The NY Times Magazine has a short piece this weekend called The Facebook Exodus. In it the author, Virginia Heffernan, cites a number of anecdotes about people quitting Facebook. I am sure there are people quitting Facebook.
But the most recent comScore numbers tell another story.
In the month of July 2009, almost 370 million people worldwide visited Facebook, up 155% from July 2008. Facebook is a global juggernaut. It is the fourth most popular website in the world after Google, Microsoft, and Yahoo!
This reminds me of all the Twitter Quitter stories I've read. People do quit Twitter. A lot of them. But here are the July comScore numbers for Twitter.
In the month of July 2009, about 52mm people worldwide visited Twitter, up 1880% over July 2008 and Twitter is the 47th most popular website in the world according to comScore.
Here's the deal, churn is part of online media, particularly social media. People come and go. Some stick around, some don't. These stories about quitters are true of course, but they miss the big picture. More and more people are using these services every day.
Of course, Facebook and Twitter and every other web service out there should pay a lot of attention to why people are quitting their services and address those issues. The best services reduce churn over time. I don't know the exact numbers, but when we first invested in Twitter, it seemed like only 10% of first time users stuck around. Now that number is much higher and is one of the reasons why the service is growing faster.
So when you read these quitter stories, take them with a grain of salt. And make sure to look at the bigger picture at the same time. It may look very different.
Disclosure: For those that don't know, our firm is an investor in Twitter and I am on the board of the Company.
Reminds me of the media’s focus on AOL’s churn during it’s growth phase in the late ’90s.
yup. there’s a bear case to be made for every stock.
try any telecom or switchless reseller in the 90’s – i cut my teeth on churn analytics! its a very important notion (or is it an accepted term? help any MBA)
while i agree with you on the general point, the visitor growth data doesn’t actually refute the issue of lifecycle length.
I agree. There’s no denial more people are visiting twitter.com and using the tool every day–but there’s likewise no denial 60% of new Twitter users quit within a month. It’s logical to conclude as the curve goes up, there’s a corollary curve going down.
that’s all true, but many people quit these services only to come back and find them useful later. that’s what happened to me the first time I used facebook.
THAT IS THE same with me for facebook, twitter, and even etsy
etsy rocks – i just did my new born (my first – stella) room with etsy stuff – the whole experience from discovery to purchase was great – very authentic. big up fred!
Wow. That’s great to hear
i’m still struggling fred. i see it and get it. i dont see the scalable business – i am dying for some one to educate me. i’m not being flippant – i just need help in this particular case i guess.
Fred, do you think as linked data standards like FOAF and RDF mature, that the “walled garden” approach to online communities will give way to open platforms? Platforms like Facebook seem sort of reminiscent of BBS’s and AOL/CompuServe which were ultimately replaced by the web browser. I could see individuals defining profiles, social graphs, etc. and then replacing the proprietary platforms like Facebook with a standards-based semantic social browser.
maybe. some people like the all in one feel of facebook whereas others prefer the more open feel of twitter. i think we’ll move to more open over time, but open is not for everyone
Just like there are multiple definitions of free (beer/speech) there are multiple definitions of open. There is open as in ‘door’ (no access control) or open as in ‘compatible’ (Lego).It perfectly possible to construct a platform which is open (compatible) without being open (door).Just saying.
I can make some really good cases for walled gardens…
Curious to hear more about this. Maybe if you’re the gardener?
You helping me start up or paying me, I can’t give everything away here you know. 😉
Neither. Just making a suggestion
Make ’em on your blog and leave a link here
http://www.shanacarp.com/es…Fyi, I know factually that Lawyers are making mailing lists by bar practice area inside cities informally. So, there is apparently a need for walled gardens…
Good argument. Although it depends on what your definition of open vs closed is. You are arguing for privacy more than anything else. I could imagine a totally open network from an apis/platform perspective that guards privacy closely
Why do you think I want to study here. Long term, I think there is a really desperate need for such products where on the front end you can’t tell. On the back, they are as secure as you can get until the human mistake factor gets in. I also realized before I came here, that most do not look at how people behave with technology or objects and how we are changing with it, or changing technology to adapt to us. I try to only advocate what I sort of see and study.Most people also don’t realize that professional products are premium products. You can’t avoid at some point paying to be part of your professional organization and the tools of its trade, as frustrating as it may be. People will pay eventually as you build your core of dedicated average users, especially if you can professional organizations to sign off. If you build the privacy oriented social networked propitiatory networked trading platform for Goldman, they will pay, and others will want it…same with the best document creation and sharing tool among law firms and their clients, it sort of doesn’t exist.
