Nobody Has A Million Twitter Followers
January 5, 2010
Last week, I wrote a bit about what it's like to be on Twitter's suggested user list. The response to that post has been really gratifying, and I wanted to share a bit of what I've learned, as well some of the more interesting responses.
First, to recap: I had about 18,000 followers of my own back in October, when I got added to the suggested user list. (Let's call these "organic" followers.) If I'd have continued my normal rate of growth, i'd have about 25,000 followers today, but thanks to being on the list, I've got close to 300,000 followers. Surprisingly though, I only get as many retweets and replies as I'd get with my organic number of followers.
I thought at first that maybe the list wasn't valuable to me because I'm not a celebrity; maybe I'm just noise, but could bigger brands find some value by having a large number of followers?
The Results Are In
As I hoped, my initial post about my experiences inspired others on the list to chime in with their findings.
- Creative Commons, despite being a stalwart organization at the intersection of technology and intellectual property, saw no increase in responses after being added to the suggested user list.
- NBC's Today Show is one of the signature brands of broadcast media. But being on Twitter's list? Didn't do anything.
- What about Starbucks, one of the definitive examples of a powerful worldwide brand? Nothing.
I mentioned in my earlier post, that Kim Kardashian is being paid $10,000 a tweet to promote sponsors on her Twitter account. But what are those sponsors paying for? Because, while she clearly has influence over a certain community, and her Twitter page says she has about 2.7 million followers, I think the reality is obvious: Nobody has a million followers on Twitter.
Does that mean Twitter's follower counts are lying? No. Instead, Twitter accounts that have over half a million followers listed actually represent (at most) a few hundred thousand people who've chosen to become organic followers of someone, along with millions who are passively along for the ride. Some of them are inactive users, some are spammers, some just ignore the noise of the accounts that don't interest them, like spam in an email inbox. But they can't count as "followers" in any meaningful sense.
A few people have asked what my goal is in writing about the experience of being on the list, and why I am offering up prizes to encourage asking questions about it. Well, perhaps the best way to articulate it is that I think the list is being used as a useful fiction for distorting the value and promise of this new medium.
The Million Dollar Gift
There are incentives to promoting the fiction of the suggested user list, of course. If I were the brand manager or Chief Marketing Officer for some big company that got on the list, I bet I'd be proudly trumpeting to senior management that "our social media efforts are bringing us thousands of new followers a day on Twitter". Somebody's gonna get a huge bonus for being the beneficiary of an act of random benevolence. Hell, I'm a pretty persuasive guy — if I found the right (i.e. sufficiently desperate) media outlet, I could probably have sold my Twitter account to somebody for half a million dollars. Well, at least I could have until last week.
And the list preserves a certain amount of power and influence for Twitter itself. (Twitter the company, not twitter the medium.) Because, for every one of the organizations i quoted above mentioning how the suggested user list provided them no value, I got a private message from another list member confirming these findings but not wanting to be quoted on the record.
People being afraid to publicly state their opinion about something of little value for fear of antagonizing a particular company is a clear sign of a completely unhealthy dynamic. I don't think the folks at Twitter would retaliate for public criticism by removing people from the list, because Twitter execs are both extremely busy and fairly thick-skinned, but it shows how insecure people feel about having won the follower lottery. (And how pageview-obsessed publishers are: Every entity that was afraid of being removed from the suggested user list is in the business of publishing content online.)
Fact Check
CNN famously reported on Ashton Kutcher beating them to be the first to get a million followers on Twitter; Today's celebrity reporting often includes a mention of a celeb's follower count as a matter of course. But I'm hoping to encourage some skepticism, to provide a basis for fact-checking that demonstrates these pronouncements are inherently suspect. It's a bit like when I worked at a newspaper: Every reporter thought "Well, our circulation is a million copies, that must mean a million people read my column." Facing the reality that only 10,000 of those people read the column, or that perhaps only 1,000 of them were reading the advertisement on the opposite page, forced a useful and important reckoning into some false assumptions that were underpinning that industry's workings.
The truth: Nobody has been able to point me to a single Twitter account that's earned over 250,000 followers on its own. Nobody's been able to point me to a Twitter account on the suggested user list that's gotten favorites, replies, retweets or responses from a larger number. And nobody's been able to demonstrate why the inflated follower count numbers should be used as a measure of anything but the growth in signups to the core Twitter service itself. [Update: I had suspected some popular artist like Nicki Minaj, the Lil Wayne protege who has famously rapped about her Twitter following, might exceed these numbers. As it turns out, the highest organic follower count I've found is from teen pop heartthrob Justin Bieber with over 800,000.]
