Is #Julia a debacle for Obama?

For a campaign with as strong a reputation as Team Obama has for social media, there’s been some gaffes lately.

You might recall a few weeks ago, we had the #DontDoubleMyRate hash tag.  That worked out OK, with most of the tweets being pro-Obama:

The conservatives tried to (and claimed to) hijack it, but the large number of tweets from college kids in favor of it overwhelmed the opponents.

So far so good.  But only so far…

Then, we have the #forward hashtag.  That didn’t do as well:

Only 25% of the tweets were on this topic.

Now comes the #julia hashtag. My oh my, there was hardly an Obama supporter to be seen:

It’s just not a positive conversation.

Is this a debacle for Obama — is his social media team losing its grip?

No, not really.  The #Julia tag is not an official tag of the campaign.  Rather, it’s a creation of conservatives who want to criticize the Obama campaign.  It’s not surprising that it’s overwhelmingly negative — it’s the invention of negative intentions.

Should pro-Obama supporters have jumped in to hijack the tag for Obama? That’s not for me to say, although I think hashtag pissing contents are at best juvenile.  So I guess I would say who cares…

The one thing this is not, though, is a debacle for Obama.  It’s a bunch of conservatives who like to hear themselves tweet…


I looked at the ~46000 tweets between 4/27 and 5/4 that mentioned #julia.  I took a random sample of 381 of them and scored them as either being conservative, liberal, or who knows.  From that sample, I estimate that 80% +/- 5% of all 46000 tweets were anti-Obama at a 95% confidence level.  (In fact, the confidence is higher because of the strongly anti-Obama response, but I’m too lazy to calculate it at the moment).


Obama Gets a Huge Boost from Jimmy Fallon – Week in Review

On Tuesday night, 4/24/2012. President Barack Obama went on the Jimmy Fallon show, and produced the best ratings in years for Fallon and a big spike in twitter mentions:

Click to Enlarge

We can see the next day Twitter was abuzz.  There’s been criticism of the President from the conservative side of things about the appearance, but it’s clear as a political move it was shrewd.  We’ll see in next week’s analysis whether Saturday’s appearance at the White House Correspondent’s dinner will have the sam effect.

Mitt Romney, on the other hand, got a bit of a boost from his primary victories on Tuesday, but it was no where near what the President saw.  Romney really needs to push harder on Social Media if he’s going to be competitive in the fall.  Fortunately, he has plenty of time to work on it…

Is Romney ahead of the President on Twitter? Do Elephants Fly?

At a closed door fundraising meeting Sunday (4/14), Mitt Romney is reported to have said:

“We are behind when it comes to commentators on TV. They tend to be liberal. Where we are ahead or even is on twitter and on the Internet.”

Is he? That’s an interesting question for several reasons.  First, there’s no single definition of ahead or behind.  Second, because any real metric is a bit hard to calculate.  Third,  Twitter is a fairly volatile place, and so picking the timeframe might influence things one way or another.

Let’s work through these issues and look at what the statistics show us.  Maybe we’ll come to an answer — or maybe we’ll decide it’s too hard to answer a simple question!

First: What is the right definition of ahead?

One definition of “ahead” is a sort of gestalt kind of thing: it does seems like Mitt just had a good twitter week.  That’s not an unreasonable way of looking at things, but it’s also not really something you can prove.  I’m sure Mitt is happy about the recent conflict between Ann Romney and Hilary Rosen. Mrs. Romney got a lot of press and a lot of tweets that may have turned the tide for a while in the war on women debate.  However, taking advantage of your opponent’s gaffs is not really a strategy (unless your opponent is predictably gaff-prone).  A few weeks from now, this will have faded away and be replaced by some other outrage.  We might assume a pro-Romney shift in Twitter, but it could fade.

Another definition of “ahead” might be # of followers.  I’ve seen this mentioned a few times by others.  The problem is it doesn’t really delve into why people are following you.  They may love you and hang on every word.  They may hate you and are looking for you to slip up so they can pounce on it.  I think the combativeness of political discourse means that # of followers is a very tough metric to draw conclusions from; you’re just not going to know a lot about why people are following you.  If Mitt was Tide Soap, followers would probably be a great metric.

