Implications of the 90-9-1 Rule

By Jake • June 16th, 2009 • 26 Comments

Last week’s post on the 90-9-1 rule was pretty popular. It bounced around Twitter and FriendFeed, and thankfully, Disqus’ Reactions feature allowed me to track comments on it.

So, like any good blogger, I’m going where the traffic is.

The 90-9-1 rule interests me for a number of reasons beyond the obvious applications it has to driving participation in Connect.

It’s an interesting study in psychology, especially when compared to social networks that inject trust into the equation. I really would like to see similar metrics for symmetric (ahem, classic) networks like Facebook and MySpace because of the addition of trust to the social equation.

Browsers don't trust the 'tubes either

This offhand mention in the Harvard Business School study alludes to trust as a motivating factor for participation:

“On a typical online social network, the top 10% of users account for 30% of all production.”

As enterprise adoption of social networks advances, their participation metrics will also be intriguing. Do more people participate when you add another comfy trust blanket on top of the network? I have to believe that just providing a place to share information inside the corporate firewall drives participation, if only because it is sanctioned by the company.This gives people a safety net of security, policy and repercussions that you don’t have on the consumer web.

This is a bigger deal than you’d think. We initially launched Connect in June 2006 with LDAP integration to ensure everyone had an account. This is standard operating procedure within an enterprise and a win over the consumer web. OpenID is a reality in corporate environments by necessity.

Even so, we hit several walls with people trusting us with their credentials, which drove us first to add SSL and then to support Oracle SSO. So, even inside the firewall, people are guarded with their credentials, and rightfully so.

Trust is a weird thing. If the metrics are correct and 70% of people create 90% of the content on typical social networks, this is huge when compared to 90% creating only 10% of tweets and 95% of blogs being essentially abandoned.

What else could it be, if not trust?

In the asymmetric follow model, anyone can follow you and @ reply you, pretty much the same with blogs.

Trust dictates how and if you respond. If I know you and trust you, I respond differently than if I don’t know you. It’s human nature. People trust other people they know. They do not trust the Internet.

Consumer networks create the semi-illusion that what you say and do is among friends, whereas blogs and Twitter do not. At least Twitter users shouldn’t have that illusion, especially after we’ve seen some famously embarrassing fails.

Granted, there are loads of cases of fail on social networks, but the key point for me here is that trust prompts participation. Good judgment is ancillary to this fact.

It’s important to remember that New Web is still in its infancy, and participation doesn’t come overnight, especially after hundreds of years of the old media model, i.e. a few sources broadcast, you consume. The prevalence of email doesn’t help either.

For a large majority, the ‘tubes is a place to consume information, old media style, and read/send email with parties you trust. It’s not anything like that cocktail party we were promised.

This is bad for New Web. If 90% of New Web content is created by a precious few of us, how boring, opinionated and derivative is that?

I want new content and information. I want more opinions and disagreement (ahem, not flame wars), more collective intelligence.

Maybe this is too egalitarian, but that’s the promise of the ‘tubes, right?

So, what to do about it?

I’m not alone in believing that Gen Y will shape the future of the web and drive more people into participation. So, there that. Being a digital native isn’t for everyone, but people will gradually be drawn into New Web by its participants. Once there, they’ll pick and choose what’s for them and what isn’t.

The overall result will be increased participation across all age groups.

The growth of Facebook across older demographics shows this is already happening, so maybe I’m just echoing what is already working. It’s too bad Twitter doesn’t have similar demographics. That would be interesting. Although I did read somewhere that Twitter adoption has been slow among those under 25.

Take that with a grain of salt, since Twitter doesn’t collect age data, making me wonder about any study’s results.

Texting seems to be the accepted mode of operation in younger demographics, which makes sense, and hey, there’s trust in there. Surprise.

Anyway, my theory is that participation begins as imitation.

