The 1/9/90 Rule of UGC & Why It’s OK to Have Lurkers

Posted on Aug 24, 2010 | 24 comments


Two days ago I wrote about Quora.  I’ve been loving the product even if it sucks up some of my time.  I prefer my time go into a very focused Q&A website than into a more generic Facebook.  What Quora has done is wrap social networking around Q&A with the more clever next-gen UX I’ve seen.

But my post today is not about Quora, it’s about an answer that I wrote on Quora.  I stumbled upon the Question, “What-is-the-motivation-behind-writing-public-reviews-and-tips-as-on-Amazon-Yelp-Foursquare-StickyBits-etc” and took the bait.  I wrote the following answer:

“I have always believed that UGC (user generated content) website users fall into three categories that follow the 1/9/90 rule. 1% = power users, 9% = casual contributors, 90% = lurkers.  We all get benefit regardless of our roles.

1. Power Users – These are the people you’re likely asking about.  They spend inordinate amounts of time contributing to the website.  They might be moderating categories on Wikipedia, writing 100′s of restaurant / bar reviews on Yelp, checking-in and commenting on every Foursquare venue or even writing entire transcriptions of TV shows on ViiKii.net.  Or let’s face it – writing lots of answers on Quora.

These people use these networks for a variety of reasons but it relates to:
- enjoyment from being a creator rather than just a reader
- creation of social status within the organization for having contributed
- rewards or perceived rewards for achieving status (kind of like collecting airline miles)
- self promotion in order to gain status that might either help with future job prospects or to drive traffic to ones website for primary business
- to meet friends / other people that are similarly inclined because they, too, are “power users.”

I tell people who built UGC websites that you really need to cater to the 1% users.  They need to have the right tools, social status, rewards and stickiness to your product because they don’t want to abandon their creation.  You live or die on the power users because they build the most compelling content and help promote your website (because it helps them).

2. Casual Contributors – These people are uninterested in achieving status on your website.  They had a very positive or negative experience and they want to tell the world.  They are passionate about a topic (like this one for me!) and they feel inclined to spend some time contributing.

For casual contributors the system MUST be quick and easy.  They don’t want to figure out how your complicated stuff works.  They don’t want to register for everything and they don’t care about your points or game mechanics.  As you scale your business they are tremendously important because at scale their contributions really add up.

3. Lurkers.  Most UGC sites try to spend time converting lurkers to contributors.  Don’t.  90% of all users will never contribute anything to your company.  They are there to ingest content.

I wish Twitter understood this better.  If they did then they would run marketing campaigns to let users know that “it’s OK to turn up and just consume content. Twitter’s great for that, too.  You don’t ever need to send a Tweet to love Twitter.”  I never understood why they don’t communicate that more broadly because I think most people’s fear of Twitter is that they don’t want to tell the world what they ate for lunch.

Make it fun and easy for lurkers to visit.  They deliver real value to you because however you choose to monetize, lurkers will always be your largest category.”

Then two things happened.  Ming Yeow Ng asked, “Would you consider Facebook a UGC site?”  Great question.  I wrote the following update to my original answer to that question:

Update: Social networking sites have an additional attribute in that they are communication vehicles as well as UGC sites.  Therefore more people contribute as they are communicating in an IM like way with other people rather than looking to contribute community content.  I still think they follow predictable behavior with power users, casual contributors and lurkers.  But perhaps the lurker category is smaller.

And more embarassingly to me Garindra Prahandono gave me this link to an article written by the guru of web design, Jakob Nielsen about the 1/9/90 rule written back in 2006.

I’ve been talking to entrepreneurs about the 1/9/90 rule for ages and didn’t realize that it was “tacit knowledge” that I had picked up through conversations with many people over the years but attributable to Jakob Nielsen.  Although the 1/9/90 and the word “lurker” seem to be attributable to somebody else, my philosophy on building UGC has been original thinking / POV.  That said, it’s scarily similar.  From the Neilsen article I loved this bit:

“How to Overcome Participation Inequality

You can’t.

The first step to dealing with participation inequality is to recognize that it will always be with us. It’s existed in every online community and multi-user service that has ever been studied.

