Don't Ignore My Tweets

Sysomos has done some interesting “research” (for lack of a better term), and found, from analyzing 1.2 billion tweets over the past couple of months, that 71% of tweets produce no reaction, 6% are retweeted, and 23% get an @reply.

There are a few things about these “findings” (for lack of a better term), that make me wonder:

1. How others have interpreted the findings. Wired magazine ran an article on the Sysomos “study” (for lack of a better term), running the headline It’s Not Just You: 71 Percent of Tweets Are Ignored. Huffington Post ran a similar article. Only problem here, folks, is that in the Sysomos post on its site, the word “ignored” is never used. Sysomos isn’t saying that 71% of tweets are ignored — simply that they produce no reaction.

2. How Sysomos determines what a reaction is. Apparently, if a tweet isn’t retweeted or replied to, then it gets no “reaction.” If this was a valid measure, then pretty much all of television advertising would be wasted (which it might very well be) unless someone saw a commercial and immediately jumped up to make a phone call to order something or ran out to the store to buy something. Advertisers believe, however, that the ads they run create brand awareness and shape brand perceptions. Is it not possible that tweets can be similar to ads in that regard? If so, then many tweets may very produce a “reaction”, even if we can’t measure it.

3. What the distribution of “no reaction” tweets is by tweeter. What I really would have liked to see from Sysomos is what percentage of tweeters get no reaction to all there tweets, or get no reaction to a majority of their tweets. And on top of that, some segmentation of tweeters by number of tweets or number of followers. Look: If you have one follower, who doesn’t spend a lot of time tweeting, and you tweet 100 times this month, then you shouldn’t be surprised if 100% of your tweets get no reaction.

4. What any of it means. The Sysomos post is chock-full-o-stats. After looking through them, I can’t help but wonder: Yeah, so what?


6 thoughts on “Don't Ignore My Tweets

  1. Hell, I have personal experience of the opposite kind. I am tweeting much less this fall than in the summer. There are good, likely temporary reasons for this, but, my “silence” has generated multiple reactions.

    Usually people just want me to shut up….

  2. Plex: Hmm. So what you’re saying is that the act of NOT tweeting can generate a “reaction”. Now there’s something for the PHDs (pretty hip dudes) at Sysomos to quantify!

  3. The real way to segment it would be to quantify how many tweets call for a reaction.

    A large majority of tweets most likely fall under a “what I am doing right now.” Those are entirely not actionable. Marketers, social media experts, and businesses probably don’t see that as much because of the social circles they reside in. The “average” person isn’t posting interesting tidbits of info and useful links all day long, with the exception of the occasional funny video or kitten pic.

    Something like 25% of tweets contain links. That is an easy way to quantify a large portion of links with a call-to-reaction. The other easy to identify type would be questions. Looking at those two segments to see how many cause a reaction would give a lot more relevant information (though still not very useful).

  4. We’re getting about 23% of our website traffic from Twitter clickthroughs each month — not so much from my own tweets as from various fan tweets. So, branding building aside, what I really care about are clicks. Retweeting and replies are way less important.

  5. I follow several brands on Twitter, usually ones I have a personal attachment to. The tweets that enhance my perception of the brand are the ones where I can observe the brand’s interaction with their users. These are not actionable tweets in that they would require me to retweet, mention or click through but over time they keep the feel good factor going and make me more likely to promote them to my friends over other brands (if all other factors are equal).
    My passive observation of the twitter stream influences quite strongly the referrals and recommendations I make. Unfortunately this isn’t easily quantifiable.

  6. Matthias: Yep, I think you’re spot on there, regarding tweets that don’t warrant a response in the first place. That’s all the more reason why the Sysomos metric is useless.

    Anne: Thanks for commenting. Love the WhichTestWon site. I couldn’t agree with you more — it’s about the clicks. And would’ve thought that Sysomos could’ve measured that, no?

    Treena: Thanks for commenting, and helping prove my point that just because somebody doesn’t RT or reply to a tweet doesn’t mean the tweet didn’t produce a “reaction.”

Comments are closed.