Category Archives: Measuring Social Media

Measuring Social Media

Vitrue Thinks Amazon Paid Premium for Zappos Social Buzz

1 Comment

Vitrue CEO Reggie Bradford tweeted earlier today that some of the premium that Amazon paid for Zappos had to do with the social buzz, showing that Zappos outscored Amazon 5300 to 2790 on their Social Media Index.

Thought I’d cross check that idea a bit using a couple other tools: Zappos is not even a blip on Facebook, with Facebook Lexicon having insufficient data to come up with a graph showing status mentions. Next, I looked at Social Mention, showing both brands had nearly identical number of mentions and a favorable rate of 5-to-1. Finally, using Radian6, you can see that between 6/23 and 7/21 that Amazon handily had more mentions combined (592,000 to 19,000) across blogs, forums, Twitter, images, videos and comments.

The Vitrue charts make it look like all of Zappos score was due to Status updates. Zappos has three core accounts, including Zappos CEO Tony Hseih’s Twitter account (currently #45 overall, crossing over 1 million followers earlier this week. In addition, 436 Zappos employees are on Twitter (25 have at least 1000 followers, and the next 25 have at least 500). But Amazon is not a complete newbie to leveraging Twitter: their MP3 Twitter account has over 800,000 followers (currently #103), and there are other feeds like @AmazonDeals, @Amazon and @AmazonGames that compare favorably in number of followers (Although as mentioned before in analyzing Tony’s Tweets, the content of Amazon tweets is mostly deal –oriented, where Zappos talked much more about the culture of the company and made the Follower feel like an insider).

Using Radian6 again to look at the breakdown of mentions by media type covering 6/23 to 7/21 (pre-acquisition announcement), it’s true that Zappos has a higher percentage of mentions in “micro-media” (or Twitter) by a margin of 62.7% to 25.6%, but Amazon still handily had more mentions (146,000 to 12,000).

Zappos Mentions by Media Type

Amazon Mentions by Media Type

So while Amazon might have paid a premium to better understand how to better engage with customers over Twitter (among several other potential synergies), Amazon appears to have quite a sizable lead in the buzz category. Also, the Vitrue SMI appears to have totally missed the impact of blogs for Zappos, which looks odd in comparison to the Radian6 numbers.

Measuring Social Media, Social Media

Benchmarking: Top Magazines Using the Going Digital, Getting Social Scorecard

Leave a comment

As traditional publishers like Meredith, Time Inc and others watch newspapers die off left and right, the challenge is trying to stay relevant and reaching users who are increasingly shifting their consumption online. Many big publishers are dipping their toes into social media as a way to engage their users and I wanted to introduce the Going Digital and Getting Social Scorecard™ (GDGSS) to create some benchmarks and identify best practices.

Methodology

These metrics are a very crude start, but allow us to at least compare brands and understanding who is doing better than others. To start, I took the top 50 published titles based on paid and verified circulation for 2008 as reported by Magazine Publishers of America, excluding titles that are part of membership dues (like AARP and AAA). For this group, I looked at four key measurements:

  • Percent of Circulation that turns to Unique Monthly Visitors to their website: measuring how well you can duplicate the offline audience to an online audience (monthly unique as measured by Quantcast)
  • Facebook engagement: relative ranking of number of Fans to a Fan Page plus average number of interactions with items posted by the publisher to look at engagement
  • MySpace engagement: relative number of friends and comments.
  • Twitter engagement: relative number of Followers and number of Tweets over a seven day period

Companies were awarded up to 5 points for each measure and aggregated to get an aggregate GDGSS total. Ultimately a magazine is going to measure success based on how much these measure drive unique visitors, page (and thus ad) views and ultimately subscriptions. Thus weighting each of these evenly has its flaws, especially since none of the measures directly translates into the ad views and subscription numbers we desire (that is the work of back-end analytics tracking the ROI of each of the traffic sources). Let’s take a look at the top twenty companies by their GDGSS score:

Top Twenty Going Digital and Getting Social Scorecard Magazine Titles

Rnk Title 2008 Circ % Circ to Web Score FB Score MySp Score Twitter Score GDGSS Score
1 Time Magazine 3.37 mil 3.04 2.98 0.00 5.00 11.03
2 Entertainment Weekly 4.03 mil 1.11 1.40 0.00 5.00 10.14
3 US Weekly 1.90 mil 0.09 4.99 0.00 5.00 10.08
4 Money 1.93 mil 5.00 0.00 0.00 5.00 10.00
5 People 3.75 mil 3.82 0.92 0.00 5.00 9.74
6 Newsweek 2.72 mil 3.25 1.40 0.00 5.00 9.65
7 Playboy 2.66 mil 0.93 5.00 0.00 3.64 9.56
8 Rolling Stone 1.46 mil 1.25 0.83 3.69 3.05 8.81
9 National Geographic 5.06 mil 0.87 5.00 1.00 1.76 8.63
10 Cosmopolitan 2.93 mil 0.53 4.32 2.56 0.57 7.98
11 Martha Stewart Living 2.03 mil 0.96 1.40 0.50 5.00 7.85
12 ESPN the Magazine 2.05 mil 5.00 2.00 0.00 0.58 7.58
13 Seventeen 2.03 mil 0.70 3.26 2.00 1.34 7.31
14 Maxim 2.53 mil 1.03 0.50 5.00 0.11 6.64
15 Sports Illustrated 3.24 mil 3.01 0.50 0.00 3.13 6.64
16 Glamour 2.32 mil 0.46 1.46 0.00 4.47 6.40
17 InStyle 1.79 mil 0.80 0.00 0.00 5.00 5.80
18 O, The Oprah Mag 2.38 mil 1.31 0.80 0.00 2.66 4.77
19 Birds & Blooms 1.52 mil 0.09 3.74 0.00 3.74 4.08
20 TV Guide 3.27 mil 0.76 0.93 0.00 2.21 3.89