Your only real choice here is in how you shape the inequality curve’s angle. Are you going to have the “usual” 90-9-1 distribution, or the more radical 99-1-0.1 distribution common in some social websites? Can you achieve a more equitable distribution of, say, 80-16-4? (That is, only 80% lurkers, with 16% contributing some and 4% contributing the most.)

Although participation will always be somewhat unequal, there are ways to better equalize it”

So in summary, if you’re going to do a UGC site:

  • Know that there will be three buckets of users
  • Design the product to accomodate needs of all three.
  • Incentives for “power user” plus product features that make tons of iterative contribution easy
  • Easy for one-time contribution without registration or other hassles for “casual contributors” and don’t think you need to convert them all to power users.  You won’t.  Their character and use case is different.  Celebrate their contribution as good enough.
  • Make it easy for “lurkers” to get value out of your website.  This is where your website goes mainstream, addresses “normals” and becomes monetizable.
  • You can convert some people to “casual contributors” to give you a 2/18/80 curve or similar but lurkers will always be lurkers.  Just ask people who receive traffic from Stumble Upon.
  • And if you’re Twitter, LET THEM KNOW IT’S OK TO BE A LURKER!!  Trying to convert everybody to a contributor is counter productive.  If a person feels that they need to contribute to get value out of your site then they’ll probably stay away.  Simply letting me know that most people are lurkers and they will get tons of value out of your product as pure lurkers and there are plenty of other reasons to use Twitter without contributing will encourage more people to come every day.  Nobody wants to join a club where they feel like they’re not supposed to be there.

** Lurker image courtesy of Wine Library TV.  T-shirt can be ordered here:

  • drorm

    We also need to remember that some people are not as comfortable responding as others.

    One simple reason could be that English is not their native language and it takes a lot more effort to post a response, and they're concerned about spelling and grammar mistakes.

    But even if English is your native language, or you're in a forum that's using your native language, plenty of people are just not as comfortable in writing as a lot of the contributors, and so they just listen instead of throwing in their ideas/contribution. Interestingly enough, pleny of the people that are so outgoing online, are pretty shy in person, and vice-versa.

  • http://www.involvd.com/blog nickmartin

    Nice post Mark. I find the UGC user category split a fascinating one and totally agree with your thoughts on Twitter. I hadn't really thought that those commonly heard gripes from people about how rubbish the product is actually not because they're technophobes necessarily but that Twitter doesn't cater for them satisfactorily.

    What are your thoughts around products like Get Satisfaction and User Voice?

    Whilst they are not the core product of the company and are primarily there to support customer and product development there is still a UGC element to them is there not? My feeling is that the split would naturally be more in favour of the contributers. Maybe a 4/36/60

  • http://www.victusspiritus.com/ Mark Essel

    Power users are a proportionally influential yet resource intensive group. They prove that inequality is a fact of life on the web, be it from platforms, content, or hacker skill. Services that remain valuable to the other groups (lurkers / contributors) always end up having to balance features with the power users in mind.

    MMORPGs are a perfect example of this effect on a network. They spend 80% of the time generating fresh content for 1% of the users (in this case dedicated raiders at level cap).

    But more important is the fact that individual users can't always understand the motivation behind other groups. They understand their own group but the others appear incomprehensible.

    “One Percenters” are the most affected by this clustering, because 99% of the rest of the user base has no idea why they go so nutz over a given service.

    You mention:
    - enjoyment from being a creator rather than just a reader
    - creation of social status within the organization for having contributed
    - rewards or perceived rewards for achieving status (kind of like collecting airline miles)
    - self promotion
    & socialization with other “One Percenters”

    I think there may be other psychological reasons for extreme users. People like familiarity and confidence in predicting our surroundings. Extreme users intimately familiarize themselves with a service and actually shape the way more casual and lurkers interact with it.

    Who are the “extreme users” or One Percenters of the investment world? Why do they do what they do? Creating investor value or growing one's own wealth is just the game of making more than x, y, or z after achieving FU money.

  • http://loo.me pescatello

    Great post. Regarding your final part about Facebook, I recently wrote a post on my blog about Facebook's privacy problem. One of the main points of my article is the dynamic between communication sites and content sites. The more communication-oriented you are, the easier it is to spread as it's inherently viral – and the lurkers are participating. The more content-oriented you are, the less likely you are to spread easily – because they are designed for lurkers.

