ARF Social Media Insights: Can Social Media Effectively Track Influence? - Lynne d Johnson - MediaBizBloggers - Lynne d Johnson - ARF Social Media Insights

Published: July 7, 2010 at 02:40 AM GMT
Last Updated: July 7, 2010 at 02:40 AM GMT

There's an ongoing debate about whether influence can be tracked via social media. At its core, the issue revolves around the lack of a common definition for influence, especially when we're talking about social media. I'm on the side of quality over quantity, where influence is more about overall impact than the number of followers or fans, or even the number of clicks. There are definitely instances where quantity will trump quality every time, and in the advertising business those instances happen far more often because we measure things by understanding their ROI. But is social media helping us shift away from that kind of thinking?

With the first, it's possible to look at both the online and offline share of voice (reach of conversations) that a brand has, along with sentiment or likability for the brand (positive, negative, and neutral mentions) to determine the brand's SIM Score. With this model, you can also determine the SIM Score for an overall industry, say auto or financial. Primarily, the SIM Score is an accounting of your brand health: think of it as your social influence market share. And in this regard, it helps a brand to realize the impact of influence on brand affinity. It's a viable framework, given what it sets out to achieve.

On the other hand, Altimeter and Web Analytics Demystified have developed a framework which considers social media measurement meaning different things to different people—and for different reasons—and that no one social media analysis solution works as a one size fits all solution. Their framework looks at a number of KPIs that correspond to specific business objectives, including "Fostering Dialog," "Promoting Advocacy," "Facilitating Support," and "Spurring Innovation." For each KPI, such as "Share of Voice," "Advocate Influence," "Satisfaction Score," and "Sentiment Ratio," there is a numerical model that gets us closer to determining an ROI for its related business objective. A successful social media strategy can't happen without starting from the point of planning to reach a specific objective.

On the surface, both of these frameworks make total sense and lend credence to the idea that there is a method for determining influence through social media. And as social media changes daily and smartphones and portable devices become the major ways that people participate and share likes and dislikes, other metrics will certainly come into play. But together, for now, these two models get to the heart of the quantitative side of social media. But how can we understand or determine the qualitative side of social media?

Since it's the realm of the fuzzy stuff, it's often left to researchers who study consumer insights and behaviors. Yet for marketers this stuff is important too: It's not solely about understanding how many clicks an A-list influencer can drive, but also why that specific influencer drives so many clicks. If we're looking only at clicks, then we're looking at social media much in the very same way that we've been looking at all of digital media. Those older models can't explain the phenomenon of someone (or a brand) with a smaller pool of followers getting more clicks on a link than one with a larger pool? Is it the value of what that person has shared that makes a difference? Is it the trust level that person (or brand) has garnered? When we talk about how influence, word-of-mouth, and sharability are outpacing display advertising and even search, how can we view them through exactly the same lenses that we do display or SEO or SEM. Nielsen and Facebook have recently taught us that in understanding the value of earned media, we need to move beyond this sort of thinking.

"To me, influence isn't about popularity. Or even reach. It's about the trust, authority, and presence to drive relevant actions within your community that create something of substance."

In this experiment, the magazine's Influence Project will determine influence based upon clicks to a unique link created when a participant creates a profile. So the experiment measures how many people check out the profile, as well as how many people create their own profiles based upon the referral. Sure it's a social experiment that monitors clicking behaviors, but what Naslund asks is whether tracking linking/clicking behaviors is the equivalent of measuring or monitoring influence.

I don't think we're talking about a wrong way of looking at influence, but we could be looking at only one side of the equation. In measuring social media, we have to listen, observe, and study to understand who the real influencers are. Perhaps an influencer's influence isn't driven online, but offline. Here's where Razorfish's SIM Score (or perhaps Altimeter's Social Marketing Framework) can help us capture--along with the aid of engagement in a private community, an interview or survey--the offline component.

There are also instances of people being influenced, but not clicking. A blog post or tweet may influence a purchasing decision with no blog post comment left behind, or no follower or retweet to track. What Naslund's argument also bears to light is that we should think about separating influence from virability/viralness when we're measuring influence in terms of social media. Generating buzz and going viral won't always create influence. Many social media failures (here and here) generated a lot of buzz (and sometimes even a lot of social media activity) that eventually reflected poorly on the brand.

Quantity and quality can both make influence happen, sometimes together and sometimes separately. Let's consider what Geoff Livingston writes in his post, "The History of Influencer Theory on the Social Web,"

"So who's right? Where's influence, the uber-connected one percenter, trust agent, free agent? Or the person who lights the spark within his/her community of 150? Well, both are."

Both the A-list social media influencer and the niche community influencer can create impact (and influence) for your brand. But it's really those who tech blogger Robert Scoble once called passionates that will drive your brand advocacy. In that case, it's not always about how many people they're telling your story to but the actual stories themselves that they are telling about your brand—the ones that are engaging and in turn driving more people to tell your brand's story. That's real influence. And that's where quality and quantity converge. It's a blueprint for how companies should think about developing relationships with their consumers and potential consumers, as well as how they should think about measuring those relationships.