My usual advice to anyone who’s wary and unsure about how to get started is to jump in and lurk. You can’t expect to start a blog or a social network and follow a bulleted list to success. Start by doing. Read and comment on blogs. Join a social network. Join Twitter. Discover what works and, more importantly, what doesn’t.

Beyond influencing personal behavior, this creates and establishes trust.

So, for me, participation equals trust.

What do you think? I trust you to sound off in the comments.


Possibly Related Posts

  • http://www.oracle.com JordanOAtOracle

    I think the 90-9-1 rule is not about trust but more about two things: cultural conditioning, and time. From our first visit to Romper Room/Barney/Sesame Street/Teletubbies/etc. we are conditioned to be lurkers and watch to learn/be entertained/be contained but *not* participate. So, it takes a lot of emotional inertia to fight years of conditioning to be willing to give one's opinion; even when it is requested without qualification. And secondly, for me at least, I am using the web's resources to help me solve problems in a fairly transactional mode. I really have to carve out and dedicate specific time to participate in the networks that I use to help me in my day-to-day job. Perhaps if our employers, from the top down (as in corporate culture) sanctioned some percentage of everyone's work-life to include social networking than we may see a shift in the 90-9-1 effect; at least in the enterprise.

  • http://theappslab.com Jake

    I've observed that shift over the last two years on Connect, as people begin to find real ways to get work done. Connect exists without top-down mandate, which I think has helped it grow virally, uncovering the good use cases.

    Trust is a key component to pushing the 10% (who create content on blogs, Twitter) up to 30% on social networks. I agree consumption is the default behavior for many, but not so much in younger people, who are no longer as used to it as we are/were.

  • http://empoprise-bi.blogspot.com/ John E. Bredehoft (Empoprises)

    Jake, both you and Jordan seem to have an underlying assumption that people are traditionally consumers, and that participation is now more prevalent than it was in the past. But if we go back to pre-industrial times, there was more participation (with some well-defined limits), such as New England town meetings, Greek democracy, etc.

    I wonder if the consumption culture is more a product of the modern age; it's notable that Jordan cited television as the initial example (ironically, Sesame Street was theoretically supposed to elicit participation in its viewers). If this is the case, then trusted social media environments, rather than being something bold and new, would be a return to our roots.

  • http://www.oracle.com JordanOAtOracle

    John, thanks for the perspective. I definitely see how many aspects of social networking could help us in this regard. The fact that network bandwidth and laptop/desktop hardware have reached the zenith required to do ad-hoc collaborative knowledge sharing may allow us to direct MMU style groups to the enterprise; especially in the context of large distributed companies.
    I also think that sites like elance, odesk, and mechanical turk are great ideas from an organizational perspective to help enterprises find best-of-breed ad-hoc talent using social media; but here is where I think the trust issue really hits home. No matter how well we integrate social media technologies with enterprise players; if they don't inherently trust the resources or infrastructure (not my cloud, I didn't vet the people personally, etc.) the resources will only be niche players in the long run.
    So, I guess it really does come down to trust when it comes to participation between the enterprise and the great unwashed masses (in either direction). Perhaps this is where the SEO and Brand Loyalty folks will really make their money? Helping enterprises engender the kind of trust that's required to elicit ad-hoc/spontaneous participation between a customer/prospect and the organization… regardless of whether that customer is a retailer, partner, supplier, etc.

  • http://theappslab.com Jake

    Good point, and yes, I've been referring to old media in the modern age.

    I watch the advance of government bodies (federal, state and local) with interest, since I think some of them have begun to see the benefits of participation through New Web. This gets at your point about a return to the way it was, on a larger scale.

  • http://empoprise-bi.blogspot.com/ John E. Bredehoft (Empoprises)

    Your comment reminded me of one of the first examples of online participation that I ever encountered.

    I graduated from Reed College in 1983 in the midst of a recession, which made it hard to find a job in Portland (and which is why I eventually came to southern California). When I wasn't working temp jobs, I spent time watching cable TV (go Cubs!). One day the cable channel was airing some type of public affairs show, where people could provide feedback via their cable remotes. However, this was an idea that was ahead of its time; hardly anyone used the remotes to provide the feedback they were requesting. I don't even think that 10% of the viewers responded; perhaps it was more like 1%.

