years 2, 3, 4, and 5, it can be harder to tell whether your community is really healthy or not.

At Lithium, we've been working with communities for more than 10 years.  As a software-as-a-service provider, we've followed all our communities continuously over time, and learned a lot about what a healthy community looks like.  Some of this is quantitative and some of it is frankly intuitive.   So how did we convert this knowledge into a quantitative index of health?

Here's how we did it:  we asked our statistician, Michael Wu, to create a model of community health.   He gathered the best of what we know from our most experienced community experts - I had a role in this - and he tested our knowledge and intuition against real data from our ten-year database of enterprise communities.    Essentially, we played hundreds of games of "Kasparov versus Deep Blue" over the past six months, testing our knowledge against Michael's model.

It wasn't always fun being a Kasparov - I think you can guess I didn't always win - but I learned some amazing things.   One thing I learned is that human beings always overestimate the health of large communities and underestimate the health of small ones.   Another thing I learned was that that quality we associate with communities that "feel like a community" versus pure question-and-answer actually has a statistical indicator.   After going through this exercise, it's clear to me that there's a vast amount yet to be learned - and that good tools for measuring communities are an absolute necessity for any of us working with communities today.

Today we published a whitepaper that shares our statistical model - we call it the Community Health Index, or CHI (rhymes with buy).   In the whitepaper you'll find a description of the model and all the formulas.   Like any statistical effort, it's a bit forbidding for the mathematically challenged (like me), so I'll be exploring simpler ways to describe it on this blog in the coming days and weeks.  But do check it out and let us know what you think.  To see how we're incorporating CHI in our reports, see this morning's blog post by Neil Beam, who led this effort at Lithium.  

We have created CHI as an open standard - the inputs are limited to data available from any community platform - both as an aid to anyone seeking to measure their online community and as a way to open the model up to exploration and critique by others in the field.   I'm sure we'll be tuning the model over time to increase its accuracy and relevance.   But I'm thrilled we've started on this journey, and I hope you'll join us, or simply follow along.

One final note:  community health is critically important, of course.   It is a measure of community success from the user's perspective.    But there are two large areas of measurement that go beyond community health.   One is return on investment - in other words, success from the business perspective.   The second is community insight and sentiment - what are users saying, and how do they feel?    Both are on the roadmap for our Lithium Insights program this year.   As with health, there are measures for both today but there is much, much work to be done to make them more reliable, more precise, and more useful.  You can trust that we will apply the same level of rigor to those efforts as we have with CHI.   

A colleague of mine got an interesting question this week:  what is the benefit for business-to-business (B2B) communities if being open versus being closed?

I often preach the value of open communities in my work with customers.  In my presentation last year at the Community 2.0 Conference, I included lack of openness as one of the "warning signs" that a community effort is at risk of failure (see slide 8).  But it's a good question - can we quantify the benefit?

First, a couple of definitions:  a closed community is one where access is limited, and an open community is one that any visitor can access.  The most common type of closed B2B community is one that restricts access to customers or partners only.   And open B2B community allows any visitor to the website to view the community and join in on the discussion. 

Yesterday I took a look at some metrics for communities running on Lithium, to compare closed B2B communities to open B2B communities.  Lithium powers more than 50 B2B communities today.  A little less than one-third (29%) are closed, and the rest are open.  Communities in these two segments are quite alike in average age -- open B2B communities on Lithium average 1.3 years in age and closed B2B communities average 1.5 years.  There are no discernable trends in terms of industry, company size, product type, or customer profile -- both closed communities and open communities are varied along each of these dimensions.

However, open and closed communities do vary in one important dimension: volume of activity.   The difference is pretty dramatic.  Open communities have ten times more contributions (10.7, to be precise) and greater than thirty times more page views (35.2).   I pulled only these two metrics, but I'd guess we'd find corresponding differences in registrations and visits, respectively.

So what's the benefit of ten times more content?   In a support community, the benefit consists of many more questions and answers, thus building your knowledge base of user-generated content to help your customers.  What's the benefit of thirty times more page views?   If you believe the authors of Groundswell, one out of six views of an answer in your forums may produce one less call to your call center.  How much does it cost you on average to handle a call in your call center?  If you know, do the math.