    The interesting thing is that the dynamic flips if you look at this dynamic through a revenue lens. The more content-centric you are, the easier it is to monetize; the more communication-centric, the harder it is to monetize.

  • http://asalesguy.com Keenan

    Celebrate the lurkers and measure the lurkers.

    Companies would value the lurkers more if they could be measured. Once measured a value could be placed on them.

    For instance, is Twitter counting every time a passed link is clicked. Divide the number of passed links by the number of users and you start to create a ratio. Passed links representing Power Users and Casual users, vs. clicks being the lurkers minus the number of passed links.

    Now I know the level of activity of my lurkers, I can do something with this. High Lurker consumption good, low lurker consumption bad.

    Lurkers value is in consuming, know how much they consume and folks may feel better about the lurkers.

  • http://twitter.com/arniesingh Arnie Gullov-Singh

    Do you think open source projects have a similar distribution? ie 1% hardcore developers, 9% contribute the occasional patch, 90% just download and use.

  • Jfinkle

    Facebook would convert a lot more lurkers to contributors if they had spellcheck capability

  • http://www.startupboyo.com/ RichardF

    Great article, thanks for the link to Jakob Nielsen.

    I don't think Twitter really know what they want to be. Which is why they are having trouble appealing to the mainstream.

  • http://notesfromtheninjabunny.tumblr.com/ Emily Merkle

    I agree in the yet-to-be measured value of lurkers, and I am typically a lurker myself unless I feel compelled to add to the conversation (Twitter) or try to spark a debate or reaction (FB). One social media site that has killed lurkers – I speak for myself and all the people who have stopped investing time in the Q/A section and groups – is because of the focus on participation for collecting positive ratings for answer submissions, and even “expert” status. This is a bastardization of the concept of collective learning. Too many LI users are asking mundane questions over and over, or answering 150 queries per session to 1) gain perceived credibility, 2) hype their own business interests, and/or 3) use the exposure for crude self-promotion. Insanity.

  • http://macsmiley.tumblr.com/ MacSmiley

    I don't think Twitter is ignoring lurkers. Just consider it's new service called “Fast Follow”. It lets anyone follow any Twitter user by simply sending a text message to 40404 reading “follow [username].” The follower then get that user’s tweets sent right to their cell phone, no signup required. If that isn't lurker-oriented, I don't know what is.

  • Bad_Brad

    Great post.

    It's worth noting that the same is true of many products and services, not just social networks. Take the airline business. The whole revenue management and rewards program models are set up based on something similar to the 1/9/90 rule.

    But it is a good lesson for those who run social networking sites to realize that there will always be this natural imbalance and that it does have implications on how you treat each group.

  • http://twitter.com/TomorrowIllQuit I'll quit tomorrow

    Hmm, the one

  • http://twitter.com/TomorrowIllQuit I'll quit tomorrow

    Hold on a sec, did you think about the 1 percent rule or you actually read it somewhere else? ;-)
    http://mashable.com/2007/10/28/the-user-generat

  • http://twitter.com/aseidman aseidman

    I ‘ve had multiple conversations with larger companies and startups attempting to differentiate their services (travel, local, etc.) via data. When we start to drill into their data acquisition strategy it boils down to a form of UGC. These are smart product people who understand the 1-9-90 rule and have well-designed services to match. Yet, most of these of these UGC driven services have fallen into the if we build they shall come trap. Sean Parker’s refers to this class of products as the “Field of Dreams Delusion.”
    If your core value is data relying entirely on UGC is a mistake. There is a role for UGC, but betting the farm on it is not a good idea. Even companies like YouTube need to purchase content to fill out categories, improve content quality, and ensure freshness:
    “When YouTube’s sales team bemoaned the tiny supply of Spanish-language videos for it to run advertisements against, YouTube’s Hoffner called up Demand. Within weeks, Demand Studios started issuing Spanish-language assignments” http://www.wired.com/magazine/2009/10/ff_demand
    To be fair some, buying data is both inordinately expensive and usually undifferentiated. Ten other people can go and by the exact same data from the two or three data providers in the category of your choosing. The data provider side of the market needs to catch up with the needs of data services.