Its a good area to focus on
Thank you 🙂
dan,in practical use RDF doesn’t scale unless considerable architecting is performed up-front. While RDF looks great in theory, and may work well on a project with a mature data model, it’s still problematic when a data model is evolving (like a social network that may add features rapidly). Also, RDF has some significant performance issues in my experience.I think ontology modeling (like OWL) offer more promise, but require more rigor than a social media company may want to invest, especially at the start.Ontologies work well in science and engineering.I haven’t tracked FOAF in a couple of years, I may need to re-visit.
I think RDF, like anything else, can suffer performance issues depending on the implementation. But part of the value of RDF and related standards like RDFS, and SPARQL is that they actually help to eliminate the data modeling issue altogether. E.g., with a triple-store, you don’t need to redesign a bunch of relational DB tables every time you add or change a feature. I’m just starting to explore this approach as an alternative to relational databases, but it looks really compelling.I can think of a few examples of scalable & high-performance platforms that use this approach, including Thomson Reuters’ OpenCalais, Metaweb’s Freebase (one of the data sources used by Zemanta), and DBpedia to name a few. DBpedia alone stores 218 million triples.If you’re curious, Jamie Taylor and Robert Cook of Metaweb did a great presentation about Freebase at last week’s NY Semantic Web Meetup. Here’s a link:http://www.swnyc.org/index….
I think the reason most people “quit” is because they don’t know how to use the service. Facebook and Twitter do a terrible job at teaching new people how to use the service. New people just get frustrated and leave. I do understand that there are some people who just don’t like Twitter and Facebook and decide to “quit.” That will always happen, but for the most part I think it’s because of a lack of education on Twitter and Facebook’s part.
i totally agree. the twitter team calls this the “out of the box experience” and they’ve been working on addressing it in a number of ways
Being on the border of that- “out of box” makes me a little wary. A good intro might be necessary, but I always like seeing what people might do unexpectedly with products…
i have to agree with shawn. i’ve tried and tried over the last year to get tons of my friends on twitter. they all sign up and within a week quit. every single one commented “i couldn’t figure it out” or “it’s too confusing”. i then changed my approach. “sign up, download tweetdeck and never use the actual website except to block spam and general maintenance.” this has worked. i also made a small instructional step by step PDF that shows my friends how to use TweetDeck. Another great app with miserable “how this works and why this app rules” instructions. Since then my conversion rate has been greater and greater. Twitter still has the most confusing website GUI around. If i wanted to follow 100 different people from different topics, I’d quit also. you’d think a company with a supposed billion-ish “valuation” they’d figure that out.
Tweetdeck is great. Twitter.com is not twitter. Twitter the platform, including all the third party apps
Myke,Is that a PDF you’d mind passing around? Would love to be able to share around as many of my friends and colleagues have the same problem. I tend to send mashables “how to” twitter guide but it doesn’t always have the same impact.
There’s a certain amount of fall off or persistence in almost anything. The media does need to hype things to believe that they have a story. Some simple examples are people buying stadium seating to watch professional sports. ie. At Fenway Park. How many people either lost interest because of financial, health, the time, or different priorities? It’s funny you don’t hear them talking about that. There is limited seating and yet the same people don’t enjoy the view. 🙂 In social media the equivalent is always going up on Facebook and Twitter. It’s strange how they don’t focus on that since it’s easier to see the contrast. I guess they’d have to give up the one sided perspective. It used to be that journalism was comprised of people reporting on things from the different points of view which was more objective.My Two Cents.