  • http://theappslab.com Jake

    Interesting. Wow, definitely way ahead of the curve. I wonder if that model would succeed today, i.e. w/the remote.

  • http://www.louisgray.com/live/ Louis Gray

    FriendFeed is a very versatile and robust social network. I call it my central nervous system because it takes in all my activity from everywhere, and routes data outward.

    I did discuss the FeedBurner bump last week, as did some solid folks including Rob Diana and Daniel Pritchett.

    You can see my response here:

    FriendFeed Sneaks Into My RSS Stats And Hits The Big Red Button
    http://www.louisgray.com/live/2009/06/friendfee…

    This comment was originally posted on Oracle AppsLab

  • http://theappslab.com/ Jake

    Thanks Louis. Like I said, I’m behind on my reading I didn’t see any mention on Techmeme or TechCrunch and assumed the 3G S launch and Twitter+Iran had buried any coverage.

    I’m wading back in to FF because Twitter has become less useful for me, and I really enjoy the innovation the FF team continues to deploy.

    I agree with the assertion that the FeedBurner count is pretty useless now, and generally speaking, I think we can do better measuring the influence of a blog, which is a general problem with raw number counts.

    This comment was originally posted on Oracle AppsLab

  • http://www.authorityseo.com/ authorityseo

    I have started using friendfeed though after reading your post I think I will give it a try. I am always amazed at what sites like Twitter get picked up by the masses and which ones don’t

    This comment was originally posted on Oracle AppsLab

  • http://empoprise-bi.blogspot.com/ John E. Bredehoft (Empoprises)

    Well, I’m behind on my reading and my writing. I had seen Rob Diana’s and Louis Gray’s comments on inclusion of FriendFeed stats in FeedBurner (for the record, my FeedBurner stats increased by an order of magnitude), but I didn’t have a chance to read your post until now.

    FriendFeed has been my primary destination (currently tempered a bit by my addiction to Facebook’s Farm Town game), primarily due to its conversational aspects. Certainly you can find conversations elsewhere, but FriendFeed clearly lends itself to the conversations – much more than Twitter.

    In terms of reading FriendFeed content, the most powerful feature that they recently introduced is saved searches. I’ve set up several searches, including some work-related ones (mentions of various biometric technologies, mentions of my employer) that allow me to quickly find relevant content without having to subscribe to a bunch of people to possibly find that content. Obviously you can do the same thing in other services (you could save a Twitter search), but I’m finding that it’s becoming much more useful to subscribe to topics than it is to subscribe to people (with some exceptions, of course). Right now the only limitation is the imperfect nature of current search technology (for example, see http://friendfeed.com/oow which captures every occurrence of the three-letter word “oow”).

    FriendFeed also gives me choices as a publisher of content. The multiple channels, rather than making things exponentially louder, actually allows you to exert more control over what you read and what you don’t. Not interested in knowing every song that I listen to via last.fm? Then don’t subscribe to my “lastfmfeeds” group. Not interested in California’s Inland Empire? Then don’t subscribe to my “empoprise-ie” group. When a co-worker recommended that I start looking into natural user interfaces, I set up a room dedicated to that topic, and a few others with that same interest have joined that group.

    A lot of this has been learned over time, through trial and error, and taking advantage of new FriendFeed features as they become available. However, FriendFeed has enough tools to let me tailor it to suit my preferences.

    Is this enough to allow FriendFeed to grow its user base substantially? Frankly, I’m not sure. While the main purpose of the recent changes seem to be to make FriendFeed easier to the eyes for the novice, it is still a much more complex application than Twitter. Complexity isn’t necessarily a barrier to growth – look how complex Facebook is – but I’m still not sure that FriendFeed’s features, while loved by a vocal few (Robert Scoble especially), will necessarily resonate with tens of millions or hundreds of millions of people.

