Contrarian Librarian

Most of the research we do is long lead; we have time to develop and execute a reasonably robust methodology. However, given the nature of PR, we also regularly get last-minute requests that require an immediate response (10 minutes to profile a reporter, give an analysis of top industry trends, etc). And sometimes the information we have in-house is not enough.

With quick & dirty research in mind, I am always on the hunt for free tools I can use to jazz up a 10-minute analysis. Wordle, Google Trends and BlogPulse Trends are old standbys, but I absolutely love all of the new tools that keep popping up, especially around social media visualization.

Here’s one Forbes (thanks for flagging, @shanlaw!) pointed to this week: Mentionmap.  It’s an interactive tool for exploring social networks on Twitter, including connections between users and hashtags. Unfortunately, the methodology is a bit vague – Mentionmap does not specify the date range of the analysis, what constitutes a ‘communication’, or the tie strength cutoff points for appearing on the graph. With those unknowns, I can only recommend it for a quick & dirty toolkit, but it’s a cool addition to a basic 10-minute profile.

What are your favorite tools for quick & dirty research?

The Conversation Prism 3.0 was released last week by Brian Solis and JESS3. Not only does this continue to be a lovely example of data visualization, it’s remains quite informative in terms of understanding the numerous social media platforms by consumer function or use. Moving from the outside in, the graphic also outlines how brands and organizations might leverage the various platforms to inform business and communications strategies.

For me, this graphic bluntly demonstrates just how complex and layered the social web has become. At the risk of becoming overwhelmed by variety, it’s simply important to be aware that all of these options are available. Similar to what Cool Infographics’ Randy Krum said, don’t think that you can effectively target all of the categories (talk about overwhelming). Nor should you assume that any specific target audience utilizes all of these categories (or even more than 5, in most cases). This is where understanding your target audience becomes so critical.

We rely on a number of primary research methodologies (both quantitative and qualitative) to understand what social media channels and platforms are most useful influential to specific audiences. We’d love to hear more ideas, suggestions or examples related to how others get at this type of information.

We are still working hard on a method for tracking information flow. We have a pretty good idea of how to map links in content, but my headache this week has been contextual references. What if a blog picks up on a new message or theme without directly linking to or quoting from another source? How do I identify the origin of that message or theme? And even worse, how would one automate a system to track this type of reference?

Stowe Boyd this week dug into the Harper’s Magazine archives and pulled out a quote that illustrates the complexity of the problem:

“Any text is woven entirely with citations, references, echoes, cultural languages, which cut across it through and through in a vast stereophony. The citations that go to make up a text are anonymous, untraceable, and yet already read; they are quotations without inverted commas. The kernel, the soul—let us go further and say the substance, the bulk, the actual and valuable material of all human utterances—is plagiarism. For substantially all ideas are secondhand, consciously and unconsciously drawn from a million outside sources, and daily used by the garnerer with a pride and satisfaction born of the superstition that he originated them; whereas there is not a rag of originality about them anywhere except the little discoloration they get from his mental and moral caliber and his temperament, and which is revealed in characteristics of phrasing. Old and new make the warp and woof of every moment. There is no thread that is not a twist of these two strands. By necessity, by proclivity, and by delight, we all quote.”

Jonathan Lethem, The ecstasy of influence: A plagiarism via Stowe Boyd

Samara and I have been working on updating a system we use to rate and rank content creators, so I thought I would take the opportunity to talk about the framework and some of the metrics we use to evaluate the influence.

 When we designed this system a few years ago, our goals were to make it:

• Extensible. At the time, Twitter was just gaining in popularity, and Second Life had made a brief splash before quickly fading. In addition, we were finding that as blogs were added to media lists, PR clients didn’t know how to compare them with mainstream media.  We needed a system that could easily be applied to different channels and that would allow for true comparisons between channels.
• Flexible. We had collected a list of social media metrics  we wanted to include in our system, but we knew that new metrics would emerge over time, so we wanted to make sure we could add and remove variables without too much hassle.
• Comprehensive. Existing approaches were far too narrow. You cannot just look at circulation or unique visitors per month and assume that you know an outlet’s influence. If that were the case, Parade Magazine would be at the top of every PR list. We had to incorporate other concepts of influence into our system.  
• Repeatable. We wanted to create a methodology that would be easy to repeat so that we could track how a source’s influence changed over time.

