This is the fifth post in my series on Fortune 500 corporate blogs and it describes the structure of corporate blog memes. As I mentioned in the previous post, out of the 5,887 corporate blog entries posted in my study window, only 299 produced memes of at least two tiers.
The tier system is based on the Blogpulse conversation tracker which was used to identify memes and the relationship between blog posts based on the direction of hyperlinks. For example, posts linking directly to corporate blogs were classified as tier one posts. Posts linking to those tier one posts were tier two posts and so on.
The chart below shows the frequency of memes observed by the number of tiers from a minimum of two to a maximum of seven. The same sharp downward trend that was present in previous charts is also visible in the depth of corporate blog memes.
Of the 299 memes that were collected, 240 were comprised of the minimum two tiers and it was illustrated in the previous post that 163 of these were made up of just two blog posts.
A very small number of memes extended beyond three tiers and understanding the factors that separate these memes from the rest is of great interest to marketers wanting to spread content and messages through social media channels.
It would also be interesting to compare the structure and specifically the depth of blog memes with the way content spreads through the retweet function (or otherwise) in Twitter. Early statistics would seem to suggest that the depth of memes in Twitter are also quite shallow but constraints like the 140-character limit (and possibly many more) may affect attribution and therefore data collection.
The second chart below shows the number of blog entries found in each tier across all memes. As only memes containing at least two tiers were collected, it only represents a subset of the total number of posts that linked to corporate blogs. The number of tier one posts is smaller that the actual total and as a result the overall number of posts is also smaller.
The total number of consumer blog posts that can be connected back to corporate blogs probably demonstrates a more perfect downward curve similar to those observed in other areas but of the memes collected (min. two tiers) the number of posts expanded by almost double from tier one to tier two. The sharp declined observed elsewhere started from tier two to tier three.
As with previous posts in this series, there are a number of caveats specific to this data that I want to mention here in addition to the general Disclaimers, Caveats and Limitations page.
First, these figures depend entirely on the coverage and accuracy of the Blogpulse index which based on their website is continually worked on and improved. Second, like the rest of the data published in this series, it represents a snapshot in time.
I also want to reiterate that this data is a subset of consumer blog posts that can be tied back to corporate blogs because I think that can get lost when looking at charts. Only posts in memes with at least two tiers were recorded because they were the only ones useful for my study so many tier one posts were not recorded.
Posted on July 22, 2009
Of course, that depends on how frequent and how big you thought corporate blog memes might be to start with! This is the next post in my series on Fortune 500 blogs and it focuses on the number of consumer blog entries that connect back to corporate blogs to form what have become known as memes.
The vast majority (262 out of 299 or almost 88%) of memes I collected were between two and ten posts in size while the biggest meme reached 56 posts. The chart below shows a frequency distribution of meme size measured by the total number of consumer posts in each meme. As with corporate posts (and as will be shown later in other areas), there was a sharp decline in the frequency of memes as the number of posts increased.
It’s important to point out that I only collected memes where a consumer linked to at least one other consumer. I was trying to measure patterns in sentiment between consumer posts so the only memes that were useful for my study were those with at least one connection between consumers. By extension, it also meant there were at least two posts in each meme by default.
The second chart below breaks out memes of 2-10 posts and reveals 62% of the memes in that category and 55% overall contained the minimum two posts. There were 5,887 corporate blog posts collected in total and only 299 (5%) produced memes of at least two, linked consumer responses meaning it was quite rare for connections to form between individuals.
However, that doesn’t mean there was no consumer blog response to the other 95% of corporate posts, although that was the case for a good number. It was possible that a corporate blog post generated any number of consumer responses but that these linked directly to the corporate blog independently without referencing one another and were not recorded as a result. (This will be explained in more detail in future posts on meme structure.)
I think there is a very good chance that the aggregate patterns for meme size would stand-up upon further study but I also feel the need to reiterate that the data is subject to the limitations of the study methodology and tools. Memes can change over time as links can be broken and posts can be added, updated, hidden or permanently removed. The data represents a snapshot in time.
The meaning that can be drawn from using blog posts to measure meme size is also limited because it is a one-dimensional approach. Other forms of online communication were considered when planning this study, in particular the role of blog comments, but ultimately it was a necessary limitation of a small study with constrained resources to focus on blog posts.