Many of the big guns (in terms of circulation) did not score highly. In fact, eight of the top ten in circulation in 2008 didn’t make the Top Twenty Going Digital, Getting Social Scorecard, including Readers Digest (#1, 8.3 mil circ, 1.69 GDGSS), Better Homes and Gardens (#2, 7.65 mil circ, 2.98 GDGSS), Good Housekeeping (#4, 4.68 mil circ, 0.92 GDGSS), Family Circle (#5, 3.9 mil circ, 1.00 GDGSS), Woman’s Day (#6, 3.9 mil circ, 1.05 GDGSS) and Ladies Home Journal (#7, 3.84 mil circ, 2.96 GDGSS).

Still, National Geographic, #4 overall with over 5 million in circulation, made the Top Ten and got contributions from all four categories. But this may have everything to do with the content (a continual flow of science news and content which has more potential for breaking news than the traditional housewife-focused fare) and demographics (skewing to the 34-50 age brackets whereas many of the top mags skew much older). If anything, the Top Twenty for the most part suggests that companies that have a solid feed of news and breaking items have an easier time of expanding digitally and socially.

That said, even in the Top Twenty GDGSS Magazines, only three scored more than half of the 20 total possible points. The top seven actually have no contribution form or presence to speak of on MySpace, and only 25% seem to have any presence, yet you would think sites like US Weekly and People could figure out a way to work well within that environment. MySpace was by far the most difficult to measure and identify metrics, but I’ll break that down as well as the average number of Fans and Responses for Facebook and the number of Followers and Tweets in another post next week.

Let me know your thoughts on the metrics, how to potentially improve them, and what you think this means for the top publishers.

Casual Games, Measuring Social Media

Going Beyond Daily Active Users: Which Game Applications Engage Users on Facebook

Leave a comment

Six of the top ten applications on Facebook are games as measured by Daily Active Users – here are the top six games and their daily active users from DeveloperAnaltics as of June 4th:

Game Developer Daily Active Users (DAU)
Farm Town Slashkey 3.15 million
Mafia Wars Zynga 2.94 million
Pet Society Playfish 2.77 million
Texas HoldEm Poker Zynga 2.46 million
Restaurant City Playfish 1.60 million
Yo’Ville Zynga 1.10 million

But as with any game, a developer has to ask, do they have legs? Will there be a sustainable audience that makes it worth developing a sequel (if you are in the casual download game business) or extending the application to other platforms (as most start out on Facebook and then move on to MySpace, Bebo and Hi5). While daily active users and of course revenue factor into those decisions, what role does user engagement outside the game play in helping make that determination? Facebook Lexicon (below) gives us some indication of how the games are trending, looking at the number of mentions of key phrases on user walls and status updates:

What, no Texas Hold’em Poker? Well Facebook Lexicon only allows you to use two word phrases and there is not a way to filter out generic mentions versus game-specific mentions, so we’ll have to pass on them for this analysis.

The other thing about the Facebook Lexicon tool is that there is no absolute number of mentions, only a relative graph that shows the overall trend. To solve for that, I’m introducing the Facebook Lexicon Activity Pixel Index ™ (or FLAPI) that measures the number of pixels from the x-axis and to give you a number you can use to benchmark competing brands when using Facebook Lexicon graphs. You can then figure out the FLAPI per user to see how engaged individual users are with a specific application.

Here’s an updated chart with the FLAPI and FLAPI per million daily active users:

Game Developer Daily Active Users (DAU) Facebook Lexicon Activity Pixel Index ™ FLAPI per Million DAU
Farm Town Slashkey 3.15 million 131 41.6
Mafia Wars Zynga 2.94 million 300 102.0
Pet Society Playfish 2.77 million 68 24.5
Restaurant City Playfish 1.60 million 50 31.3
Yo’Ville Zynga 1.10 million 55 50.0

Note that I took the pixel distance using the LAST date where Facebook Lexicon presented data, which was May 31st. The resulting data looks a lot more interesting when you chart it like this:

Now you can see that when you take the FLAPI and adjust it for the audience, you get a great feel for which brands , relative to their audience, have users talking about their game in their Facebook status and wall pages. In the example above, if you took a 45 degree line as a baseline, both Zynga games (Mafia Wars and Yoville) would appear above the line and infer that they are better at getting their users engaged on Facebook than their competitors.

So while a game may wane based on game play, like Pet Society which clearly shows that it’s beginning to lose active users, it very well could be that the active engagement of the users with a game(and possibly the way that Zynga does it) could be actively extending the typical lifetime of a game application.

Interested in looking at how we can apply the FLAPI to other brands, including those that have only a Facebook Fan page and don’t have an application. And also how much the FLAPI is impacted based on these brands having multiple fan pages – today Facebook Lexicon does not appear to mine those pages (”Lexicon shows the number of users that posted each term per day on a profile, event or group Wall.”), but those pages could be driving user posts.