Measuring and Evaluating Your Social Media Effort
There is little doubt about the rapidly growing importance of social media. With more than a billion Facebook users, more than 500 million Twitter users, and 150 million LinkedIn users, these channels undoubtedly create opportunities for businesses not just to "talk" at customers but to "listen" and respond in highly relevant ways.
But doing social media "at scale" isn't easy. Social media isn't just one thing. It isn't PR. It isn't viral campaigns. It isn't Twitter-based customer support or help communities or brand monitoring or product launch tracking or evangelist blogs. It's all of these things and more. If you're treating social media as a single function—if you've created a single social media function in your organization—you're almost certainly doing it wrong.
The true value of social media lies in your measurement and analysis of it. Measurement is critical to both understanding and optimization in the digital realm. Are you getting a reasonable return on your social media investment? Which social media functions are most valuable to your organization? Most importantly, what can you learn about your customers and your brand from social listening?
The imperatives of good measurement are clarity of purpose, a good understanding of the levers that drive performance, and a fixed view of what counts as a success. Putting good measurement in place will force your organization to deliver on these imperatives, and along with them will come a much clearer strategic vision of what social media in your organization is for and what it can accomplish.
Measuring everything is impossible.
In complex marketing programs, the range of possible measurements is likely infinite. So if you can't measure everything, you have to choose what you're going to measure.
In marketing, appropriate measurement is driven by audience and function. If someone walks into your office and says, "Traffic to our Web site is up five percent," your first response should be "Who are those visitors?" It is crucial to know if they are customers or prospects or job seekers or students or affiliate trolls or robots.
Once you've established your audience, it's time to talk function. A customer coming to your site for support is fundamentally different than a customer coming to your site to buy more products. The difference affects the way you'll think about the traffic, the way you'll think about what counts as a success, and the way you'll want to measure that visit.
Your business goals need to be relative to the range of possibilities afforded by a customer, of a specific type at a particular point in his or her life cycle, relative to your business.
Knowing audience, function, and business goal is what ultimately narrows the range of applicable measurement. When you understand who someone is, what they are trying to do, and what your business goal is in that situation, the appropriate measurements of success will usually be obvious.
The Functions of Social Media
For some, the first task of a social media strategy may be to begin to make allocation decisions about which of these functions to invest in.
If you centralize the social media function, you're asking a single group to somehow manage a multitude of very different functions, the bulk of whose activities are owned elsewhere. That's hardly ideal. But splitting up the social media function into components that match potential business sponsors has its own challenges as well.
In many respects, this is similar to the effective organization of digital measurement and analytics. Centralization is essential to proper control and standardization, but effective execution needs to be driven by people who understand the problem, not just the technology or channel.
What might be surprising is that most of these functions require fundamentally different measurement techniques and technologies to measure their success.
Who are you measuring?
If treating social media as a single function is the worst mistake you can make when it comes to building a program, failing to think carefully about your audience is by far the most common. Interestingly, the two issues are deeply related.
Unlike traditional media, most social media influencers are topic specific. Social media has created many of these niche influencers and, within a topic, these influencers need to be treated as if they were traditional media or company spokespeople and be excluded from any consumer-focused measurement. This means that one of the most important steps in any social media measurement program is finding the right sample to measure.
There are two approaches to social media tracking: monitoring what people are saying and measuring their actions. Each has its issues.
If you concentrate on monitoring, you have no ready way to understand the value of readership. Social media participants are more often in the role of consumers than producers, but monitoring largely misses the "consuming" role. It would be nice to know that someone who read a post about your product bought it, even if they never wrote anything about you. With monitoring techniques, the only measure you have of impact is virality; namely, did something get talked about? Since talk isn't always the most important or most interesting behavior, the chance for misleading measurement and optimization is high.
Clearly, measuring actions would provide a deeper view of social media impact.
Metrics Buyer Beware
The whole point of social listening tools is that they listen to a huge variety of sources, from traditional media to the blogosphere to Twitter to communities both public and private. How can you report from so many sources without making life too hard? Toolmakers came up with a great abstraction that allows them to treat all of these sources as if they were identical—the mention.
The mention is a perfect abstraction, common across any communication. Tools can aggregate mentions, break them out by sentiment, divide or multiply to calculate share of voice, and trend over time.
However, there is no single social media measurement tool that can meet the full range of needs if you are tackling all or even most of the social media functions. The most versatile of social media tools only support two or three functions.
There is also a distinction between tools that can provide for the collection of data (the listening) and tools that can analyze the data. Many tools try to combine both functions; however, it is often better to disassociate them. It is essential to understand what functionality is important to you within the tool set, which tools play together, and which are competitive.
Many organizations make the mistake of assuming that their general-purpose social media listening tool can fulfill the customer research function. After all, these tools support data collection, keyword classification, reporting, and even sentiment analysis. Sadly, they tend to do all of these things rather poorly, particularly the keyword classification and sentiment functions. Since these are essential to the sampling and topic analysis at the heart of good customer research, they tend to be quite poor tools for the job.
The bottom line is that a large organization basing its social media strategy on a single tool is almost certainly making a disastrous mistake. The capabilities necessary to perform a wide range of social media functions simply don't exist in one tool or even one class of tools.
Social media offers an opportunity to create relevant and targeted messages that work toward business goals. As we explore the fairly new analysis of social media, we are determining what metrics to ignore, the metrics with real meaning, and how tools both help and hinder us in our task.
Gary Angel is cofounder and president of Semphonic, an independent digital analytics consultancy. He has led Semphonic's analytics practice and consulting for more than a decade. Previously, he worked in the credit-card database marketing industry, where he developed transactional analysis systems for marketing purposes.