But there's one other benefit to an open community for B2B that may trump everything else:  because of the relatively small size of the average B2B community as compared to other communities on the web, an open community is much more likely to grow and thrive than a closed community.   Maybe that's why two out of three companies say:  let's open it up!

Paul Greenberg has a terrific article on ZDNet this morning: CRM 2009 - Companies to Watch For - Second Verse, Different Than the First.  Paul has long been waving the "social CRM" banner, and his summary of where CRM vendors stand -- and where they are going -- in 2009 pays close attention to the integration of social elements into the CRM suite.  

There aren't good numbers out there on this, but I'd venture to say that Lithium has the largest base of CRM-integrated communities of any vendor.  While people speculate on when this trend will happen, we've been delivering integrated solutions with Right Now Technologies for more than two years.   So it's nice to see Paul give us a shout-out for all this good work:

They have been long established as the go-to guys for on demand customer service for quite some time. In fact, two months ago, Gartner positioned them as leaders in the E-Services Magic Quadrant, rightfully so. Unlike many of their on demand brethren, with perhaps the exception of Aplicor, they've penetrated the public sector deeply, an incredibly smart move.  They've spent a good deal of time trying to position themselves and direct and redirect strategy. Sometimes too much and with odd glitches. For example, they bought the highly functional SalesNet in 2006 for $9 million (see this Phil Wainewright ZDNET 2006 entry for some solid analysis of the deal) and then tried to call it something that "improved the customer experience." Not one of their better moves. Yet, they've moved to keep a focus, one better directed on customer experience as a core requirement of customer service. More recently, they've wisely begun to integrate some social CRM thinking and a bit of the functionality into their offering. For example, they have a tight partnership with Lithium since 2006 - which gives them the capability to provide threaded discussions inside of company-produced forums and communities.

I can't neglect to mention one other thing about Paul's article -- his attention to Sage Software, a Lithium customer.  Paul likes what Sage is doing these days, and gives credit to Sage CM General Manager Dave van Toor, who he describes rightfully as an innovator and visionary.  Not mentioned in Paul's piece is the fact that Dave has been a strong advocate for community at Sage, through his blog and through many channels both internal and external to Sage.  Good luck to Dave in his new responsibilities for North American operations of SalesLogix and SageCRM.

Last week Jake McKee created a new site, 90-9-1.com, to spur debate on the principle of Participation Inequality best described by Jakob Nielsen in his Alertbox newsletter back in 2006.   I wrote about this back in 2006 as well, and I don't think I'd change anything I said back then.   However, I would add a few things today, particularly in response to McKee's efforts and the responses it has produced so far.

Points to McKee is for stirring this debate, but points off for muddying it at the same time.  Compare McKee's pyramid to Nielsen's:

You'll see that where the segments in Nielsen's pyramid differ by degree (none, some, a lot), McKee's segments differ in kind.  Creators and editors do different things.  In other words, this is not the 90-9-1 framework Nielsen is talking about.  It's a proposal for a new framework.

How well does it work?  Not very well, as it turns out.   As McKee describes them in an audio response to a comment on his blog, creators start with a blank piece of paper, and editors "append something someone else has created."  The creators in McKee's model are the 1% and the editors are the 9%.

Now think of one of the most common types of community on the web: the technical support community.   There are hundreds, maybe thousands of these on the web.   It's probably half of our business at Lithium, and some of our customers have registered millions of users -- and served tens of millions more if you include what Nielsen calls "lurkers" and what McKee calls the "audience." 

Ok, now here's the punch line:  if you talk to anyone who runs one of these communities -- and I recommend you do, because they can tell you a lot about the future world in which every company will be in continuous communication with customers --  they will tell you that the breakdown between creators and editors is exactly the opposite as Jake defines them.  The 1% of users who are most active are incredibly unlikely to ever start a conversation.  What they do is provide answers.  It's the 9% who ask all the questions -- who start with the blank piece of paper.   

Others commenters on McKee's post have correctly pointed out that McKee's pyramid is actually an unholy blend of Nielsen's work and the kind of participation hierarchies that Forrester and Gartner have created.   To help out, Josh Bernoff at Forrester usefully explained how his participation hierarchy (Forrester calls them social technographic profiles) differs from the kind of thing both Nielsen and McKee are aiming at.  (He doesn't highlight McKee's error, but his research has focused more on community users rather than the dynamics of those users within individual communities, so he simply may not know.)