You might say you’re ahead based upon the number of tweets you make.  This might be a great metric of how active your social media team is, but is probably not a great metric of how effective you are.  It’s an input to the process, not an output. Again, this is at best a tie breaker.

One definition of “ahead” I think is meaningful is how often you get mentioned in tweets.  It’s not perfect, because if you say or do something that animates your opposition, you’re going to trigger a bunch of tweets even though they are not supporters.  But, over time, the number of mentions is at least  a valid indicator of mindshare.

Another useful definition of “ahead” is how often you get retweeted.  Unfortunately, this is not a great standalone metric — if you are not a heavy tweeter or your tweets are not particularly retweetable, you won’t score well even if you are the darling of the media.

A combination of the last two is useful, though, to see how much of the twitter conversation you dominate.  And it’s what we can go on — it’s hard to draw any other metrics out of Twitter.

So let’s start the scoring!

We’re looking at the accounts @MittRomney, which is the official campaign account for Mitt Romney, and @BarackObama, which is the official campaign account for Barack Obama.  We’re also looking at tweets taged with #Mitt2012 or #Obama2012


  • Mitt Romney: 442,029
  • Barack Obama: 14,275,493

I did say that followers are not a great metric, but you’d have to admit that Obama is way ahead here. You might give Obama an advantage since he joined Twitter in March 2007 and Mitt Romney joined June 2009.  But that’s not much of a difference, although I suspect that Romney only really got active once he announced his campaign which didn’t help.  Still, if you were going by followers, Romney is way behind.

How about by the # of tweets since the start of the year?

  • Mitt Romney: 180
  • Barack Obama: 1063

Barack Obama sends out about 6 times as many tweets as Mitt Romney does.  While I did say that the # of tweets is not an indicator, in itself, of your presence on Twitter, the comparatively fewer tweets from Romney probably trickles down into the statistics which follow:

In terms of mentions — the number of tweets by people that include either @barackobama or #obama2012 for Obama or @mittromney or @mitt2012 for Romney:

  • Mitt Romney: 60,502
  • Barack Obama: 138,881

The President is ahead by a bit over 2 to 1 here.  That’s actually an impressive showing by Romney, given his light tweeting.  But it is still a 2+ to 1 advantages to the President.

In terms of retweets, where somebody has been so moved by what the candidate says that they repeat it to their followers:

  • Mitt Romney 10,006
  • Barack Obama: 34,288

Here the President is 3+ to 1 over Romney.

Even Klout has the President with a higher score than Romney.

Net net (wow, you read this far?):

Romney is way behind on Twitter, his one good week with the Ann vs. Hilary battle notwithstanding.  If he thinks that, he needs to have a metrics based discussion with his social media team. Any viable metric you come up with is going to show him at best 2 to 1 behind and probably far worse.


For the count of # of tweets sent by the candidate, I looked at their tweets between 1/1/2012 and 4/14/2012, inclusive.  For the other metrics, I looked at all tweets sent between 2012-04-10 01:00:29 and 2012-04-16 17:00:35 EDT, a few hours short of a full week’s worth.

Tweetdown: Coke Vs. Pepsi — Coke is It!

Coke vs. Pepsi is one of the longest running, most hard fought battles in the marketplace. So I thought it would be interesting to see what kind of engagement the two brands have on Twitter.  Looking at a week’s worth of tweets, we can see the number of times each brand is referenced:

Coke vs. Pepsi in popularity on Twitter

Click on image to see larger version

And we can see that Coca Cola gets about 75% more references than Pepsi does.  An analysis of the gender of the senders shows that both brands are about evenly split between males and females (and that about 50% of the people tweeting are of unknown gender).

When you look at the sentiment expressed in the tweets, you can see that @cocacola and @pepsi tweeters are generally happy folks:

Sentiment of tweets, coke vs. pepsi

Click on image to enlarge

it is very unusual to see a large selection of tweets have virtually no tweets with a negative sentiment!