That leaves an inescapable conclusion. Nobody has a million followers on Twitter. And being on the suggested user list doesn't add value to a Twitter account, regardless of whether you're a regular guy like me, or one of the biggest brands in the world.
Reminder: I'm running a contest for ideas about how to get more data from my being on the suggested user list. I've been running Gina Trapani's smart little Twitter application ThinkTank since before I was added to the suggested user list. As a result, I have an archive of all my followers, tweets and replies going back for months.I'll provide a prize to one random person who suggests an idea of what information we should query from that data set, as well as one random programmer who contributes code to help.
Here's the prizes and how to participate:
- Have a question or specific bit of data that you'd like to know about an account on the Suggested User List? Submit it to Twitter with the hashtag #sulidea and one random person who makes a suggestion will get a $25 Amazon gift certificate.
- If you're a programmer, watch ThinkTank on GitHub, commit any updates you have to the project, and one random person who commits code to the project will win a 500 GB portable hard drive.
I'll be picking winners for both prizes on January 15th.
12:00 PM
|
47 Comments
Leave a comment
- Earlier: Life on the List
- Next: Remembering Brad L. Graham
Anil
Excellent post, as usual.
You've hit the nail pretty much on the head.
A lot of people have bots and passive followers that never engage. A lot of them probably don't even use Twitter any more, so the entire "value" argument that some people are using is a fallacy.
Michele
Back in October I bet that Drew Carey offering to donate $1M of CBS's money to Livestrong wouldnt get him a million followers. The only way to a million would be if he was on the SUL.
I guess being on the SUL wouldnt have gotten him there either.
Thanks for bringing attention to the follower scam
So what you're really saying is that twitter follower counts are about as accurate of a metric to gauge reader level as hit counts were in the late 90s (remember all those link/click/banner sharing rings?).
As someone else (I forget who) pointed out before, follower to list ratio is probably a whole lot better measure of Twitter value and influence.
That is, until Twitter rolls out the "Suggested Users to add to your List List" (SULL) feature...
Anil,
If it helps any, @aplusk linked to one of the blogs I write for a few times in the past month or so.
His click-through rate was 0.14%.
That's about 6,000 clicks. Not bad at all, but definitely indicative of the difference between total Twitter followers (4.2 million) and active Twitter followers (6,000 within 9 hours).
I would love to know if you have the subscription list of your subscriber i.e. whether they keep subscribed to you, but also whether they actively subscribe to new people, who those new friends are (outside-of-twitter friends or #ff suggestion for the SUL), are their accounts private, whether those subscriber update at all, and is it related to your own keywords, whether they retweet, what application those subscriber use (based on their tweets).
All those a pretty obvious data points, but I have tons of suggestions based on what the results are.
Very interesting articles - thanks for posting them. I've been on Twitter for a while and I've seen some people's lists expand enormously, but none of these have been on the SUL.
My interest is in fandom and I see that people such as Wil Wheaton, Felicia Day and Nathan Fillion have gained massive amounts of followers organically. They also have the tendency to crash websites when they share links because it seems that a large number of their followers are engaged.
So my question is does your "Nobody has a million followers" still ring true for these people?
I know that you are looking for metrics to see if the SUL confers any actual advantages (other than more followers). Perhaps it needs to be measured against a number of things:
1) Number of ppl RTing the account*
2) How many lists they were on
3) How many ppl follow links they tweet (using bit.ly or some other link tracker)
This would have to be assessed over time, too (ideall before and after being on the SUL compared to a 'similar' account not on the SUL).
Not every tweet is RT-worthy and some ppl tweet more than others, so this is not going to be a level playing field. But it is my guess that accounts that have grown organically will have a much higher percentage of engaged followers. At the end of the day, in business terms, this is the important thing - how far your reach is so that you can entertain/influence/educate ppl via Twitter.
I would love to see you set up an experiment to see if you could pin down the metrics and show whether being on the SUL is beneficial to reach at all. Like any experiment (esp social science) there are so many variables that finding a definitive answer is almost impossible, but that doesn't mean we can't try.
* The 'how many ppl RT' metric would have been an average because not every Tweet is going to capture the attention, so averaging over every tweet in a week, or out of a standard number of Tweets.
Great post and comments. The question becomes, at least in the organizing world, are we building followers or leaders and broadcasters? (those who re-post)?
