Twitter Statistics

Earlier this month, Dan Zarrella of HubSpot posted a highly-detailed blog suggesting that 22 tweets a day was the best way, statistically speaking, to use Twitter.  His research included the Twitter habits of more than 1.6 million Twitter users, with emphasis on the habits of the most-followed (or, in Twitter parlance, most popular) users.

Undoubtedly, this “22 tweet” proclamation was then pinned to the bulletin boards of countless marketing offices and pasted into numerous Power Point pitches.  But, as Zarrella pointed out in the blog’s actual title (“Is 22 Tweets-Per-Day the Optimum?”), that number is flawed, because it doesn’t tell you the whole story.  For example:

*  It ignores the content and context of what’s actually in the tweets

*  It doesn’t specify the time of day those tweets were sent — or seen

*  It doesn’t indicate how many people are actually seeing those tweets, vs. the potentiality of the Twitter user’s entire aggregate audience

*  It doesn’t delineate between original tweets and retweets (or “forwarded” tweets from others)

*  It implies that the “most-followed” Twitter users (like Barack Obama, The New York Times and Zappos) are being followed as a result of their frequency of Twitter use, rather than their brand-name attraction.  (Obama’s account, for example, has been dormant since January 19th.)

*  It defines Twitter success solely in terms of followers, without regard for whether one’s audience will take action (in terms of sales, marketing, activism, direct feedback, etc.) based upon that user’s tweets

Unfortunately, marketers, consultants and advertising agencies are so eager to quantify the impact of social media — if for no other reason than to understand what percentage of their bugdet (and time) they think their clients should be investing in it — that they’re looking for any statistical shorthand that can help make their work (and their clients’ decisions) easier.

But just as you can’t quantify a baseball player’s impact on the game based solely on his batting average, you can’t summarize social media’s impact on your business based strictly on a pre-determined criteria of “followers,” “views” or “comments.”  The “22 tweets” results are interesting and, in some cases, may be useful, but when it comes to conversation-based metrics, mathematical statistics only tell one part of the story.

Image by ©aius.


This entry was posted on Wednesday, March 18th, 2009 at 1:30 pm and is filed under Conversation Starters. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
5 Comments so far

  1. Tyler Hurst on March 23, 2009 12:04 pm

    Agreed. It’s crap. Social media, twitter especially, can’t be measure quantitatively. What he found was an average.

    The most followed people on twitter are usually followed by a large amount of people who have no idea what they’re doing, spammers who are auto-followed or just because they are popular.

  2. Terence on March 23, 2009 5:21 pm

    Another consideration: how many people who follow more than, oh say, 50 users, are really absorbing very much of that. I agree CONTENT and CONTEXT are king. Lots of TDD, twitter deficit disorder, out there.

  3. Justin Kownacki on March 24, 2009 9:46 pm

    Everyone uses Twitter for a different reason, so I wouldn’t say that *absorbing* every bit of information is paramount to everyone’s experience. But it is true that the more people you follow, the more messages you’ll need to sift through in order to find the ones that are relevant to you. (Fortunately, there are tools to help you do that, but it’s still an issue of quantity vs. quality.)

  4. Peter Chemisov on March 25, 2009 10:51 am

    Absolutely agreed. That 22 average is a generalization. Meaning, do people really need to be like the most popular users and do they really need to tweet 22 times per day to achieve their goals. Which, this study assumes, is to be like Barrack Obama, Zappos or the NY Times.

    It does, however, underline a certain point. That is, to be popular to any degree, you need to be twitting. Period.

  5. timothy vogel on March 25, 2009 8:15 pm

    In a nutshell, any single metric designed represent a predictably reliable “score” of complex multivariate network behavior is an exercise in futility. And, might I add, almost always the “fruit of a statistically-poisoned tree”, to doctor an old adage.

    The same business world that is dying to make an easy buck in the Social Networking world are the same purveyor of consulting fees and fat-and-sassy contracts designed to promulgate such a naive view of how people actually use the internet in the comportment of their everyday lives.

    The Myth of Statistics? Bologna!

    More like “The Myth of Intelligentsia Originating From the Hubris of Untrained Morons” if you ask me!!!


Name (required)

Email (required)


XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Share your wisdom