10 Stunning (And Useful) Stats About Twitter
UPDATE: Follow me at @rohitbhargava if you liked this post!
Last month a social media analytics provider named Sysomos released a comprehensive report on Twitter usage. The problem with most analysis on Twitter, though, is that it is limited by the minimal amount of data that Twitter collects. So, to fill the gaps, most reports do things like guessing gender based on real names or pulling data from keywords in people's biographic information. This often yields some questionable results - and the Sysomos report is not immune to this (for example, they find that 65% of Twitter users are under the age of 25, but base this on only the 0.7% of users who actually disclose their age).
Looking past these small points, the report does share some fairly interesting observations and stats as well if you dig a bit deeper. Here's my read on the 10 standout conclusions that the report offers to help you (and your brand) better understand the potential uses of Twitter:
- 21% (One Fifth) of Twitter accounts are empty placeholders. These are the percentage of Twitter accounts that have never posted a single tweet. They may either be registered simply to hold a username for later use, or be experimental accounts started up but never used.
- Nearly 94% of all Twitter accounts have less than 100 followers.
In a finding perhaps consistent with the newness of the tool as well as
the fact that many people may currently have an account simply to start
experimenting with the tool, Sysomos found the vast majority of Twitter
users have an extremely low followership.
- March and April of 2009 were the tipping point for Twitter. During these months, Ashton Kutcher launched his quest to get to 1 million followers faster than CNN, Oprah started using Twitter, and the steady flow of new users to the site continued. For many, it offered a safer and easier way to get their feet wet with social media, 140 characters at a time.
- 150 followers is the magic number. In a particularly interesting data point from the survey, Sysomos found that Twitter users tended to "follow back" all their followers up until about 150 connections. Then the reciprocation rate fell off dramatically, which seems to indicate that this number may be the crossover point where people shift from using Twitter for more personal use to using it more for "lifecasting" their thoughts and actions to a community of people who they feel varying levels of connection to.
- A small minority creates most of the activity. A steep curve of a small minority of actively engaged content creators generating most of the activity on a site is common among social networks, but it is steeper and more pronounced on Twitter. 5% of users account for 75% of all activity, and 10% of users account for 86%. This seems to suggest that the site has managed to engage a mass audience beyond those who typically engage with social media.
- Half of all Twitter users are not "active." If you take a general description of being "active" on Twitter to mean that you have posted a tweet at some point in the last 7 days (1 week), then the survey learned that 50.4% of all Twitter users fit this category. If you remove the 21% from point #1, this leaves about 30% of users who have an account and have tweeted before, but happen to be inactive now.
- Tuesday is the most active Twitter day. One of the most useful data points from the report is that it clears up the common question of which day of the week is the best day to tweet something. Sysomos found that Tuesday stood out as the most popular day for tweets and retweets, followed by Wednesday and then Friday.
- APIs have been the key to Twitter's growth & utility. In terms of tools that people are using for Twitter, Sysomos found that more than half (55%) of all Twitter users use something other than Twitter.com to tweet, search and connect with others. This may, in part, be due to Twitter's notorious reputation of failing/crashing, but also is a credit to all the third party applications that have been built on top of Twitter and do their fair share to bring new users to the service.
- English still dominates Twitter. When exploring Russia as
part of a class that I am teaching this summer at Georgetown, one of
the barriers we learned about was the difficulty of fitting some
Russian language words into just 140 characters. Twitter is, however,
extremely English-friendly. As the Sysomos report found, the top four
countries on Twitter are all English speaking (US, UK, Canada,
Australia). Of these, US makes up 62% of all Twitter users, followed by
UK with nearly 8% and Canada and Australia with 5.7% and 2.8%
respectively. The largest non-English speaking country on Twitter?
Brazil with 2%.
- Twitter is being led by the social media geeks. This particular finding should likely come as no surprise, but 15% of Twitter users who follow more than 2000 people identify themselves as social media marketers. These individuals are more likely to post updates every day (sometimes more than once per day) and also use Twitter more actively for direct communication.
Bonus Geographical Stat/Quote: "The cities with the biggest Twitter populations are New York, Los Angeles, Toronto, San Francisco, and Boston. Los Angeles is the fastest growing city on the list."
Download the full report from Sysomos at http://www.sysomos.com/insidetwitter/
I actually kinda expected these results. Most of my friends on Twitter are barely even tweeting or just there to follow like 5 celebs... then the interest dies off.
The ones that im highly active with are mostly social media marketers or the more tech savvy ones.
