eMetrics Day 1 — Let’s Look at the Tweets!
By Tim Wilson on in TweetReach, Twitalyzer, Twitter with 3 Comments
Update: I misstated @johnlovett’s follower count in the initial post. This was a fatigue-driven user error on my end — not bad data coming from either tool employed in this analysis and has been corrected!
Picking up on Michele Hinojosa’s quick analysis of tweets from the first day of the Omniture Summit, I thought I’d take a quick crack at Day 1 of eMetrics. I used TweetReach and a “tracker” (query) I set up a couple of weeks ago for that.
Now, I was a bit short-sighted, in that I set up the tracker on Eastern time. But, we still cover the main bulk of the tweets by selecting March 14th for the analysis range, so I’m not going to lose any sleep over it. The high-level summary:
Let’s take a look at some of the more interesting tweets, as identified using a few different criteria.
Just looking at raw exposure of the tweets, @SocialMedia2Day really dominated with their tweets. Now, @SocialMedia2Day has over 59,000 followers, which means every tweet gets recorded as that many impressions — even before anyone retweets (and there are more followers who might retweet). According to Twitalyzer, @SocialMedia2Day has an effective reach of 175,226, which puts the account in the 98.2nd percentile. The top 3 tweets, just based on raw exposure:
Notice that the top tweet had 10 retweets — 10 people in @socialmedia2day’s network thought it worth repeating. And, it’s a pretty good point content-wise.
@comScore also has a high follower count — more than 24,000, and an effective reach from Twitalyzer of 46,474 (94.3rd percentile). So, after all of the @socialmedia2day tweets comes a list of all of the @comScore tweets. Jumping beyond those as anomalies, of sorts, we get the top tweets by “individual” contributors:
John’s Code of Ethics tweet was retweeted 9 times and garnered almost 30,000 impressions. Nice! We care about acting responsibly! John’s tweet generated its exposure through retweets, as he has around 2,500 people following him…which is a lot of people, but only 1/4 of Ken, who has 10,000 people following him (and he’s following 10,000 people), so his tweets generate ~10,000 impressions just from him tweeting them.
So, looking at raw retweet volume is an indication of how naturally interesting and repeatable a user’s followers (and any followers who retweeted) found the tweet to be. The top retweeted tweet was retweeted 11 times:
Again…a pretty sharp observation.
Shifting around to the top contributors, TweetReach again provides a list based on the exposure generated by each user. The top 35:
We covered that @SocialMedia2Day, @comScore, and @kenburbary have a very high follower count, so let’s take a look at the next two. First, @michelehinojosa, who has just under 1,000 followers, an effective reach in Twitalyzer of 18,852 (89.7th percentila), and tweeted about eMetrics 127 tweets over the course of the day (tweet detail sorted by highest to lowest exposure):
Note the top two tweets were retweeted multiple times…and they’re worth sharing!
And, finally, yours truly — a bit under 1,200 followers, and a Twitalyzer effective reach of ~3,000 (although it jumped up to north of 89,900 starting on March 9th, which is twice what @comScore’s effective reach is, and they have 20X the followers; I need to ping the Twitalyzer folk to help me understand how that happened). My top 5 highest exposure eMetrics tweets for the day:
The second tweet — which was just a humorous observation — was interpreted as a “reply” to @jimsterne…but it showed up as the second-highest exposure tweet. That’s not exactly high-value content — more of a chuckle for those in the room who were watching the #emetrics stream. And, interestingly, I got a direct message from a follower midway through the day that they were unfollowing me as I was clogging their stream. I’m somewhat sensitive to that, but, with tweets being, essentially, public note-taking for me at conferences (and the enticing opportunity to then analyze and summarize those tweets after the conference, so it’s actually shared public note-taking), I suppose I’m okay with that.
Overall, this (very quick) analysis seems to reveal that the most engaging (egad! scary word!) tweets were one that stated, succinctly and eloquently, truths about our profession. I also I would’ve liked to generate a word cloud of all of the tweets (appropriately cleansed)…but that’s simply not as quick and easy as I wish it was!
What do you think?