ExperienceCurve by Karl Long

Avatar

Social Media and New Marketing Strategy

Engagement vs. Popularity Metrics in Blogs

Lots of people have been talking about the right metrics for blogs and other kinds of social media for a couple of years now. It is in fact a desperate need in the blogging community because apart from the obvious ego-surfing, feedback and benchmarkeing is absolutely critical to managing and growing a blog. Also for anyone running a corporate blog, how do you show ROI and how do you show growth, progress, results? More to the point if you are managing a blog how do you provide the right metrics in place to drive the right behavior on the blog. The old adage of you can’t manage what you can’t measure holds true, as does the idea that you only get what you measure, so you’d better be measuring the right thing :-)

Most of the metrics thus far have been based upon typical web-site metrics, unique visitors, page views, incoming links etc. which are fine measures if you are only interested in popularity. Now this is fine if your business is pleasing advertisers because of course they are still obsessed (however misguidedly) on eyeballs. But popularity is a very misleading measure if your aim is something other than popularity, and is especially meaningless when you compare part time blogs, to professional with a whole editorial staff.

Many folks have compiled lists of blogs ranking them using various publicly available information. Mack Collier’s Top 25 Marketing Blogs (recently expanded to include Social Media blogs) uses Technorati Authority to rank these blog, and the Ad Age Power 150 uses a multi-metric including Technorati, RSS subscribers, Google Page Rank and others, but IMHO still essentially measures of popularity. Seth Godin’s blog is at the number 1 spot on the Top 25 marketing blog and the Ad Age Power 150.

Now in comes Aide RSS a company that aims to provide measures of “engagement” (see Mark Ghuneim on Measuring Engagement). Now engagement has been a watchword in the marketing and advertising community for a couple of years now and I think there is a great deal of consensus that customer engagement is a more meaningful measure in the world of social media than measures of popularity. Beyond Social Media it is becoming clear that it is engagement over time that is one of the secrets behind building great brands in the web2.0/social media space, I really like the work David Armano is doing on Micro Interactions and Direct Engagement, it is those Micro Interactions which form the basis to measuring engagement.

Anyway, Aide Rss has been doing some engagement measurement using their soon to be released API and they used Mack Colliers Top 25 Marketing blogs as a baseline for that test. They published their results as an image, so I took the liberty of transferring it to a table and calculated the relative gains and losses of these blogs. I think the results are pretty interesting and there are some big moves in what for the last year has been a pretty static list. In the table I’ve got the Top 25 marketing blog standings based upon technorati rank and then on the right the Aide Rss Engagement rankings, and in the last column a +/- for where the blog has moved.

Here’s how they calculated engagement:

So how did I go about it? With our custom-designed API — sorry, hasn’t been publicly released yet — I analyzed each feed, which accomplished the following:

  1. counted number of posts published in each of the last two months (so essentially for May and June)
  2. counted numbers of each type of engagement we analyze, e.g. 200 clicks, 5 comments, 12 trackbacks, etc.
  3. weighted each engagement type for level of engagement
  4. added up the engagement scores for all engagement types for all blog posts for each month to calculate an overall engagement score for each month
  5. calculated an average engagement score based on dividing total engagement score by number of posts per month
  6. calculated the percentage increase or decrease in engagement for the blog’s content month over month.





  Popularity Technorati Score   Engagement    
1 Seth’s Blog 9,223   Chris Brogan 47028 +3
2 CopyBlogger 6,270   Seth’s blog 39535 -1
3 Chris Brogan 1,935   CopyBlogger 33696 -1
4 Search Engine Guide 1,471   Daily Fix 9933 +4
5 Logic + Emotion 1,288   Search Engine Guide 7670 -1
6 Duct Tape Marketing 946   Duct Tape Marketing 7037 0
7 Influential Marketing 834   Logic + Emotion 4362 -2
8 Daily Fix 761   Social Media Explorer 4254 +14
9 Brand Autopsy 717   Six Pixels Of Separation 3901 +3
10 Church of the Customer 661   Conversation Agent 3869 +1
12 Conversation Agent 625   Drew’s marketing Minute 3150 +2
13 Six Pixels of Separation 619   The Viral Garden 3086 +5
14 Drew’s Marketing Minute 605   What’s Next 2770 +2
15 Jaffe Juice 603   Influencial Marketing 2387 -7
16 What’s Next 475   Damn, I Wish I’d Thought Of That 2289 +6
17 Diva Marketing 439   Techno Marketer 2104 +7
18 The Viral Garden 438   Brand Autopsy 1864 -8
19 Greg Verdino’s Marketing Blog 427   Church Of The Customer 1809 -8
20 CK’s Blog 418   Greg Verdino’s Marketing Blog 1712 -1
21 Damn! I Wish I’d Thought of That! 415   The Social Media Marketing Blog 1481 +5
22 Converstations 402   Jaffe Juice 775 -7
23 Social Media Explorer 389   Diva Marketing 772 -6
24 Techno Marketer 385   Converstations 612 -2
25 Every Dot Connects 378   Every Dot Connects 464 0
26 The Social Media Marketing Blog 376   CK’s Blog 320 -6

(BTW no idea why the table looks so bad, I guess k2 is overriding everything)

I think this is a pretty fascinating experiment and I for one am very excited for when Aide RSS release this API. The one thing that I think is interesting here is the number of blogs who I consider very authoritative (and personal favorites) like Jaffe Juice, Brand Autopsy, and Church of the Customer, that had quite significant drops on the engagement scale. Obviously this is just after a cursory glance at the results and doesn’t take into account blog design or other factors, but I wonder if the more popular a blog gets the less conversational it becomes?

Anyway, I coded all this by hand so there may be errors, especially in recording gains and losses so please let me know if there are any problems.

4 Comments, Comment or Ping

  1. I also like the attempt to measure engagement rather than popularity. It should be pointed out that Aide RSS only looked at the last 60 days of each blog, so it’s not an overall picture but more like a snapshot. Still, it’s valuable information and something to look forward to implementing.

  2. Karl:

    I think you’re onto something with popularity and drop in engagement. We used to have more conversations at my place, now if there are more than a couple of comments it’s a lucky day! However, I noticed that many of the engaging posts were not business proper, they were a bit more personal and off the cuffs. So of course those would get a higher engagement level. But are those posts more useful to individuals? Probably not, I’m not *that* interesting. I’d venture that actionable business information would have greater value.

    And here’s the thing, what if people who get value form it do not link or do not have blogs? What then? How do you measure that? What’s your take?

  3. Karl - fascinating. many thanks for your time and energies putting this together. it was no simple job! thanks for sharing the results with the community.

    to valeria’s point, i also notice an increase in comments when posts are either more personal or the ‘information’ is wrapped around a story. interesting that although seth does not have comments turned on his engagement score is the 2nd highest ranked.

  4. Thanks guys, yeah there are some really interesting things going on here, and what kind of posts attract the most comments is always interesting.

    I often find that often if I try and cover all the bases in a post and try and be authoritative I get much less in the way of comments and conversation. Some of it has to come down to voice, some blog posts sound more conversational and more open. Not sure if that comes with practice or some people have a more conversational editorial style.

    @Toby totally agree about Seth, he still has a very high engagement score, but I still find it fascinating that he has an unassailable position from the standpoint of popularity, and this metric is the only one that has moved him from that top spot. I don’t think it means a great deal, but it does indicate that this is a very different metric that is being used here, and I think that’s exciting.

Reply to “Engagement vs. Popularity Metrics in Blogs”

Recent Photos