Posted by: Nicholas Guan | Monday, July 06, 2009 at 12:44 PM
Not very surprising:)
Posted by: Sohit Karol | Monday, July 06, 2009 at 01:29 PM
Great post! Interesting facts. I responded with my thoughts on our CCG blog: http://capitalcommunicatorsgroup.blogspot.com/2009/07/from-influential-marketing-blog-10.html
Posted by: Robin Ferrier | Monday, July 06, 2009 at 03:33 PM
How fascinating...I had no idea.
Posted by: Phoenix Ad Agency | Monday, July 06, 2009 at 04:15 PM
The country stats (#9) seem doubtful to me because Japan is missing. Twitter staff have stated more than once that they were especially successful in Japan and Japanese is the only language beside English the Twitter UI is available in.
Posted by: Ole Begemann | Monday, July 06, 2009 at 04:57 PM
This is a great post - thanks!
Some of these points aren't that different from the web at large though, and I'd hesitate for people to poo-poo Twitter just after reading these numbers.
For instance, almost every internet innovation (success or failure) has a large number of social media/marketing geeks at the forefront (#10).
Likewise, I bet more than 94% of blogs get less than 100 readers, consistently (#2). Forrester Research would probably back up the idea that a small group of people create the most amount of noise online as well (#5).
It's just interesting, in the context of Twitter, how easily it is for these stats to weave a story that may or may not actually be there.
Posted by: DJ Francis | Monday, July 06, 2009 at 05:15 PM
Wow.. that is incredible information. Thank you for sharing!
Posted by: Tracy | Monday, July 06, 2009 at 08:10 PM
1 a day? I post almost 300/400 a day.
Posted by: ste | Monday, July 06, 2009 at 08:27 PM
I am very surprised to see how many people have signed up and are inactive on Twitter. It's a great forum and lots of sharing.
Posted by: Saferdates | Monday, July 06, 2009 at 09:00 PM
Nice article. It points out alot of good stats about the viability of Twitter. It's getting alot of media backing from numerous TV channels such as "Hit us up on Twitter and let us know what you think!" type on the most popular shows. I hear alot of people using Twitter as a verb, and once that happens; I think things are pretty much here to stay for at least about 5-6 years.
James F.
#1 Twitter Backgrounds Site
Posted by: James F | Monday, July 06, 2009 at 09:11 PM
This is quite interesting facts about twitter. Thanks for the share.
Posted by: Rahul | Monday, July 06, 2009 at 09:53 PM
I'm a teacher and there are loads of us on Twiter sharing resources with each other, and with all the people who are using Twitter to learn English. Tuesday is actually our version of #followFriday, so I wonder if #TeacherTuesday has anything to do with the statistical increase in tweets and retweets on Tuesdays!
Posted by: BCHK_teacher | Monday, July 06, 2009 at 10:40 PM
Who gives a shit?
Posted by: You Know Me, eh? | Monday, July 06, 2009 at 10:46 PM
I found this very informative and helpful. Thanks!
Posted by: Find Tips | Tuesday, July 07, 2009 at 12:02 AM
As my friend Kevin would say, oh my god, you mean another social networking tool that follows a power law?!
Posted by: Ben | Tuesday, July 07, 2009 at 12:38 AM
Very amazing stats, most amazing is 1st point
Posted by: Himanshu | Tuesday, July 07, 2009 at 05:46 AM
"150 followers is the magic number"
That was to be expected, knowing the all-round impact of Dunbar's number. People who are familiar with the idea of the 'monkeysphere' will know that Dunbar's number is a theoretical cognitive limit to the number of people with whom one can maintain stable social relationships. No precise value has been proposed for Dunbar's number, but a commonly cited approximation is 150.
See Wikipedia for more information.
Posted by: Stijn | Tuesday, July 07, 2009 at 06:33 AM
#2 actually misses and important point: many people use Twitter as another way to keep in touch with their friends, and so the fact that they have less than 100 followers doesn't all all mean that all those people aren't using it. The vast majority of my friends use it every day but follow far fewer than 100 people.
Posted by: Reid | Tuesday, July 07, 2009 at 07:57 AM
I tend to agree with DJ Francis "It's just interesting, in the context of Twitter, how easily it is for these stats to weave a story that may or may not actually be there."
What I'd like to know:
1. What happens to the stats if Spammers and the @Oprah factors are removed.
2. How much "nesting" is in the active user community (ie: how many "user groups" are just talking to themselves - over and over again?)
But it is becoming obvious that the key to Twitter monetization will be uncovered once someone figures out how to leverage the 90+% of relatively inactive users.
Posted by: Steve Dodd | Tuesday, July 07, 2009 at 09:00 AM
It's not too surprising and very similar to the reports that David Sifry used to do about the State of the Blogosphere.