    This comment was originally posted on Oracle AppsLab

  • http://theappslab.com/ Jake

    Wow, thanks for the search tip. That is very useful, too bad it captures a lot of Twitter overlap, but still, nice protip.

    Groups and lists do help a lot to control the noise. I’ve not played with them very much, good reminder.

    I’d like to know how Facebook’s redesign has impacted the casual user, since that would be telling for FriendFeed’s long-term success. I think it’s too early though. FF may be months away from a Twitter-type explosion. Knowing the team, they’ll find ways to make it both useful for n00bs and quiet for early adopters.

    Thanks for your insights.

    This comment was originally posted on Oracle AppsLab

  • http://www.facebook.com/people/Ben-Turner/784110612 Ben Turner

    Why does there have to be one rating? Why not let the users manipulate their own data and create their own formulas (and, of course, share them) to see which metrics are most representative and productive?

    I’m pushing that route for Galapag.us: http://www.facebook.com/pages/Galapagus/1003399…

    This comment was originally posted on Oracle AppsLab

  • http://oraclenerd.com/ chet

    “…met in meat space”

    I like it.

    This comment was originally posted on Oracle AppsLab

  • http://theappslab.com/ Jake

    I’m all for a feedback-based component, but leaving it all to users creates inconsistency, doesn’t it? Plus, generating reputation for people makes them care about it, whereas if it’s all on the users, it’s harder to get started.

    Look at social networks for an example. Did you one day decide to join Friendster, MySpace, Facebook, LinkedIn, Twitter or were you invited? The vast majority of people were invited to join.

    So, I prefer an algorithm that factors in user-to-user kudos style points, but also uses other criteria with equal or higher weighting.

    This comment was originally posted on Oracle AppsLab

  • http://theappslab.com/ Jake

    Wish I could take credit for that. Alas, someone said it to me last week.

    This comment was originally posted on Oracle AppsLab

  • http://oraclenerd.com/ chet

    A follow-on question would be, do users even care?

    We’ve discussed this before, I’m not so sure most people get the whole trust thing. We work in technology therefore it’s pretty high on our list…but my parents (yes, they’re on Facebook)?

    I do try to counsel friends and family about dialing the privacy controls up…

    I find it hard to wrap my mind around at times, the vastness of the ‘tubes (as you would say), and I ponder this fairly often.

    Not really sure if I had a point or not…just not sure if everyone gets the “trust” issue. When you couch it in those terms, I have a hard time getting it.

    This comment was originally posted on Oracle AppsLab

  • http://theappslab.com/ Jake

    Good point. I’m not entirely sure how much typical users care, but I think this is a pretty easy change to describe, i.e. if you choose to share a post with “Everyone”, Google will eventually find it.

    Actually, Google will probably find it and rank it higher than your personal website of family pictures or your online resume.

    The other controls are a bit more difficult to explain.

    The problem is that even within a network of people you know, activity can leak out into other parts of Facebook and now, into search engines. Facebook (and other networks) give the illusion of a closed conversation among friends, which encourages sharing personal stuff, e.g. these pictures of us get wasted at the BBQ, status update that I’m on vacation in Maui, etc.

    We all know these risks, but average FB users do not.

    This comment was originally posted on Oracle AppsLab

  • InternetStrategist

    I consistently find it curious that people feel more comfortably judging people they meet in the physical than they do getting a feel for people online whom they most likely know far more about (unless they’re in the habit of having background checks done on people).

    Con artists in person are far more difficult to detect offline than on because there is a record and inconsistencies are made more obvious.

    This comment was originally posted on Oracle AppsLab

  • http://theappslab.com/ Jake

    Interesting point. I suspect con artists online will evolve over time as the low-hanging fruit disappears, e.g. 419 scammers, and social networks with trust built in will be a natural place to start.

    This comment was originally posted on Oracle AppsLab