Lies, Damned Lies, and Tweetdeck
[Ed. note: Yes, I’m on sabbatical. Yes, I’m incredibly happy. And yes, I’m unplugging much more than usual. So some posts over the next month or two will be fresh and posted on the fly (like this) and some will be ones I’ve got scheduled. I hope you like the mix!]
So this morning I Tweeted about some new Nielsen research on Twitter users.
In a nutshell, Nielsen found that only 40% of new Twitterers stuck around after a month and, comparing that to the retention rates of other social networks, pondered how that could help predict Twitter‘s future.
But then @heatherread clued me in to a VERY important point that the ever-intelligent (and, btw, very cute) @mashable picked up on in this post.
In short: they weren’t counting Tweetdeck.
The longer explanation: Nielsen’s stats don’t seem to take into account one of the fundamental weirdnesses of Twitter as a social network: The more you use it, the less likely you are to use its own website to access your account. You’re going to download one of the many free applications that work better and prettier to get you that information.
This is not the normal way a social network is used, and that seems to be what’s tripped Nielsen up. No matter how much you’re addicted to Facebook (and who isn’t?), you’re still using Facebook.com or Facebook Mobile. There is no third-party app that you’re accessing which is pulling your information to it.
But Twitter’s completely different, because, frankly, its own UI is not nearly as good as those that have been built around it. Using me as an N of 1: I might visit Twitter.com once a month to check something in my archives, but if my computer is on, I’ve got Twhirl open. Ditto Twitterberry if I’m on my Blackberry. (I’m not a Tweetdeck girl.) I suspect most users are the same.
I haven’t seen @nielsenwire respond either way about their methodology, so I can’t say for sure. But the point is, it’s important to notice that the Nielsen stats may not take into account that the logical progression of a user’s experience on this social network is completely different than that for others.
Just another reminder to critically consider what the easy numbers actually mean before you take them as gospel. Apples and oranges are awfully different.
HUGE thanks to the lovely and talented @heatherread for pointing this out.
fantastic post, sarah!! Glad you came off sabbatical for this.
@Steve Awesome to see the clarification! And as you say, even more awesome to see the interaction made possible by social media.
The more social media permits interaction, the more we all have the responsibility to use it to be involved and engaged with the information we seek, not just passive consumers. So cool to see an example of that at work!
Sarah,
We heard the tweets so David Martin did an update on the Twitter Quitter bit with tweetdeck, et. al looped in.
http://blog.nielsen.com/nielsenwire/online_mobile/update-return-of-the-twitter-quitters/
Steve / NielsenWire.com
@Juanlulli – GREAT point. A person’s value on Twitter is what they’re bringing to the conversation. One-time-only or every-now-and-then doesn’t cut it – what matters is what you’re adding to the ongoing stream. That’s inherent to Twitter.
And I think that relates to why there IS – I agree with you – a high rate of attrition. Twitter only works when and if you find people who provide useful content for you. Twitter in itself is valueless. Think of it as a riverbed filled with water. The people you follow are what fill it up. If you find Twitterers who entertain and inform you, then it’s beneficial to your life. If not, then not. But finding the right mix takes time. And I think a lot of people aren’t explained that, and that’s why the attrition rate is so high.
Thanks Sarah. I understand much better.
I’m not a technologist, so I’m left still wondering, if Tweetdeck logs into your Twitter.com user account via your Twitter userID and password everytime you turn on Tweetdeck, why would user activity via Tweetdeck not be captured on Twitter.com.
Oh well…BUT, I have a much larger question. Aren’t we missing the point altogether?
I think one way or another the data will still end up showing that Twitter has high attrition. Question still remains, why does a user have a harder time surviving and thriving in Twitter than in FB.
I have a thought on this. On Facebook, your value — the value of your persona on FB -is in the asset of your Profile. But on Twitter, the value of your persona is in the currency of your conversation and information you bring to the audience. The latter is more difficult. What do you think?
(PS I’ve left similar thinking on http://www.briansolis.com in response to his blog post that you flagged.)
help.
Your Tweetdeck account is set up with your Twitter.com userID and password. Every time you turn on Tweetdeck, it activates your login data on Twitter.com to access your Twitter.com activity and to pull it on the Tweetdeck dashboard. So if it logs into Twitter.com using your userID and password, why wouldn’t user’s daily activity performed through TweetDeck not be “on” or “captured on” twitter.com?
Of well, I don’t understand technology, so I’m not sure if I posed the question clearly enough.
But I do have another *much more* important follow up. Are we not missing the point altogether?
One way or another, I suspect and anticipate, that the numbers *will end up demonstrating* that users do have a harder time surviving and thriving on Twitter than on Facebook or MySpace.
Please do me a favor.
Take a look at my comment (#7) to @briansolis post
https://www.blogger.com/comment.g?blogID=21217704&postID=5962824150948576585
and, while it’s probably too harsh, given what you’ve helped to bring to my attention, does share why I think the numbers will ultimately still show higher attrition rates for Twitter.
@Juanlulli – Good question! Yes, third-party apps do route through Twitter.com – but pulling information from a website on behalf of a user won’t always trip a site counter the way that a user actually logging into a website does.
The question is what stat Nielsen used to come to this conclusion: “more than 60 percent of U.S. Twitter users fail to return the following month”. If they just checked how many people logged in to the website, then it’s inaccurate. But if they were able to see how many accounts were accessed even through a third-party app, then it’s right.
I’m not sure the latter is possible, though. And this question has been brought up in the Nielsen post’s comments, but as of this writing they haven’t addressed it with an update.
Hope that helps clarify!
I don’t understand. Every tweet that is chirped from tweetdeck is routed through the twitter.com network. Otherwise, how is it that every tweet from tweetdeck is captured, processed, and displayed on twitter.com? If this is true, then the Nielsen numbers are 100% accurate. Can you elucidate. @juanlulli