The attrition rate is what's interesting - and what are the indicators of long term adopters.
I always used to say to people that if they wanted to stand a chance of getting into my blogroll then they needed to be both active bloggers and blogging for longer than 3 months (because most of the attrition occurred in the first 3 months following sign-up)
I wonder if the three months barrier is the same for Twitter
Also how many of the Twitter accounts could actually be counted as active?
Posted by: Katherine T | Tuesday, July 07, 2009 at 09:28 AM
Interesting stats that I haven't seen before. Twitter certainly is led by a lot of social media geeks and their apps, but it's also led by the celebrities who embrace it, such as aplusk.
Posted by: BIll Hartzer | Tuesday, July 07, 2009 at 09:56 AM
If you are "following" 2,000 people, you aren't following anyone.
I won't follow back anyone who follows more than 200 people unless they really have exceptionally interesting things to say.
Posted by: Tony Lawrence | Tuesday, July 07, 2009 at 10:05 AM
So, in summary; nobody cares. Only the "twitter experts" care about twitter and retweet each other's tweets, talking about tweeter and what to do to get re-tweeted.
It's not really surprising. Twitter is like an atrophied RSS channel, combined with a passive IRC channel where everybody is talking to an empty channel without having a real conversation.
Posted by: JJ | Tuesday, July 07, 2009 at 11:00 AM
You assume that Canada is an all-English speaking country. Canada has two languages.
Posted by: Gray McCarty | Tuesday, July 07, 2009 at 11:10 AM
This is a really great top ten list, extremely useful information. There was a lot of good info on placeholders, active users and other demographics of Twitter users, I was wondering if there is more information about turnover or churn rate. I would imagine that it is pretty high given that almost half of the accounts are inactive and would like see how it compares to other social networks. Anyone can post their own list to our site http://www.toptentopten.com/. The coolest feature is you can let other people vote on the rankings of your list. We are @toptentopten on Twitter.
Posted by: Vince | Tuesday, July 07, 2009 at 11:31 AM
Good info, will be useful for many :)
Posted by: A Maui Blog | Tuesday, July 07, 2009 at 11:38 AM
That is means social media is very very important.
Posted by: alex | Tuesday, July 07, 2009 at 11:41 AM
Great post!
I guess that someday twitter will start to clean up the house, disabling inactive accounts.
Thanks, regards
Antonio
Posted by: Antonio Caldas | Tuesday, July 07, 2009 at 11:47 AM
I guess I should be happy with over 300 followers. But, it still seems like such a minuscule number compared to what I would like to have. But, then again, I am trying to use Twitter to market my business which develops core curriculum resources for school (K-12) using Web 2.0 technology.
Posted by: Andrew Pass | Tuesday, July 07, 2009 at 11:48 AM
"I won't follow back anyone who follows more than 200 people unless they really have exceptionally interesting things to say."
I don't follow back anyone who doesn't have something interesting to say, period.
"everybody is talking to an empty channel"
Not really. But if you don't like Twitter, you are free to ignore it, along with blog posts about it.
Posted by: TXCHL Instructor | Tuesday, July 07, 2009 at 12:12 PM
Very informative article! I've been using Twitter to market my business as well as build up a mailing list in order to expand my online business and found that the relationships developed on Twitter often start off with something as small as "Hey, I noticed you _____ from your Bio. I enjoy doing that, how long have you been doing it?" and this eventually leads into longer convo's and eventually connecting with these people via Skype, it's great how Social Media is allowing us to connect with people like never before.
ps- To learn how you can dominate Twitter head over to http://yourtwitterinfluence.com for a Free 48 Minute Audio interview with a Top Twitter Marketer revealing his secrets to success
Posted by: Chris Hughes | Tuesday, July 07, 2009 at 12:51 PM
Good dialogue, thanks for sparking. I think the missing piece, which is often the case in emergent tech, is objective research of the "how" and "why" of twitter users.
Anecdotally, I know many who are quite happy to 'lurk", to review info sources on a regular basis. So they become non-participants only because we are incorrectly measuring them.
Regarding the 150 rule, I personally found it harder to keep up on evaluations and follows as the base and traffic grew (I am picky who I follow, they need to bring something to the table). This only gets harder as numbers increase.
I too would like to see stats on % of spam, or MLM tweets/tweeps.
David Ian Gray, www.dig360.ca
Posted by: David Ian Gray | Tuesday, July 07, 2009 at 01:36 PM
Social media geeks run twitter? This is no piece of surprising information...the most exciting thing here was the tuesday thing...why tuesday?