I find it odd that Ms. Heffernan chose to only publish comments from people who would support her original thesis, regardless of what the numbers are telling her. Was there not a single person in her circle of friends who could offer a positive use comment?Facebook, for all its flaws, works very well for what it was designed to do: keep people in touch with each other. It worked very well for me when I traveled around the World because I could update my status and upload photos, all the while knowing what my friends and family were up to without having to resort to writing multiple email messages. Now that I’m not traveling so much I may use it less often but it’s still there to do the job I expect to, albeit with a deluge of cause invitation and mafia requests, but those are just a direct reflection of the network I have created, not Facebook’s profile of me.
People also decide to quit smoking, quit drinking, watching tv and so on yet those products still make money year over year. Most of the people that quit can/will come back as long as the product is compelling enough.Service such as twitter and facebook have to become less of a luxury and more of a utility to life. They have to become so useful that you have a better chance to quit driving than facebook
that’s a good goal. i think mobile phones have achieved that. social nets, not yet.
An interesting take, and I do think you’re right that the importance of quitter numbers are overblown. But I also came across an interesting post today called “Back to Digital Reality” that cites some other numbers from Forrester & comScore, and helped bring me back to the current landscape.Search is, of course, still dominating, both in attention and in spend. But more interesting to me was the realization that Myspace is more popular than Twitter. I had to do a double take there. I suppose it’s hard for those of us in silicon alley (or valley) to see on a day to day basis, but Myspace is still quite the force from an attention standpoint, even if not a technology standpoint.We’re still in the expansion phase of social media, where people are discovering it for the first time every day. At some point, that has to end, and social media becomes a mature concept, where every new visitor to Twitter, Facebook, or whatever, is already familiar with the idea of creating a profile and connecting with like-minded individuals.The real question is: at that point, what happens?Does the fragmentation of online persona persist, does a Twitter or Facebook win, does openID or an alternative gain momentum, and does subscriber attrition — quitters — overwhelm new member rates?Back to Digital Reality: http://www.kenburbary.com/2…
good point about myspace.
What about daily users and engagement like time on site for both? I’m impressed by Twitter’s growth, but want to see how it compares to FB.please turn off disqus on the iPhone. It’s hard to post.
facebook is superior in terms of time spent (8 mins to 4.5 mins per visit) and visits per month (21 to 4)one thing to note is many of the power users on twitter use third party apps to access twitter so its not totally an apples to apples comp. but even so, facebook does better on engagement and loyalty.i think the right answer with disqus on the iphone is the make it work better on the iphone, not to turn it off
For those charts above would be really interesting to see it broken out by source: Google/SEO vs. passed links vs. type-in/community traffic.What those charts disguise is any drop-off or slowing in member/community interaction, or rapid growth in SEO traffic.
Fred,I am a bit surprised – usually your articles are quite rigorous but this one is an exception. The graphs that you are showing do not speak at all about the number of people quitting and in fact can easily obscure the longevity and persistance of a new phenomenon such as Facebook or Twitter. In other words, looking at these graphs, it is hard/impossible to determine whether Facebook / Twitter are indeed finding a core group of users who will keep coming back day after day after day. There are other metrics that one can point to to demonstrate that the community is alive and kicking.I am just calling out the fact that you answered a different question or at least used data which do not speak to the question being asked. Now, on the issue of sensationalist articles, such as this one, using anecdotes to overblow an issue – I think you are spot on!
valid point. see the reply to another comment for the time spent and enagement numbers
Yogi Berra said about that restaurant, that it is so crowded, nobody goes there anymore.I cant speak for everybody, nor do I have any idea about the growth curve but personally the service have been disappointing. All I get is spam invites. It is bulky and clunky to use as a communication tool, etc. A telephone book has an address of every person on earth, so what? A telephone book still is only good to sit your child on.
I’d be less concerned about a low number of one-off quitters and more concerned about a growing number of inactive users.
it’s the same thing with these social nets. people just stop using their accounts
I imagine FB & Twitter have their own proprietary metric of at what point inactivity=quitter. Or is it not until the overt action of closing the account = quitter?