I'm interested in the conversation also about building 1,000 true fans as it relates to social change. What does 1,000 or 10,000 true leaders look like?
http://www.kk.org/thetechnium/archives/2008/03/1000_true_fans.php
Marianne Manilov
The Engage Network
co-founder
www.engagenet.org
Insightful article. I believe it's almost impossible to design a metric to create a valuation of the number of followers a user has. I think RTing scale is an ineffective foundation for a metric as well. People RT for various reasons and those reasons often have very little to do with the Tweet itself. Folks often RT without even reading links in the Tweet. Add to this dynamic the growing number of spammers who do nothing but RT in order to seem legitimate and the metric is already skewed.
The most important thing about Twitter is list building. Twitter can be used as the first qualifier to building a personal list of active and engaged fans clients or customers. If you have 1 million followers but only 7% sign up for "more information" you have a clear idea who your active followers are. Twitter is the first step in the process of list building. I don't think it will ever be more than that.
To test this theory I will be conducting several experiments of my own in the next few weeks. Stay tuned
Twitter.com/AngeloBell
This is why I ritually purge my Twitter followers of ne'er-do-wells and non-entities.
Thus �by my math� my current roster of 105 followers (@BartKing) is actually four times that.
I've been noticing that a number of my most recent Twitter followers have 10K+ followers ... and yet when I briefly peruse their tweetstream, their tweets don't strike me as particularly profound or insightful.
Today, I browsed around the tweetstream of one of them, and discovered several references to twitterway.net, which appears to be a pyramid scheme to effectively commoditize Twitter followers (or followe.rs, as it says on the video). Who needs "suggested" lists when you have a service where everyone who joins is automatically followed by everyone else who has joined already?
Thank you for this series and service.
It's funny, but before the suggested users, there was another kind of list which was generating similar, but smaller followers.
Twitter used to order the little icon's of followers by signup date. Which meant that those of us who were friends of twitter, who were early adopters, always showed up first. I got thousands of followers that way. Once they changed it, my 30 or so new followers a day dropped to a much lower number, like 5. Often i'm losing followers on some days.
You, Chris Messina, and everybody in the SF blogging scene, were early adopters. We all got an initial boost because of it. The current way of doing things, ordering by most recent update, i think, is more fair. i guess.
You're 100% correct. The term follower is a little misleading, unless you understand that a 'follower' does not equal a listener.
I could have sworn BNO @breakingnews was never on the SUL. I looked once and they wer enot
Who needs the SUL to have useless followers? I've got around 140 total and maybe 10 ever actually read the people they follow. The rest either signed up, followed, then never came back, or follow too many people to see anything and are just friend whoring for status.
I like the idea of limits on the number of people you can follow. Let's be honest here, nobody who follows 1500 people is actually reading any of it.
Limit that, and all those people who don't actually do anything cease to exist.
Great post and thanks for sharing. You've gained at least one organic because of it.
Felicia Day (@feliciaday on Twitter) has 1.6 million followers -- all organic.
Good point that it is really hard with twitter to determine how many readers you really have. Studies have shown that more than 55% of twitter users have never sent a tweet but are they reading posts? Hubspot reported recently that only 2% of twitter traffic is retweets so probably not a good measure of how many people really read your tweets. Clicking on links might be a better metric. Its unfortunate that Twitter does not have tools to show pro users how many people saw a tweet in their stream.
Not sure if the metrics are accurate here. Twitter's more like a pro football game: there's, say, 1,000 people on the field - players, refs, media, etc - and 110,000 watching, and 90% of them are actually watching the big screen in the stadium more than the actual game on the grass in between beers and supernachos.
That's what we Americans do best: spectate. We like to watch, not so much to engage. Plus, it's not like we all have something to say, or know how to write...
And that's why I disagree with the premise here: lots of folks have millions of actual followers - AKA, watchers - whether they choose to retweet or mention or tweet their lunch menu or not.
tw-frank1569
Not sure if the metrics are accurate here. Twitter's more like a pro football game: there's, say, 1,000 people on the field - players, refs, media, etc - and 110,000 watching, and 90% of them are actually watching the big screen in the stadium more than the actual game on the grass in between beers and supernachos.
That's what we Americans do best: spectate. We like to watch, not so much to engage. Plus, it's not like we all have something to say, or know how to write...
And that's why I disagree with the premise here: lots of folks have millions of actual followers - AKA, watchers - whether they choose to retweet or mention or tweet their lunch menu or not.
tw-frank1569
The principles are always basically the same. It's pretty easy to pump up numbers, but without targeted engagement, it's no different from getting 1,000,000 clicks through to a website, with a high bounce rate or a near-zero conversion rate to revenues.