Credit to Bernoff when he says that looking at real data makes 90-9-1 a messier proposition, though not (in my experience) an invalid or unuseful one.  Here are a few things our data at Lithium tells us (using Nielsen's definitions for 90-9-1):

90-9-1 is scale-sensitive.  In larger communities (say, >100K), heavy participants are likely to be many fewer than 1%.   That may be one reason that YouTube total participant base (that is, the 1% plus the 9%) is 0.16%.   For another data point, Marc Smith's research on Usenet newsgroups found that .05% of Usenet users are heavy contributors.   Alternatively, the number can be higher in smaller communities (say, <200), because of the effects of peer pressure and closer relationships. 

90-9-1 also varies by modality.   It's truer of forums than it is, for example, of video uploading, where the some barriers still exist to participation.  (Not every user has a camera that captures video or knows how to upload videos they create, hard as it may be for us in the geek-o-sphere to believe.)   That may be another reason why YouTube is at 0.16%.

90-9-1 is very handy in planning your outreach efforts.  In other words, it does, as Richard Millington points out, give you a useful way to think about how many people you need to drive to the home page of your community in order to have an active community.  I call it a "planning assumption."

90-9-1 is not necessarily, as Millington suggests, a "huge opportunity."  Not one, if you're running a community, you should bet your job on.  I'm not saying it can't be -- I'm just saying the odds are not in your favor.  

More on this later, I'm sure.

Forrester's Jeremiah Owyang is trying to aggregate site usage stats for social networks -- useful stuff.  

He mentions that his post from back in January on MySpace and Facebook stats gets considerable traffic.  I've noticed (actually, Scott noticed) that a post I did back in December 2006 on the same subject is still one of the most highly-viewed pages on the Lithosphere.  I guess people are just hungry for numbers. 

In fact, that old post of mine is still pretty useful.  It's useful not for the stats it includes -- social networks have grown just a little bit, if you haven't noticed -- but for the reference it makes to the Wikipedia page that tracks such things.   And that Wikipedia page is useful not because its stats are particularly up to date, but because the references contain some good sources for current info on social net usage.

Here's an example:  the reference for Twitter is a site called Twitdir.  This site will tell you faithfully how many registered users there are on Twitter.  (Today, the answer is 3,328,420.)   I say faithfully because Twitdir uses Twitter's API to compile its directory.

If you're a real Twitter stats junky, you might also enjoy Twitter Facts: Facts and Opinions on Twitter and the Blogosphere.

Have a good, current stat?  Go over and help Jeremiah with his project.

I'm attending Defrag in Denver this week and enjoying the opportunity to listen and learn.  One thing I thought you might be interested to see is a new version of the participation hierarchy that Charlene Li and Josh Bernoff presented in Groundswell.  Used to be a ladder, now it's a pyramid.  

Groundswell version:

Charlene's new version, as presented at Defrag today:

Li's full presentation is on Slideshare - enjoy.

ADDITION 11/7/08

I remembered that Jeremiah Owyang posted Gartner's version of the above in a blog post back in August:

Jeremiah's post also generated a long and interesting comment thread -- go check it out.

One of the great things about being a software-as-a-service company is all the data we have about community performance.  We have the very first message ever contributed to a Lithium customer community almost ten years ago, and complete performance data for every community since then -- which today means hundreds of communities covering a wide range of types, size, and age.   I can look at performance across the entire network of Lithium communities (almost 8MM users a day), or I can compose a group with similar characteristics and do some benchmarking. 

Looking across performance data from many communities allows you to see commonalities that aren't otherwise visible.  I've prepared a simple test below to illustrate.   

Four of the most common metrics that companies measure in their communities are registrations, page views, contributions, and searches.  These metrics provide a balance between passive and active participation and allow you to see -- albeit at a very high level -- the relative health of the community over time.    

Interestingly, if you plot each of these metrics over one year's time, the trendline for each metric has its own distinctive shape.   In fact, if you spend a lot of time with community metrics, you might be able to identify which metric is charted on a particular graph solely by the shape of the curve.   

Let's give it a try.

Here are four charts representing each of the metrics I identified above.  In the comments section, I'd like you to tell me which chart shows which metric, and explain what it is about the trendline that led you to that conclusion.  I'll respond later today with the answers.