The approach that we came up with was fairly simple. We would collect metrics in four different areas:

• Reach. Traditional PR measures of direct readership. Examples might be circulation, unique visitors, or Twitter followers. 
•  Buzz. Measures of sharing behaviors. Examples might include inbound links, popularity measures such as Digg or Reddit, social bookmarking presence, or retweets.
• Engagement. Measures of reader interaction with content. Comments are a good example of engagement.
• Content. Measures related to the relevance of content. Includes the frequency and depth of coverage of a client’s brand or industry.
• Audiences. Measures of the audiences reached or targeted. Preferably based on primary research, but this could also be extrapolated from reviewing the content.

In all, we ended up with a system of about 30 different variables, all grouped under one of these attributes.

I’ll dive more into each attribute and underlying metrics in future posts.

FYI, we presented this material at Internet Librarian 2009 and also wrote about it on FUMSI, if you want to know more.

I admit it – I am kind of obsessed with data visualization and infographics. I spend more than a good portion of my spare time looking at sites like Chart Porn, Flowing Data, Cool Infographics, Information Is Beautiful, and information aesthetics, and marveling at the amazing visuals (such as the following).

Working in the PR and media industry, we are quite adept at creating compelling storylines and written content that bring otherwise boring data and information to life. But the old adage remains – a picture can tell a thousand words. When we take the time to package our research results in a visually compelling format or infographic, experience has proven that we immediately capture the attention of our colleagues and clients.

Still, at times it can be challenging to determine how to represent the data and findings in a manner that not only resonates with a specific audience, but also in a way that doesn’t distort the data or generate confusion as to what it means. Does anyone have tips or best practices for how to consistently evaluate whether data lends itself to visual presentation or is better left to a different format?

I’ve been working on a project to visualize the flow of information through media sites and across channels. My goal is to capture linking data and timestamps to show the origin of a story (whether through an outlet, blog, tweet, whatever) and then to plot who links to that original content, who links to the linkers, where the story crosses into different channels, etc. Sort of like this: 

(I know there are tools that do this or something similar already, but I am trying to find a more customized approach so that I can layer on other data.)

What I need to make this happen is a good linking search tool. I like Topsy for Twitter, but blog/online linking is a problem. Here are a few tests I ran early this week, using an Engadget post as my test case. Posts were considered relevant if they were: original content  (no reposts) and linked to Engadget in the text body or citations (no ‘related posts’ or ‘around the web’ sections). I also skipped forum content, to keep my life easier, although that might be something to keep in mind for later.

Google’s Link Search
Returned Results: 0
Comments: Zero results? Seriously? I don’t get it. Tried both the Advanced Search box and the link: operator. Sigh.

Google General Search
Returned: 81
Relevant: 0
Comments: The good news is that Google had a few good forum results, although I could have used a service like Boardreader to capture those links. The bad news – none of the four top tier outlets I knew had linked to the original post appeared in the results.

BlogPulse
Returned: 149
Relevant: 30
Comments: BlogPulse included many many reposts, which were excluded here. Also, this is blogs only, and only blogs indexed by BlogPulse, which is a fairly limited set, imo. It did return the top blogs I had previously identified.

Yahoo Site Explorer
Returned: 9598
Relevant: 2 in the first 100 entries
Comments: Ok, I got lazy and only looked at the first 100. But here’s the thing – it included at least a dozen results that posted before the Engadget article, sometimes YEARS before. I’m not comfortable with that.

Frustrating results overall, although BlogPulse wasn’t bad. What other linking tools am I missing?

This is my very first blog post! 

Sad that it took this long, because I have been working in PR research for five years now.It’s much easier to sit back and observe than to jump in and participate.

I’ll keep it easy to start, if that’s ok. And what’s the easiest type of post? The “this is cool” link. So here it is:

Check out WTF Is My Social Media Strategy for an awesome and cringe-worthy sampling of marketing-speak. The list of buzz words is painfully accurate, and I think someone may have bugged my office to come up with the collection of phrases.

Welcome to my world!