In the past year and even in the last few months, the extent to which online memes are distributed across many social media platforms has only accelerated (I don’t think I need to mention Twitter). Companies need to monitor their corporate reputation and marketing campaigns within the broader social media landscape which I’m sure will continue to provide a moving target.
Posted on July 15, 2009
One area that was not the focus of my study but resulted in some interesting data was the number of posts published by each corporate blog. That data is analyzed in this post although, in addition to general caveats, there are clearly limitations in looking for trends in a snapshot of data.
Almost half (49%) of all corporate blogs published no more than 5 posts during the two month period in which I studied them. In fact, 81 (21%) of the 389 blogs tracked did not publish any posts at all. The chart below shows the range of posting volume for Fortune 500 corporate blogs.
Although I am trying to present and describe the data as informally as possible on this blog, I would have liked a little more precision on this chart. It uses a line chart for a frequency distribution which is not orthodox (for good reason). In addition, the first band starts at zero and floats away from the y-axis.
I chose to use a line chart in order to illustrate the sharp downward trend because it repeated itself in data on meme size and consumer participation (see future posts for more on these). It was also a very wide distribution because of one blog (by AOL) that posted 546 times during the two month window.
Overall, 91% of Fortune 500 blogs published less than 40 posts during the two months and 80% published 20 or less in two months. The complete spreadsheet of post totals is posted on Swivel.
The chart below expands the frequency of blogs posting between zero and 40 times in a more traditional histogram format. It is possible to see the same sharp downward trend in frequencies within the 0-40 range and it is most pronounced going from zero to one post.
As I said at the top, by far the largest number of blogs (81) published no posts at all which meant only 308 of the 389 blogs included in the study published posts. Although, those 308 blogs accounted for 70 of the 76 companies found to publish blogs.
This might suggest that companies with blogs are posting with some regularity (at least every two months) but perhaps publishing multiple blogs increases the chances that posts become less frequent or stop completely.
However, it’s hard to draw significant meaning from this data because it represents a snapshot taken over a two-month period. In addition, corporate blogs with no published posts in the three months prior to July 2008 were excluded as already inactive.
A longitudinal study tracking publishing activity over a longer period of time and collecting more detailed data points would be needed to validate any observed trends on the posting activity of corporate blogs.
Related Pages
Posted on July 8, 2009
At the time of data collection in September 2008, there were 76 Fortune 500 companies publishing at least one corporate blog as I defined them. There were 391 blogs in total although this included some portals/platforms that encompassed from a few dozen to several thousand individual employee blogs.
The list was dominated by technology companies and technology-related industries. The chart below shows the top 15 companies by the number of corporate blogs for each.
A full list of Fortune 500 corporate blogs identified in this study is available here. The list is probably already a little old given how fast things change. There was at least one more recent survey of Fortune 500 blogs that counted 81 companies blogging although selection criteria may differ.
The chart below shows that 66 of the 76 companies with blogs in my study had between 1 and 10 blogs and 8 of the remaining 10 were associated with no more than 20.
Related Pages
Posted on June 25, 2009
Introduction
I conducted a survey of Fortune 500 companies for my Masters thesis and I wanted to share some of the data it produced in the hope of making it useful. In order to do that in a blog format, I plan to keep things as simple as possible by not approaching it in an academic way. That might prove to be a mistake but we’ll see what happens.
Background
I don’t plan to post within the context of my thesis because it’s probably not interesting or useful for very many people. Instead, I’ll focus on presenting basic data on corporate blogs. However, I have added a few pages on my study to provide some context for anyone that might be interested:
The Data
I plan to post all of the underlying data on Swivel which makes it possible to cut, slice, comment, share and download public data. It looks like Swivel is undertaking a pretty major redesign at the moment so things might change but I’ll try to make sure all data is available somewhere.
Work-in-Progress
This is a work-in-progress. I’m still in the process of uploading data and I’m still adding links to several the pages listed above. This will probably continue as I post new charts and interlink pages and posts where helpful.
Questions or Feedback
If you have any questions or feedback on anything posted here or in this series, you can email me at phillip.d.baker@gmail.com and I’ll get back to you as soon as I can (usually pretty quickly).
Posted on June 24, 2009
My name is Phillip Baker and this is my personal blog about finding value in a world of free information.