Finally, there’s a health balance between retweets and new tweets mentioning both brands:

Tweets vs. Retweets for Coke and Pepsi

Click on image to enlarge

The top 5 retweets for coke are:

total Tweet Text
412 RT @CocaCola: A smile is the best gift you can give (an ice cold Coca-Cola is a close second!)
236 RT @CocaCola: Sometimes a high five is so good you need to high five again to celebrate it!
51 RT @Foodimentary: On this day in 1886 @CocaCola is said to have been invented by @DocPenberton
32 RT @CocaCola: To do’s this weekend: skip, run, & enjoy Spring!
27 RT @CocaCola: Discovering a good tune is loved by all folk.  It’s fun Making_ _ _ _ _With_ _ _

The top 5 retweets for Pepsi are:

total Tweet Text
190 RT @pepsi: Happy Birthday Lady Gaga! We <3 you, mother monster! 26 looks good on you!! (cc: @ladygaga)
98 RT @pepsi: “I stand for freedom of expression, doing what you believe in and going after your dreams.” – Madonna
71 RT @pepsi: When life gives you lemons, ask for a gift receipt and get a Pepsi.
69 RT @pepsi: Let’s make memories tonight. #Cheers
60 RT @pepsi: Happy birthday to Elton John AND Aretha Franklin! #PepsiIcons #Legendary

I think it’s interesting that Coke’s retweets are coke centric, while Pepsi’s tend to mention celebrities.  Clearly a reflection of their marketing efforts.  It’s also interesting that, for Pepsi, virtually the only thing being retweeted are @Pepsi tweets (16 out of the top 20 retweets are from @Pepsi), while for Coke there is a broader mixture of sources (8 out of the top 20).  That generally indicates a more vibrant community for Coke.

Still, when I look at the activity for both brands, there is nonetheless a relatively low level of engagement with the overall Twitter community.  When you look at the engagement that TV personalities get, for example, it is way beyond what these two venerable brands get.

This shows an area which each can gain much more value from Twitter; the value of getting higher levels of engagement is immense.  If you read Robert B. Cialdini’s “Influence: The Psychology of Persuasion“, he talks about how taking small steps to demonstrate commitment can lead people to a deeper commitment.  In the case of Twitter, getting more people to tweet on Coke or Pepsi  will result in not only those people forming a deeper attachment to the brands, but serve as “social proof” for others as well.  It’s all well and good for the brands to tweet out, but it is those tweets back that drive real engagement.


All tweets mentioning @cocacola or @pepsi for a 1 week period ending roughly at 1pm on 3/31/2012 were collected and analyzed. Gender was assigned using a mechanical process that looks at first names and terms used in the description.  Sentiment was scored mechanically by looking at the choice of words and phrases in the tweets. Note that it’s very hard to gauge the sentiment of tweets, and so the large neutral ratings for each brand should only be interpreted as the inherent weakness in mechanical sentiment scoring.

Tweetdown: The O’Reilly Factor vs. The Ed Show

Bill O’Reilly and Ed Schultz go head to head at the 8pm time slot on Fox News and MSNBC.  Earlier, I covered the relative twitter activity of Maddow vs. Hannity, and we saw how Sean Hannity’s followers were much more active on Twitter than Rachel Maddow’s.

I thought it would interesting to take a quick look at Bill O’Reilly vs. Ed Schultz.  Ed implores his viewers to tweet him in a poll every day, and so I was curious if that produced a greater response.

During this week (ending 3/23/2012), you can see the relative activity of each show’s viewers:

Twitter Activity of followers of the O'Reilly Factor vs. The Ed Show

And we can see that Ed’s requests of his viewers is followed.  And that led me to my next question: outside of the time Ed’s show is playing, how active are his followers?  So I added up the number of tweets each received by the hour of the day:

Tweets by hour of day for O'Reilly and Schultz

Ed’s commanding lead in tweets comes purely from the engagement he generates from his audience during the show (the show is also on at 11pm ET, which generates a second spike towards the very end of the day).

So the winner in this Tweetdown is Ed Schultz!

Bill O’Reilly: Why not engage your viewers on twitter? Ask them to tweet you what they think of a topic each night and we’ll have a rematch!

Methodology: I looked at all tweets from 00:00 on Monday to 23:40 on Friday that had either @oreillyfactor for Bill or @edshow OR #edshow for Ed.