Most companies only count someone who has logged in during the past 30 days as active
Great points, Fred. I agree, like they say in PR: “bad press is good press” — that’s what I think about whenever I come across these quitter stories.As mentioned in previous comments here, as the web is rapidly changing I’m sure we’ll increasingly turn to other metrics for measuring traffic. For the real-time web in particular, it would be great to see a data-set quantifying the number of daily status updates over time, respectively in both Twitter and Facebook. Similarly, in other apps like Foursquare, it’s interesting to see the number of “check-ins” a particular venue gets versus unique visitors over time.This short-term quitter phenomenon also reminds me of the AIDA model of marketing (http://bit.ly/YthJo). The users that just come and go quickly are still part of the “Attention” and “Interest” phases; if the network is truly growing organically over time they’ll eventually come back to take real “Action”.
i like that model. it describes the way that i become a user of most services
“This short-term quitter phenomenon also reminds me of the AIDA model of marketing”That reminds me of this: <center><object width=”425″ height=”344″><param name=”movie” value=”http://www.youtube.com/v/y-…”></param><param name=”allowFullScreen” value=”true”></param><param name=”allowscriptaccess” value=”always”></param><embed src=”http://www.youtube.com/v/y-…” type=”application/x-shockwave-flash” allowscriptaccess=”always” allowfullscreen=”true” width=”425″ height=”344″></embed></object></center>
Ah, sorry. I should have known that wouldn’t embed. Here’s the link to the video: Alec Baldwin’s Showstopper on “ABC” and “AIDA” from Glengarry Glen Ross.
Fred, I feel like you’re missing the point. This article isn’t about the big picture, it’s about emerging pockets of resistance against the trend of Facebook. Popularity doesn’t necessarily equal value or meaning.
that may well be true but it may also be true that resistance isfutile and facebook will be as dominant in social media as google isin search
Reminiscent of Yogi Berra who famously said about the Copacabana “Nobody goes there anymore, it’s too crowded.”
Once before mentioned by Fred, at first it is harder to find personal value in Twitter. Figuring out who to follow, getting others to follow you, can all be a challenge which makes people establish accounts but abandon them. It is much easier on Facebook, with the friend suggestions etc. Quicker adoption and perception of value, but perhaps a quick burnout after the initial fuss is over. With Twitter, what I see bringing people back after they’ve signed up is the various affinity groups (Howard’s Stocktwits being I think most successful so far.) It is happening with so many others. The walled garden of Facebook is one approach, and it makes people feel safe. Once the world shrinks (almost like being a part of a fraternity in a large University) it feels much more like the community one feels on Facebook. I notice it happening more and more with Twitter.
the word i picked out in your post is ‘churn’. this is one of the first business dynamics i learned at my first job out of school. I learned it from absolute pro’s. this concept is so important when measuring a business. telecoms of old basically used it as the absolute measure of business performance. of course i might be exaggerating – and a smarter person may site a business school class that examined this – but its been a fascinating and very helpful tool that i have used over the years in many many situations.
I’m starting to think that the real story here is that the NYT and other major media continues to miss the big picture on all sorts of topics. ‘They missed the big picture’ may well be there epitaph.
If that’s true, they’re not alone. I don’t get the enthusiasm about Facebook or Twitter either. Is the drop off rate analogous to that of blogging? I.e., anyone can start, and lots of folks do, and the prospect of connecting and sharing your thoughts with countless people over the Internet is enticing. And then most people find out that no one is listening to them.It seems that active (i.e., roughly daily) blogging is becoming mostly the province of a relative handful of professional bloggers or bloggers who (like Fred) see a value in blogging related to their business. Might the same be happening with Facebook and Twitter?Maybe the general social media analogy is to public access T.V.? That, too, gave regular folks a chance to broadcast their content. For a small handful of them, it led to bigger and better things (I can only think of two, both food-related, off the top of my head — Rachel Ray and Isa Chandra Moskowitz — but I wouldn’t be surprised if there are a few others). But for everyone else, the deafening silence mocking them from the aether caused them to drop off.
