The Biggest Social Graphs
It’s easy to have Facebook on the brain with its mojo over the past year, but Saul Hansell’s post on Inbox 2.0 made me think a bit about where the largest social graphs are.
Here are worldwide UV numbers from Comscore:
I couldn’t figure out how to get comscore numbers on certain services like Yahoo Mail, gmail, and AIM and ICQ, all of which have huge databases of users and the people they know.
Wikipedia says that AIM had 63 million users worldwide in the middle of 2006. I suspect that number has gone up.
And Mike Arrington published this chart in November of 2006:
So, here’s a back of the envelope guess at the largest social graphs on a worldwide basis:
Yahoo Mail – 250mm
Microsoft Hotmail – 200mm
mySpace – 107mm
Facebook – 75mm
AIM – 70mm
gmail – 60mm
AOL Mail – 40mm
These are just guesses and not based on much other than a few numbers I quickly put together and some gut instincts.
The opportunity is certainly there for Yahoo, Microsoft, Google, and AOL to turn their mail and instant messaging services into social nets. But they have to get moving.
Comments (Archived):
Hi Fred, people sometimes seem to forget the massive amounts of e-mail users. I never really understood why e-mail hasn’t been modernised. It has a lot of capabilities but with spam, full inboxes, too many red ! and the lack of handeling multi media and multi streams it is not up to the task. I have written some thoughts on this a while back ago, and I think it is good to see that Yahoo and Google are really looking to increase the social aspect of it. Looking at the number of users the potential is there, now lets see if they can innovate the concept of e-mail into something better.
you should add china’s tencent. per the earnings release today they have 288.7M active IM user accounts and hit 32.6M peak concurrent IM users in Q3. http://biz.yahoo.com/prnews…they also have several hundred million email accounts
The web email systems have the users but have not shown the ability to leverage their user base. This recent Research Recap article shows that social nets in the UK now have more traffic than the big email sites (according to Hitwise) and surprisingly (to me at least) are generating more clicks to etail sites than the email sites. Considering how many emails the big etailers send out (all with click-thru offers), I found that incredibly surprising. The Hitwise charts tell the story.http://www.researchrecap.co…
What about Skype
Fred – In case you didn’t see this chart in yesterday’s Washington Post (http://www.washingtonpost.c…. It cites US only comScore data from September 2007.
That’s a great chartThanks mikeB
Very informative chart. Thanks for sharing.
Fred, these are all great numbers but what are the ACTIVE numbers? i have about 10 email accounts spread across all of the major ‘free’ services that I’ve forgotten about over the years. Do my numbers count in these stats?I’d be much more inclined to go by these numbers if I saw activity rate – and i think they would play out differently
Fred:The problem with this strategy would be the resistance to create new profiles on a new network. I think the recent success of facebook was due to the junk-yard environment of MySpace, and the new “serious” person profile attractiveness. It appealed to many demographics, instead of just the millenia generation. I use facebook, and abhor MySpace, but I would never now move to another platform and create another profile.That said, the Open Social concept of standardized API’s where I can cartry my profile, reallly appeals to me. So when I walk into a very small, niche based social network, it will recognize me and my freinds, likes, dislikes, etc.I would think that this strategy might convince me that I need one FINAL profile, based on Open Social standards, that will come with me forever. So in other words, the new social networks should be highly customized, based on an endless list of interersts that exist out there. If MSN launched one, I woudl resist unless it was the lasty one I ever joined.
I think you are right on Fred about the opportunity to develop social networks around established email platforms. After all, FB and MySpace are in essence large communications environments with unique ways to share information, ideas, etc. Google, Yahoo and Microsoft will all have their plays, however, I think a lot of the real innovation will come from start ups like Xobni. The Xobni guys are onto some big innovative ideas around email and I think they, as do other start ups, have the opportunity to influence the game on how email platforms can extend themselves into the social networking arena.
Fred,good post is it worth adding Skype into the mix too. On my last Skype call all of 10mins ago I noticed that 10.3mm users were logged on at that time. That must give their masters at ebay a huge in on the social scene.Saffron
I guess i missed Arrington’s ComScore numbers from Nov 06, but i am surprised that Gmail is bigger internationally than in the US.substract the US numbers to WW numbers, and you gmail gets 10% of the total webmail accounts compared to 6% in the US.reg the post, the big 4 have had this huge potential for years and even when tried (msn spaces) it doesn’t seem to be taking off…
It would be interesting to see where the cable companies (as a group) who provide email fall on this list. I have a Time Warner Road Runner email address that I can check via webmail as well as a yahoo and a company address.
Fred, three points: (1) If you’re looking at comparable social graphs, you should look at IM. IM represents a “dual handshake” model that is similar, if not exactly the same, as how social networks faciliate friend connections. With email, there is no permissioning system, so I can add any old contact to my contact list. (2) Not only is IM closer in principle to social networks re: establishing friend connections that lead to a social graph, it is larger and growing faster than email. And (3) Comscore absolutes are misleading. I encourage you to look at global user market share to get a more accurate view on the biggest social graphs.
I don’t think we should be thinking of all of these individual graphs. Effectively each one of these services uncovers a little more about what your _real_ underlying social graph is. Email is important because it is both passive (you don’t email someone explicitly to ‘friend’ them) and intimate (emails often contain really important information about the nature of your connection). So when you want to utilise your social graph the information from your inbox adds really important context (but not the be all and end all).A big other category of this social context is photographs. At the moment this is a limited opportunity because tagging is a pretty inefficient way of identifying who’s in what photo and our ability to understand the context / location of a photo is only just emerging. But imagine when you have the perfect storm of face recognition, context identification (party, office, holiday…) and accurate geo-tagging. That’s some pretty powerful augmentation of your connections. Once again, it all comes down to combining these sources. Bring on the middleware.