Social Media Measurement
by Katie Delahaye Paine
Many people groaned when KDPaine & Partners announced our social media classification of 27 types of conversations and 19 types of videos. At first it does seem like we're adding unneeded complexity to an already confusing field of study. But measuring social media is a lot more complicated than counting up key messages in a few print articles. And these categories really are useful.
Classifying the types of conversations and videos is actually a way to make social media measurement simpler, not more complicated. With these categories, the bewildering array of social media interactions is reduced to functional types that can be examined and compared, apples to apples.
Below are five examples of how we employ them to help our clients make better decisions. We hope that you discover additional ways in which they can be useful. So, let me know if and when these techniques come in handy for you. (And if you'd like to talk about how KDPaine & Partners can help you measure your social media, call Mike Layton at 603-696-6098.)
1. Measure across platforms
The beautiful thing about these 27 types of conversations is that they can be used to measure interactions in many places: Facebook, Twitter, Plurk, FriendFeed, My Space, on blogs, in your living room or on your own private network. Do people have different types of conversations in different places for different reasons? Now you can find out.
2. Figure out what to pay attention to and what to ignore
Different types of conversations are important to different organizations. For years in traditional media, we've classified articles according to standard article types, including letters to the editor, feature stories, industry roundups, and wedding notices. Invariably, we ignore the latter. Anyone who counts a story about "Sue Smith, manager of product X at Dell today married John Smith at Hewlett-Packard..." as part of their "competitive analysis" is foisting bad data on his or her audience. (I actually saw a demo of a measurement program from a prominent PR agency that included many such articles.)
Similarly, not all blog postings, Tweets or comments are relevant to your situation, and using our classification system enables you to make rational and consistent decisions about what to include or not include in your analysis. Now you can pay attention to what's important to you, and ignore what's not. For example, if 33% of all conversations on Facebook are offering something for sale, and you're MADD trying to track your online reputation, you can probably ignore, or at least skim, most of those want ads. On the other hand, if you're Prada and trying to protect your brand from counterfeiters, you probably want to read them all.
3. Figure out if the conversation is trending in your favor
Many corporations are now monitoring their reputations online, but aren't doing anything with the results. For instance, Southwest, Comcast and Dell are actively participating in Twitter as part of customer service programs to solve problems and respond to issues. I recently asked Southwest how they were measuring the success of their program. Their response was that they were tracking number of followers and key word searches. But how do followers and key word mentions contribute to putting butts on planes?
The real important thing for Southwest to learn is whether or not conversations about them are getting friendlier or more hostile. So they ought to count up the number of people expressing criticism or dismay. Then count up all the ones that express support, and compare the ratio. Is the good stuff outweighing the bad stuff? More importantly, they should look at the trend and compare to blog comments to see if they correlate. Southwest has the advantage of being able to also track online ticket sales with unique URLs, so in theory they could also track correlations with actual reservations.
Even if you don't have a sophisticated sales tracking system, you can use KDPaine & Partner's 27 standard discussion types to more accurately track not just the tone but the content of your conversations.
4. Decide how best to participate in social media
We recently had a request from a client for recommendations on starting a Facebook page for recruitment. Our advice was to first study what was being said about their brands on Facebook. What were people already saying about them? Knowing that most of what is shared on Facebook is video, we suggested they analyze the type of video and content to better understand what is acceptable and most likely to be shared. In this case, there were lots of photos of one of the company's products and some serious fans, but not a lot of video. We suggested a video contest to allow the fans to express their brand loyalty in their own way.
For another client, we compared the degree to which video which we classified as "promotional video" or "sightseeing/tour" was being picked up. To provide a benchmark, we compared the client's videos to that produced by peer organizations. What we found was that the competitors' videos were far more popular and more frequently shared, indicating an area of opportunity for the client.
5. You can figure out if you're getting what you want out of your social media program
If you only measure success via eyeballs, or even friends and followers, you really have no idea whether you're engaging your audience in a way that does your organization any good. If you post a serious educational video about your product that gets downloaded 1 million times you might think of the program as a great success. But if people are sharing it as a joke, and their comments express criticism, you might want to rethink the investment.
If you set up a Facebook page to recruit young talent and get 1000 friends in the first week, that's great. But if those friends spend their time on your page sharing personal information and don't apply for jobs, it's probably not a good use of social media.