Social Media Metrics

I’m gearing up for the SXSW Interactive Conference and Festival in March. It promises to be a great conference. One of the panels I’m most interested in, “Social Marketing Strategies Metrics, Where Are They?” will be led by Tom Parish, a colleague.

Metrics, as applied to social media, are rightly heating up, not that this is a new thing. Shaping the dialogue are the conversation’s level and the insight of people whose opinions matter.

The issue isn’t why, what, or how we should measure social media. There are answers for every one of those. We measure because it leads us down a path of sustainable results toward a process of continuous improvement. Even if we start aimed 180 degrees from where we want to go, with the combination of a plan and some basic metrics, not only will we get turned in the right direction, we’ll actually get there. More practically, if we can’t express our objectives and methodology quantitatively, we have zero chance of selling in our ideas. This doesn’t mean that we run by numbers alone; we don’t. But it means that at some point as a CMO you’ll be up against the COO competing for dollars, and your CEO and CFO will be the ultimate arbiters. They speak in points, share, dollars, and return. When in France, I’ve found it’s really helpful to speak at least a little French.

What we measure it equally obvious. We measure the things we can understand and we can apply to our plans and, thereby, self-correct over time.

Finally, how we measure is actually the easiest of the bunch, but you’ll have to read further.

Social media metrics’ challenge has little to do with social media and whole lot to do with the way we’ve applied metrics, after the fact, to traditional media. It’s a rare claim outside of perhaps direct response that says, “She saw this spot and therefore purchased that product.” Period. No debate.

In baseball, I can see someone hit a ball into right field and watch someone score. I get that connection. In marketing, it’s just not that clean. What is well understood and documented is that exposure to a message increases the likelihood of a purchase, an attitude changed, or an openness to learning more. Make no mistake, each of these drives the purchase process. At the same time, each is dependent on external factors ranging from the economy (can I afford it?) to availability (do they have my size in blue?) to social factors (if I show up in this, what will my friends think?). All play a role and defy direct measurement.

Back to social media. One of last year’s pod camps, led by Beth Kanter in Boston, produced a nice list of the social media metrics that lend themselves to quantitative measurement.

Look down the list: page views, visits, time spent, search engine entry. While page views as a metric has its own issues (AJAX (define) programming technology and other technologies calling rightly into question the validity of the page view as a core contemporary metric), on balance these other metrics are easily obtained and tracked. Of course, the numbers themselves aren’t indicators of success or failure. It’s on this point that the quantitative measurement of social media has its biggest departure from traditional media. Simply, social media is a moving target. Traditional measures are absolutes: one million impressions is 10 times better that 100,000, at least up to the point of diminishing returns (a.k.a., saturation).

This leads to one of social media measurement’s fundamentals. Think back to trigonometry and differential calculus: trigonometry is all about continuous functions. Give me an “x,” and I’ll tell you the “y.” This works great for predictive models operating at a relatively macro level, think mass markets. Social media, by comparison, is about very small groups and the impact they have or feel, albeit, as a result of events in a larger social setting. Differential calculus takes over and makes the case for not only the measures that Kanter has listed but also for trending them over time to interpret them. If traditional media talks in terms of velocity (how quickly you can get your message out), then social media is best understood in terms of acceleration (the change in how fast your message spreads over time). Rather than planning a campaign with defined start and end dates and a certain spend that’s guaranteed to produce a specified exposure (reach and frequency), social media is an ongoing effort that builds and converges toward an objective. By understanding what’s happening now on the social Web and measuring over time, you can see the trend emerge. The dynamic trend, rather than static measures like reach and frequency, becomes the quantitative guidepost for your social media program.

How do you get started? Carol Meyers, Unica CMO, explains it this way in her blog: “When you’re in a competitive market that demands innovation, you have to get in the trenches to help innovation along” Relying purely on existing metrics isn’t necessarily the way to go; it tends to favor the status quo. Meyers continues, “You can be incredibly efficient at doing the wrong things.” Here are my suggested steps:

  • Don’t launch a social media program based on a hunch. Gear up first. Start with Kanter’s list and build from there. Measure everything you can. You’ll throw away a lot, but you’ll also uncover the core metrics that apply to your business.
  • Start examining trends. Get a sense of the current conversations, and track them. Unica, Cymfony, BuzzMetrics, and New Media Strategies can all help you. They have proven solutions and the business savvy to guide your next steps.
  • Based on your acquired knowledge of where your customers and prospects socialize, innovate. Follow Meyers’s lead, and make sure your sensors are turned on. Sure, you’ll stumble, but you’ll catch yourself and thereby set up your own success on the next go-round. Learn to use the new channels in the same way that pioneers of mass advertising learned to use traditional media.

In other words, don’t do what they did, do why they did.

In the end, the social tools and social media metrics are available. Start by measuring the current state, then pick your point of entry. Meyers’s blog cost her $175 plus 40 hours of her own time. It paid for itself on her first engagement. Your results may vary, but for $175 how far wrong can you go?

Related reading

How numbers affect conversions
A hand writing on a blackboard with INFLUENCE IS POWER written on it in stylized text, framed by a design of cogs and lightbulbs.