Not that you should use Twitter to flog stuff, but you see the point.
These ratios are always basically similar.
I never used to believe these folks who had 500,000 email newsletter subscribers, either. A handful really do, the rest are just inflated.
Cheers...
@andrew_goodman
I run a fashion twitter account that grows 30-60 new followers a day and have over 10k followers (a lot for my niche). I have seen very little incremental click throughs and other obvious metrics that have been pointed out in the article and comments section as my following grows.
I have been thinking about this for the last couple months...
Another possible metric is if your tweet is viewed or not.
Just because it doesn't get "RT'd", added as a favorite, or the link isn't clicked on, doesn't mean you are ignored. With everyone following hundreds and thousands it is difficult for a follower to have an action for each tweet.
The only way I can think of is to somehow track tweet impressions. It wouldn't be too hard through the twitter web interface.
Twitter would just track if a tweet was between the top of the current conversations and the "more" button at the bottom of the followers conversation page.
But would still be very difficult to track on other platforms.
I think there has been a paradigm shift within the Twitter community that roughly coincides with the implementation of the S.U.L. I do not think that there is a causal relationship between these two things, rather that they are both the result of Twitter's rapid increase in popularity.
The shift I see is that "organic" users seem to be more interested in the conversational aspect of Twitter, where the newer "non-organic" users are more into the consumption of tweets. People are not following Ashton Kutcher, Kim Kardashian and Britney Spears because they think they are having a conversation. Rather, they are the equivalent of tabloid reading fans, satisfying some voyeuristic urge. They are not replying or retweeting, since that is the action of a conversationalist.
So, I think it is VERY POSSIBLE that some people have a million followers, despite the increase in replies/retweets. But these are a different type of follower than the "organic" follower, and you should not expect the same behavior from them.
Just because you're not getting more RTs doesn't mean no one is listening.
I follow @wilw and enjoy many of the things he says, but I rarely RT. Why? Because everyone I know is following him. I'd just be spamming.
I use RT when I want to share something that I don't think my followers would have seen on their own.
The other factor is that such large users are usually not very participatory. A business could afford to have a whole team watching and replying to tweets, but on the few times that I've had something to say back to a celeb I've never been one of the lucky few that get a reply. So I don't reply, I consume. I *am* consuming though.
One thing in your post screamed out at me:
"People being afraid to publicly state their opinion about something of little value for fear of antagonizing a particular company is a clear sign of a completely unhealthy dynamic."
The problem is, these days potential employers will look up your online persona -- Facebook, Twitter, blogs, whatever -- and judge you on what you post. If you post a critical message about HP, for example, you can very likely kiss good-bye any chance of ever getting hired by HP. The complaint may be legitimate, something they've heard about and corrected, but you can bank on the fact that hiring managers are VERY conscious of the fact that this person or that made a public and non-positive comment about the company.
So yes, some people will naturally be shy of posting anything critical of a company, because down the road some time that person may want to WORK for that company, and will forever be denied that opportunity.
Sorry, but that's just the way life is in this world today.
//Steve//
I am so glad you wrote this. I was skeptical of someone I know when they were added to the suggested user list and started to gain a huge following. They tweeted a link to my blog and I only saw a dozen click-throughs. I knew they were either spammers or one-time users...either way they were not, as you say, real *followers*.
Thanks for these thoughts and points, Anil and everyone else.
If I could add a subtitle to this article, it would be, "... And That's OK."
Nobody needs a million Twitter followers.
(At least, not yet ...)
The font on your blog post is so fucking annoying. I'll never read your blog again because of it. WTF
Yeah, my wife said that, too. Should be fixed now.
I am the Online Producer for 106 & Park, BET's music video countdown show.
We have 877,168 Twitter followers and counting, without being on the SUL.
I think in another 3 months we will pass 1 million absolutely earned followers.
See our account here: http://twitter.com/106andpark
...and please follow us! :)
Do you we need any more proof that Twitter's effectiveness is NOT going to be measured by how many followers you have?
Twitter is best as a listening tool.
Respond to people and add value to their lives wherever you can.
Listen and Love.. it's as simple as that
Didn't Kutcher do his race with CNN before the SUL? Those folks had to choose to follow. Doesn't take away from the central point (I was surprised early in '09 by how little traffic came through RTs from some of the majors), but it does make it a little less sensational. If I'm wrong about the race, sorry, it's late. I've enjoyed your posts on this issue, perhaps with a dose of schadenfreude.