  • Christian Hudon

    When you're writing that casual contributors “don’t want to register for everything”, does that include the suggestion that casual contributors shouldn't have to register to one's web site before posting content? I can see how that increases contributions (says this casual contributor to the comments of this blog, who didn't have to register to post this). On the other hand as soon as your site reaches a certain size I'm not sure how practical that would be, no? (I think it would make it harder to manage the site, for one.) Or do you mean something more specific by that quote?

  • http://bpang.posterous.com/ Beatrice Pang

    Mark, this is intuitive but still super helpful to remind our team as we are building a social media business.Thank you!

    I have one thought regarding your comment on Twitter's marketing strategy. I have a slightly different view. I think Twitter was brilliant at its early stage to market it as a tool for microblogging/publishing. You first need great content to attract lurkers. Encouraging everyone to tweet more, though not effective at attracting the true lurkers, did both attract the first two buckets of people disproportionately more and bent the curve a bit by expanding the first two buckets. I myself actually became converted over time from a lurker to a casual contributor and now a more active user. How you design the product, market it, and choose your first user can have a big impact on changing the size of the two buckets. Now that Twitter has achieved in attracting the power users and a significant portion of the casual contributors, it's the right time to go mainstream and change the marketing message to attract lurkers.

  • http://lmframework.com/blog/about David Semeria

    Wow! This was an awesome post most, Mark. Real insight there, thanks.

  • http://giffconstable.com giffc

    catching up on your blog — nice quora response. Totally agree. The basic ratio of creators vs lurkers is remarkably consistent across different kinds of UGC mediums.

  • http://www.ribbonfarm.com Venkat

    This is one area where I wish somebody would right a book-level treatment that traditional company managers read.

    It is the area where I find the MOST miscalibration of expectations. People go “Oh, you had a 100 new registrations last week, but only 2 of them contributed? Your technology is not working.”

    Even the LOWEST thresholds to participation, like Twitter, don't solve these apparent “problems.” Even the simplest way to contribute (a one-bit signal of like/dislike 0/1) will attract a 1:9:90 distribution (or close cousins…). On 'like' buttons, I am in the 9%.

    Businesses just have to get used to 1:9:90 as a basic social environmental condition. Only very special circumstances can break it.

    Venkat

  • http://www.ribbonfarm.com Venkat

    I too, was guilty of the 'tacit knowledge' state (I got the rule from Jack Herrick of WikiHow.com a few years back, but never thought to look if it came from somewhere). So I did some digging beyond Nielsen. There is a ton of good material out there on the 90-9-1 rule, including empirical tests. I made a trail about it. This post from Gaurovnomics shows the range that can occur, based on studies of the Lithium community while this one on Twitter claims that Twitter has 3x the participation. And this academic paper on something called SCOUT claims that it is possible to achieve 33-66-1 through careful design.

    So looks like unlike the law of gravity, this one CAN be broken or bent to some extent.

    Part of the problem seems to be fuzzy statistics and part seems to be the issue of ignoring a lot of control actuators that community managers may have access to under special circumstances.

    That said, I still think 90-9-1 probably holds for general public communities on the Internet beyond a certain size.

    Venkat

  • http://www.hrabaconsulting.com/blog hhotelconsult

    This is brilliant. Please tell me… where is the data? You said that you wrote 90%, but I don't know where that information is from? From Forrester's Groundswell, I remember 70%. Do you have any data? Is it hyperbolic or metaphoric assumption, or is rooted in real info?

    oops.. all that info may be here… http://www.useit.com/alertbox/participation_inequality.html

    sorry I didn't see it, I leave the comment edited if other people missed it too!

  • Giuseppe De Giorgi

    Great post Mark.

    I would point out that we also have a related rule in our sport sharing platform between:
    Organizers = power users
    Players (and sometimes organizers) = casual contributors
    Players (only) = lurkers

  • http://www.hrabaconsulting.com/blog hhotelconsult

    This page has two comment sections. And the oldest comment is from 3 years ago? so odd.

    Not sure why I am posting it here, but your article got me thinking of some big issues with Facebook, and the lurker vs narcissist, and the real interaction happening. So … thank you. Here's my article:

    Narcissism, Brand Pages, and the Challenge of Facebook:
    http://www.hrabaconsulting.com/blog/2010/09/01/facebook-brand-pages-community-interaction-what-do-we-know/

  • brad

    Question, how do you start a UGC site? It’s kind of like a chicken and egg problem. Who wants to come to the site in the first place if there is no content?