Posted by: Astrid Celibataire Endurci | Tuesday, July 07, 2009 at 01:47 PM
Good point that several people leaving comments made is that just because a Twitter account has no tweets doesn't necessarily mean it is inactive. Some people may be planning to use it, but not gotten around to it - while others may simply be creating one in order to read aggregated tweets from their favourite sources.
If anything, the interest in this post demonstrates there certainly does seem like lots of interest in having better stats around Twitter.
Posted by: Rohit | Tuesday, July 07, 2009 at 02:56 PM
With regard to your point #1, perhaps those 21% are those who realise that there is no reason for people to be interested in what they say and so thankfully stay quiet, instead of those who think that people are interested in 140 characters of tripe!
Posted by: Black Katz - rent in London | Tuesday, July 07, 2009 at 04:28 PM
I think the comment about 150 followers being the breaking point is slightly slanted. I think that as a user becomes more popular they start being followed by more and more spammers or people looking for a follow back from popular users.
Personally I have just over 200 followers and I noticed that recently I dont follow back as much because the number of fake, empty or spam-like accounts and followers has greatly increased.
Posted by: Caleb | Tuesday, July 07, 2009 at 07:41 PM
So having over 2000 followers - more than a 1000 more than I follow is not the usual.
Posted by: Robert MacEwan | Tuesday, July 07, 2009 at 09:56 PM
Regarding #7: Tuesday is the most active Twitter day: It's important to remember that your Twitter followers are likely from all around the world, so if you post to Twitter from Los Angeles on Wednesday at 12:00 noon, it's Wednesday 3PM in New York, Wednesday 8PM in London and Thursday 3AM in Beijing. If you're trying to reach a certain geographic region, plan your posts to match high Twitter activity times in their timezone.
Posted by: Twitter Marketing | Wednesday, July 08, 2009 at 02:58 AM
Not surprised about the 150. It is really difficult to keep track of any more than this of the users you follow are active. pop ups every few seconds make following a full time job!
Posted by: Rene Rencontre | Wednesday, July 08, 2009 at 03:45 AM
really great collection of facts. thanks a lot for sharing :)
Posted by: Subodh | Wednesday, July 08, 2009 at 04:52 AM
Blogs are so interactive where we get lots of informative on any topics nice job keep it up !!
Posted by: dissertation sample | Wednesday, July 08, 2009 at 06:09 AM
Interesting, but not surprising. Thanks for sharing though!
Posted by: Tracy | Wednesday, July 08, 2009 at 03:12 PM
#1 misses the people who sign up to follow others but not post. Twitter's replacing RSS in some ways.
Posted by: ceejayoz | Wednesday, July 08, 2009 at 03:36 PM
Very interesting stats, Twitter should systematically delete inactive accounts.
Posted by: Jennifer | Wednesday, July 08, 2009 at 03:39 PM
I guess I am in the top 10 percent. I use Twitter to get my brand name out there by tweeting and retweeting relevant information about the language industry. I do wonder how effective it really is over the long term.
Posted by: English Profi | Wednesday, July 08, 2009 at 03:51 PM
Thanks for the info. I wonder why Tuesday's a favorite though. I'm not surprised about the 150, especially if these people are really focused on following all those. It's hard to keep track of everything when you have a big number.
Posted by: Nate Holland | Friday, July 10, 2009 at 12:35 AM
Hi, I am a Hong Kong student from The Hong Kong Polytechnic University. My major is Marketing and Public Relations. One of my current subject is studying Online PR. My groupmates and I built a blog to talk about Viral Marketing and how it can imply to anti-drug.
Please take a look on our website and comment on our post if there is any ideas, thx!!!
http://viralmarketingspd4290.blogspot.com/
Posted by: Otto | Sunday, July 12, 2009 at 03:27 AM
Great post, it explains the results I get on twitter!
Dr Wright
http://www.wrightplacetv.com
www.twitter.com/drwright1
Posted by: Dr Wright | Monday, July 13, 2009 at 10:54 AM
Great post, very interesting numbers.
Posted by: Jack Zufelt | Monday, July 13, 2009 at 09:02 PM
Having under 100 followers doesn't mean you're "new" or "just experimenting" with twitter. It means you're having personal conversations with people you know instead of being a link spamming new media whore.
Hope this helps.
Posted by: Hamilton-Lovecraft | Monday, July 13, 2009 at 09:22 PM
Nike jordan shoes 10
Posted by: clmm8899 | Wednesday, July 15, 2009 at 10:20 PM
Interesting pie chart. Great post indeed.
Posted by: TJ | Friday, July 17, 2009 at 01:14 PM