Says who, you aren't a celeb? You're part of the twitter-created nouveau-celeb elite, don't you know? Your apparent discomfort with this status is admirable and your attempts at coming to grips with it and trying to demystify it are quite exemplary.
However (you knew that was coming, didn't you), whatever metrics you're using to guage "effective" followership will still yield your pre-SUL stats mainly because you've done your best to stay true to who you are. You keep your tweets mostly tech, I imagine, so those that followed and/or interacted pre-SUL will continue to do so (and do so at the same rate unless, of course, the nature of your tweets changed drastically). If you were to try something nice and low like tweeting a pic of you and Tiger Woods (there I said it) having lunch with some swimsuit models, say about a month ago, it would shock you how many hits, RTs, or @mentions you'd get. Your ability to distribute "news" (or just influence stuff) has grown in multiples post-SUL, whether you admit it to yourself or not. Mention @anyone in a couple of tweets at twitter rush hour and I bet they'd pick up at least 10 followers.
Personally, I think it speaks very well of you that you're trying to "tear down this wall" (re @davewiner) and use your "powers" for good.
I'm @udeme, you should follow me :)
QQ: Does anyone else have a slight issue with the term "follower" or is it just me?? ...also, Anil, why dont you have a "sign in with twitter" option? odd.
Not even @tonyrobbins ?
Interesting article. The idea of twitter is that by following people and viewing that time line, you understand what's going on 'upstream'. i.e. What your follows are saying and doing. Very few people understand what is going on 'downstream. i.e. What your followers are doing. Unless you follow your followers or they choose to engage you or you hunt around your followers list and do a little research. Here's some questions:
When did your followers last tweet? i.e. Are they still active?
How often do your followers tweet? i.e. How are active are they?
How many follows do your followers have? i.e. Are they evening listening to you?
How many followers do your followers have? i.e. What's the benefit if they RT?
I think people with tons of followers would be very surprised (and humbled) to know the answers to these questions.
-Steve
Hashim, 106 & Park is on the Suggested User List.
Oh Ev, why did you curse us so?
Was worried that I couldn't get a dollar from each of my followers for my 168 Project film, but it seems that I have other problems that I can't even begin to imagine....
Great article Anil!
I remember reading an article in b-school "Social networks that matters: Twitter under the microscope". It divides your followers into friends and followers and says just because you have a large number of followers does not mean that your ideas will get propagated. It is a quick read and I recommend it.
I think everything comes down to it is not how many friends you have but who are your friends. Quality over quantity always wins.
@feliciaday is on SUL
As far as I know @realjohngreen has never been on the suggested users list and he has over a million followers. He's a YA author who has been very active online, especially on YouTube. See: http://twitterholic.com/realjohngreen/
I think that there should be an algorithm to determine the "real-ness" of a twitter account.
1. It should monitor the number of tweets per day
2. It should monitor the number of new "add's"
3. It should monitor the follow frequency (there is bound to be a pattern for a live human being)
4. Any accounts that don't fit this "profile" should be "warned about their deletion in 15 days". Because let's face it, if you haven't Tweeted in 15 days than you're probably not 'active'
The reason Twitter doesn't want to / Won't do this to clean up their site is they've taken ~160M in VC and need to show INSANE growth, not just good growth, but INSANE growth for that kind of money pre-revenue.
My bet is that the "real" number of twitter accounts is 50-60% of what's reported by them and if those accounts were deleted, you'd see the celebrity accounts get cut by 50% or more, b/c they are the one's attracting more than the fair share of the spam, bots, and less-than-casual users of the Twitter service.
Cool! Keep up the good work and thank you for sharing.
No one seems to be asking the question - is there even such a thing as "organic" followers? I have been active on Twitter since October and I average between 10 and 35 new followers a day. While some of these are "genuine" approaches having come across me through RTS and searches, I KNOW that many of them are following me using a key word auto follow app, or one that follows friends of my friends (like Tweetspinner) - all on auto pilot. These are hardly organic followers surely? They don't know who I am and I would estimate that of my 1500 or so followers, I engage with no more than 50 or so. I sense that the majority of people are just looking for MORE FOLLOWERS and don't give a toss about interaction.
I need to do a full purge to mine the true conversationalists - or find the time to build a list so I can at least filter out the noise!
Anil, I have to agree with you. I have almost 8,000 followers, all with using no automated techniques. They come in at about 10-30 a day and I follow them back. However, I notice little change in actual engagement from days when I had half as many. Mostly I put out my writing tips and interact with those who are interested in interacting with me. I doubt that verymany actually follow everything I say except those who have added me to a list.
it is better knowledge to get