Check us out on Facebook at MediaBizBloggers.com
Follow our Twitter updates @MediaBizBlogger

To communicate with or to be contacted by the executives and/or companies mentioned in this column, link to JackMyers Connection Hotline.

Influence needs to be measured as far as "Influence to what?", meaning that influencing people with no purchasing power to click on a link or sign up for something is completely worthless.

Influencing people WITH purchasing power but no DESIRE to purchase the particular product a company's selling is worthless.

This is the problem with all kinds of contests. Regardless of the amount of talent or the worthiness of a particular pitch, if you leave the determination of who's "influential" to the number of people that they can get to click on a link, you're only going to end up with the winner being the person with the largest online network, not the person who can make things happen with a Call To Action that will be of any value to a client.

Posted at 10:53 AM on Jul 9, 2010 by Bill Cammack

Lynne, I also tried Klout yesterday and found it pretty interesting.

As marketing fragments into more and more micro-efforts, it's only natural to want to measure each.

What does this effort contribute vs. that one? Where should we cut our losses, and where should we double down?

But if we have a macro goal (e.g. shift opinions from "Toyota is unsafe" to "Toyota is reliable again"), will micro-measurement prove to be useful?

A problem with measuring each individual component in isolation is that it can't tell us much about how the components interact in an integrated campaign.

Said differently, we may find that it's like trying to understand how beef stew tastes by analyzing each of its ingredients in isolation.

We may learn a lot about the characteristics of carrots or potatoes, but we won't know a whole lot about the stew. :-)

Posted at 07:09 AM on Jul 13, 2010 by Tom Cunniff

Great questions and discussion, Lynne--particularly as to what constitutes an "influencer." The old model of "Celebrity = Influencer" (to over-simplify just a bit) is off the mark; each of us has the capacity to be a champion for a brand that delights us in a small or large way. We know this from experience: movie or restaurant recos at the water cooler, picking a garage-floor paint after a chat with a neighbor, etc. Geoff Livingston is dead-on in his observation that an "influencer" can be someone whose articulate passion shapes lasting behavior in a relatively small group of people. Smart marketers will keep both the qual and quant sides in mind as they build their brands.

Posted at 08:55 AM on Jul 13, 2010 by Pete Healy

@Bill Cammack - I read your blog post and listened to you on Cathy Brooks Social Media Hour. I liked a lot of what you had to say. I'm sticking to my guns though -- I know we can't measure everything, but measurement can be a part of the process. We just have to get to a common understanding of what influence is. I think its definition has changed dramatically with the implosion of the social web and especially now that we're heavily into the age of the Brand Called You.

@Pete Healy - We're definitely on the same page here. Just bc Ashton Kutcher is the leading Twitterer it doesn't mean that he's influential. He's a celebrity, of course all of those people are following him. But can he ignite people to action. Sure in some instances, but not in every. Influence depends on the situation -- and especially on the people.

@Sean Williams - I read your piece too and I welcome and thank you for sharing your insights. This is not traditional marketing. We can't think of this stuff like we've thought of every communication channel or advertising vehicle before. That's because it's not that. It's something else. Something more along the lines of trust agency. What Chris Brogan and Julien Smith talk about in Trust Agents. Social media has made everyone have to be much more honest, and also much more customer service oriented. It's about building trust through relationships. You can't measure that. It's like how many times a day you hug your kid -- does that mean if you hug him 10 times he'll love you more than the day you hug him only 5. You can't put value on relationships the same kind of way as you can clicks. Clicks are not going to measure relationships, or influence or inspiration. They can measure the reach of that message, but not the impact.

Posted at 07:18 AM on Jul 14, 2010 by

Lynne, I couldn't agree more that we're all learning. For me, that's the most fun part of this :-)

Among the things I think we'll need to learn is the ability to switch from micro to macro thinking and back without losing our balance.

We'll need to examine the veins on a single leaf on a single tree. But then we'll have to compare that learning with what we can see from the 3,000 foot view above the forest.

Most people are at their best at either extreme -- I tend to be more of a 3,000 foot view thinker. But both views are valid so we'll need to team up with (and trust) people who have strengths we don't have.

In short, we're going to have to grow into this. It's a challenge, but it's also the opportunity :-)

P.S. Great to meet you at the ANA Digital and Social Media Conference today. Thanks for saying hello.

Posted at 01:27 PM on Jul 14, 2010 by Tom Cunniff

@Sean Williams - Thanks for that link. "Guidelines For Measuring Relationships in Public Relations," looks like a viable framework.

@Tom Cuniff - Thanks for stopping by again. I learned a lot from your comments on the panel at the ANA pre-conference yesterday, and I'm sure these things will be taken into consideration in the work that the ANA, ARF, and WOMMA will do together.

Posted at 12:24 PM on Jul 15, 2010 by Lynne d Johnson