If I were the reporter I would be more concerned with the loss of revenue to The Times and The Chronicle than the use of either Twitter and Facebook ! Many of my friends no longer subscribe to the Times, or buy it on Sunday but are using Twitter-for its ease of use-My Accountant in Arizona a Wharton Graduate uses Facebook and Twitter to correspond with his son who lives overseas ! He has his family pictures on F-book-In Michigan we get the same crap from the Old Media-they are like roosters crowing from the top of a dunghill ! HEH
“No one goes to the ballparks anymore because they’re too crowded.” – Yogi BerraAny service like Twitter or Facebook that gets HUGE publicity will have a tremendous amount of people signing up to see what the big deal is. I think we’d all agree that neither service (no service for that matter) is for everyone. But with these two juggernauts, everyone is going to try them. And of course, the people that never would have used it anyway but tried it because of Oprah, well they’ll quit.This is not a sign that these guys have massive problems, its a sign that they are incredibly popular. If the press stopped talking about them for a week, BAM they’d suddenly have a massive increase in retention.What’s staggering about these two companies is the people that really use them, use the sh&t out of them. Like all the time, when they first fire up their laptop and when they finally close it down.
“Facebook is good for finding people, but by now the novelty of that has worn off, and everyone’s been found.” NYTIMESjust as myspace used to be the next big thing and now wasn’t even mentioned in the posting, soon everyone on facebook will have moved on to something cooler and fresher.the true problems with facebook and twitter is the lack of ability to monetize these services so they actually make money for their investors. there is no model of profitability for either service and advertising is NOT the solution
Facebook’s self serve ad platform rocks. I think you are wrong about that one
it may rock–but it does not generate significant income to make facebook a profitable business.if you take away the microsoft money that subsidizes facebook (and now the russian money) you would have a barely profitable (if profitable at all) company.
Yeah, but its scaling rapidly. I don’t think it makes sense to focus on what is as much as what will be
just as myspace used to be the big thing, soon everyone on facebook will have moved on to something cooler and fresher.
I’ll say it once, I’ll say it again – the internet in tech time is like a teenager. If it were human, it’s roughly 16 years old.Wait until it discovers this idea of “adulthood.” It’s going to be complicated and sucky and needing parenting until then, and will you wait out some of the growing pains?
the focus should be more and more on who uses, and how much, the services….it’s about understanding usage habits in order to grow overall social media role and effectiveness
“churn is part of online medial” AND also part of every business that has customers, subscribers etc. The mobile industry religiously manages churn as one of their key metrics.I think it’s funny, and a bit disingenuous when people focus on churn outside of the big picture. The real story is net adds and net gains. It’s similar to a expense/revenue ratio. Are your adds (new subscribers) out gaining your churn, if the answer is yes then your business is growing. Focusing on the fact that you have churn at all is missing the bigger picture.Improve the churn:adds ratio’s and your headed in the right direction. AND ignore those using churn by itself to spin the sky is falling stories. It’s just not accurate.Hurricane Danny ruined my two days of my vacation on the Cape. BUT, my vacation was 10 days long. The 8 days of sun is the real story.
I think there’s a crying need for the ability to deconstruct media and brand narratives and find out where the real motivation comes from. Who’s getting bought and sold. Who’s making money and who’s losing money. Let’s face it, content, including news, is getting paid for by someone. The media’s complicity in the war on Iraq, for example, was predicated on the idea that war is good for the news business AND good for Halliburton. The context in which the narrative unfolds is actually means more than the content. Matt Drudge has access to exactly the same content as a million other online aggregators. It’s the ability to tint his content yellow, and his business relationship with News Corp that point down the path to critical thinking about the Drudge Report.
I want to share a superficial reflexion on twitter/FB with you and see what you all make of it:UGC metrics are more important in twitter than on FB; For example: the number of photos uploaded in FB is less relevant than the number of tweets on a specific topic. Photo pages can barely be monetized nor can they be aggregate it in a larger context where they could eventually be monetized. Tweets, on the contrary can. If you aggregate all tweets regarding the Iran protest you get content that is relevant to users/readers that are not necessarily directly related to the content (I may not be friends with I guy who created the tweet…I may not be Iranian…But I am interested in the news). Moreover, content can be aggregated (a la hash tag) to provide a valuable complement to editorial news feeds..
