Measuring Social Media

by Chris Kenton on December 7, 2006

I want to go a step further in breaking down the dialog over social media metrics, in the wake of the Factiva roundtable. The questions I asked yesterday were what should be measured, why should it be measured, and what will the impact of those measurements be. Let’s start by looking at who wants to measure social media in the first place.

When a social media channel first takes off, the measurements that are important first are those that are relevant to the Content Provider. Take Marketonomy. I want to know what kind of people are reading my blog. I want to know what they find interesting. I want to know how to better target my content to build a better relationship with readers. Currently, those metrics are not very robust. I know how many people subscribe to my RSS feed, but I know next to nothing about them. I know which of my posts get read the most, but with the small ratio of comments to readership, I’m not always certain what drives those posts to the top. The metrics I wish I could get today are demographic. I don’t need to know *who* my readers are personally, but I would like to know have a general profile of their professional background and areas of interest.

Once a social media channel has some traffic, new measurements come into play. If you want to bring on advertisers or sponsors, you need to demonstrate the value of your channel over others. In print, you point to subscription demographics and circulation numbers to justify ad pricing. On the Web, you’ve got traffic. Of course, these metrics are open to gaming. Publications find creative new ways to inflate their circ numbers, just like Web sites find creative ways to inflate traffic. Content providers package pretty numbers to push up ad prices, and ad buyers poke holes in those numbers to push the price down, and the same game will certainly develop with social media.

Just like the current dialog in Social Media circles, traditional media has mounted periodic campaigns to introduce new metrics that justify more ad spending. I once worked for a magazine that tried to measure pass-along rates, and I’ve put the same concept into play with email campaigns using source codes. But these metrics are usually a sign that the provider doesn’t have leverage with buyers, which means they probably don’t have high demand. Just as traffic/circulation has been the bottom-line metric for traditional channels, I would argue that it’s going to remain the primary metric for most social media channels too–the more traffic you have, the more demand there will be to reach your audience, the less granular your metrics will need to be, simply because you’ve got waiting buyers.

But the complicating factor in all this is how a channel grows and maintains a robust audience, because once it reaches critical mass, it has a hungry mouth to feed. You’ve got to find worthwhile topics to cover, people to interview, or at least engaging people that will drive dialog and audience participation–and if they’re really that interesting, other people will want them too. That’s where other players in the value chain for delivering content often come into play, and other metrics matter to them. Whether you’re dealing with an interview prospect’s secretary, or an executive’s PR rep, or even if you’re relying on users creating content, the question is: Why should anyone bother driving content into your channel? The answer is that there must be some value they can derive from it, and one of two metrics will demonstrate that value.

For anyone actually engaging in social media, the important metric is likely Participation. The more people are engaged, the more people I have to talk with, or at least listen to. For anyone interested in getting content in front of your audience, like a company or PR group, the important metric beyond raw traffic is likely Influence–which is the power to impact the opinions, attitudes and behaviors of a target audience. In Social Media, traffic, participation and influence are certainly related, although the relationship doesn’t seem all that clear yet. You can have high traffic and low participation, and still have signficant influence–just as the Wall Street Journal has for generations. You can generate influence with low traffic and high participation–just as my former job at the CMO Council did with programs engaging a small group of leading marketing executives.

I’ll dig into the dynamics of influence more next time. For now, I’ll just repeat the argument that if you have significant traffic, you’re going to have high demand to reach your audience. More granular metrics will help you raise the value of reaching your audience, or compensate for a lower traffic numbers than competitors.

{ 4 comments… read them below or add one }

Rob Fields December 7, 2006 at 8:57 pm

Some really strong examples here, Chris. Kudos!

Any discussion of participation or engagement will have to incorporate reorientation on the part of marketers. You’re right that it might be futile to apply traditional metrics and concepts to the world of social media. Better to look at other sites linking to a particular site or page; number of comments or trackbacks to blogs. These begin to give an indication of influence and engagement. The other part of the reorientation that I mentioned will be to get marketers to clarify their objectives for getting involved in social media. If social media is about “them” not you, then using straight ROI measures will be useless. However, if you’re looking to become an active and valued part of a community and part of a conversation, then other measures will have to be developed, since there’s certainly value in being in this space.

Chris Kenton December 11, 2006 at 9:24 am

Some good points, Rob. I think the challenge with measuring influence is that metrics available at the moment are really more about reach, rather than influence, and they’re easily manufactured. Trackbacks and posts as raw numbers are easily corrupted by SpamBots. If someone blocks trackbacks or forces some kind of verification before posts are accepted, their raw numbers may be very low, but in fact their influence in a social community might be quite high–in part maybe even because of the higher bar for posting that filters out garbage. Additionally, *who* is being influenced matters quite a bit, which typically requires some type of demographic. If I’m highly influential among entry-level marketers with no budget influence, I don’t have as much mojo as I would influencing senior marketing executives holding the purse strings. Without some kind of demographic, that’s a hard thing to prove.

Ironically, I think we’ll start seeing some very traditional tactics emerge as bigger companies try to make sense of this space. To wit: I expect we’ll start seeing some brand awareness studies conducted by big companies among users of social media sites, looking to measure the impact of targeted social media campaigns.

Nelson Bruton May 20, 2008 at 7:34 pm

Chris,

I really enjoyed reading your post. Here are some of my thoughts.

You can measure a social media campaign only after you determine the objective for the social media campaign. Influence and interaction and results are the ways in which a social media campaign can be measured. Each has quantitative and qualitative elements. Below are my initial thoughts on this subject. Please bear in mind that there are probably more to add to each category. (Help, advice, and collaboration is appreciated)

INFLUENCE
Quantitative – 1. the number of people in the network 2. the number of networks/social communities/platforms 3. the growth rate of your network

Qualitative – 1. who is in the network? 2. what is the motivation for people joining the network? 3. what ideas are discussed in the networks

INTERACTION
Quantitative – 1. the number of communication methods within a platform 2. the number of scheduled tasks(eg. messages, replies, comments, bulletins, blogs, etc)

Qualitative – 1. the types of communication being sent out 2. who are you targeting with a particular message?

RESULTS
Quantitative – 1. number of leads generated 2. number of sales generated 3. number of new contacts made 4. revenue generated

Qualitative – 1. types of leads generated 2. types of contacts made

Would you mind if I reposted some of your content on inSocialMedia.com ?

Respectfully,

Nelson Bruton

Chris May 21, 2008 at 11:06 pm

Hi Nelson–

Thanks for your comments. Sure, repost away.

I think we should clarify one important point at the start. When you talk about campaigns and measuring lead generation, you’re talking about objectives measured in marketing ROI.

When I was writing the original post, I had my publisher’s hat on and was thinking of business objectives measured in Brand Equity–building a media property able to drive persistant revenue from readers or advertisers.

So while I think there are certainly similarities in how you would measure social media from those different perspectives, they are distinct points of view.

I think the division of quantitative and qualitative measures is helpful. I can think of quite a few more metrics I might add in general… like frequency, authority, relevance. But you really have to look at each metric individually to see if it really adds meaningful value.

I’ve seen lot’s of deep-dive discussion on the practicality of how you measure influence. Is Technorati meaningful? How reliable are trackbacks as a measure of influence?

Finally, I think you have to include readership without explicit engagement in the Interaction category. Social media is a tremendous driver of search engine result traffic, which means even beyond the standard ratios of real-time participants to immediate readers, many more people engage by reading the content for months after the dialog ends.

Wish I had time for a more well reasoned response. Thanks for the post.

/chris

Leave a Comment