I’m getting in a little late on this post, but as a 24 year old who started using Facebook early on in college when it was only available to college students, I see Facebook differently now as do many of my peers. As Facebook has expanded access to everyone now, alot of the original college students who used it from the beginning are now using it less and less frequently. Alot have stopped using it at all actually. My general feeling would be that the users that are leaving Facebook are in the 22-26 year old range, and alot of the new users would be in the 40+ yr old age group. These are just my thoughts having talked to my friends about their lack of use of Facebook now, and also having my parents friends tell me that they have just signed up for Facebook and use it alot. Facebook is losing users in their 20’s, and gaining users who are in their 40’s. I’m not sure if thats a trend they would like to see continue-
I think you are right. Even my kids who are late teens and love FB and use it more than any other web service have started blogs and twitters to get more control of their online presence. I would imagine 20 somethings would want that even more. I think some of those 20 somethings end up on tumblr
People aren’t necessarily “quitting” because these sites are doing anything wrong. In fact, in some cases it is because they are doing everything too well, and people find themselves sinking unhealthy amounts of time into their social networks.Perhaps this sort of rehab isn’t what you were initially thinking in the demographic of “quitters,” but this sort of user-on-hiatus definitely exists and usually returns.
My son needs a rehab for his farmville/farmtown habit!
No debating Comscore numbers, but how do those numbers paint an accurate picture in terms of long term sustainable growth and retention? Rather than short term growth and brand equity we ought to ask ourselves how do most if not all SN services build long term customer loyalty. Something “old world” businesses do quite well. Beyond data retention, there is no brand loyalty. Leaving a social network today does not mean a user leaves his or her friends, thus the abandonment of SN-x is a practice widely accepted in our culture.We talked about growth numbers, brand equity, and the sale of MySpace as a social network, now it’s a hybrid music/lifestyle/entertainment portal. Still a major player but I’d be surprised if it weren’t anything more than a marketing arm for NewsCorp a few years form now. Geocities went dark and it barely registered a blip outside of niche tech news reporting sites. And briefly following GC, we had Friendster. Once the cool factor dies and the curiosity wanes, we use comscore numbers to illustrate the quarter to quarter decline in use as users leave in mass.Now it’s Posterous evangelized by Rubel and Scoble and Tumblr which seems to skew to that lucrative 12-24 demographic. And even if you follow Geoff Cook’s recent TC article about higher concentrations of teens on Twitter compared to Facebook it still doesn’t change the fact that most users would leave Twitter in an instant if presented with a viable alternative micro-messaging service. And we’re not far away from that.
I think facebook is the place to look. Its been dominant for five years now and growing more so every day. I never had real friends on myspace. I do on facebook, I do on twitter, and I do on linkedin. I think friend networks are powerful and do build defensibility over time
Facebook has a lifecycle: it started as small core for early adopters, puff out as mainstream users get dragged in by users, and then see growth is tapping down as people get turned off by spam. That’s what happens to Friendster and many other social networks. Facebook is now just a time sucking hangout or I’d say that it’s a new spam heaven. Almost every few hours I’ve been getting messages from my Facebook acquaintances with links to some phishing sites also these messages are not sophisticatedly crafted but randomly drafted. The more such messages, the less useful Facebook becomes as a communications portal.Such a spam free zone is it, isn’t it?
Facebook is not leveling off. Its growing faster than ever
Facebook is not leveling off. Its growing faster than ever
while i think this is an interesting issue to investigate, i also believe there’s some vetting of facts to consider. or at least, a need for the distinction of metrics.while the comscore graph does indeed show a different story–increased usage vs fleeing subs–it is measuring a different factor. the nyt article suggests that subscribers are bailing, not uniques. whereas comscore is showing that more and more uniques are flocking to facebook.the detail to keep in mind here is that it’s hard to draw more than the correlation that uniques = subs. a larger subscriber base on a platform inherently means that the platform draws more interest, even from casual or disinterested browsers. i wonder how many of those uniques came to facebook to see what was going on.. vs join.i think a more interesting chart would be one that tracks a given month’s uniques over a quarterly basis. if the bulk of them are still there after the quarter, then we can assume they’re mostly subs. but if they trail off, then we know what’s going on.my 2c
Journalists should be careful about making generalizations based on a small set of qualitative interviews with users.
I missed this article from the WSJ, dated from August 25th, with the exact same line: http://j.mp/i9ukC