How the Pros Measure Social Media Marketing Success

From Mashable:

The notion that marketing costs can’t always be understood is an ancient one. John Wanamaker, a department store mogul who died in 1922, once mused, “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.”

The same could be said for social media marketing. Though there are a lot more metrics than Wanamaker could have imagined back in his day, the tricky part is determining the right ones to use. Is it retweets? Facebook Likes? Or maybe just product sales?

To get a sense of where social media marketing metrics are right now and where they might be going, Mashable recently contacted some people on the front lines. The group we chose includes:

• Sarah Hofstetter, senior vice president of emerging media and brand strategy at 360i
• David Rosenberg, director of emerging media at JWT
• Dexter Bustarde, senior web analyst at Digitaria.

From left: Sarah Hofstetter, David Rosenberg and Dexter Bustarde

A lot of campaigns seem aimed at accruing Facebook Likes. But is that success? What’s the point?

Dexter Bustarde: Earning a Facebook Like, by hook or by crook, is a success, but it’s a success at a very tactical level. If I have a client with a campaign completely focused on getting Facebook Likes, I’ll do what I can to show that this tactic should be working towards a broader strategy (using Facebook to tell people about your brand), which in turn should drive towards a business goal (getting people to spend money on your brand). Far too often, I’ll see Facebook campaigns that are very successful at getting a lot of Likes during the campaign, but I’ll have no concept of what to say to people after the campaign is done. When you earn yourself a Facebook Like, you’ve successfully opened up a line of communication with a potential customer and his or her friends. What you communicate after that is where we should look for real lasting success.

David Rosenberg: The value of a Facebook Like is directly connected to the anticipated actions the brand hopes its newfound followers will take on behalf of the brand. Is bigger better? If the brand is engaged in social selling or F-Commerce, then this more traditional CRM approach might lead a brand to goals such as include a large following. However, if the brand is looking to attract its most loyal fan base to use for product development, R&D and research, the quality of the fan reigns supreme over the quantity.

Sarah Hofstetter: The question you’re posing is actually interesting in and of itself. The question above referred to Facebook Pages as a “campaign,” but Facebook is a platform for ongoing conversation, which is really not a campaign at all. Putting your brand on Facebook — or any other online community — is an invitation for conversation, not merely a reach-and-frequency play, and it shouldn’t be measured as such. In fact, the whole premise of Pages is that they’re free real estate from Facebook where you still have to earn attention. Depending on your objectives, you may be interested in inviting lots of people to a continuous mass event, and you’ll even pay to get them there in media or coupons; or you’d rather focus on quality of engagement to foster a real relationship with the people who matter most, and invest your efforts in getting a continuing dialogue going between yourself and your consumers — and even better, connecting fans of your brand to each other. Ultimately success is defined by objectives


What about Twitter? How do you measure success there? If it’s the total number of mentions, how do you take into account negative sentiment?

Hofstetter: Twitter is interesting because the way people use it can vary. There are broadcasters, there are “listeners” who follow and read groups of tweets in some sort of aggregated way (celebs, friends, news, etc.), and then there are a select few who do both … and actually have conversations. Influence + mentions = second degree followers, which is why influence can be both a positive and a negative.

Rosenberg:There are different ways to measure success. If you are a news organization, watching a news link spread organically is a measure of awareness, engagement and traffic drivers. But if you are a luxury brand engaging in an influencer program, having the “right” people share your content may be your tactical goal. Having clear business goals and tasks will help solve for the approach a brand should take.

With regard to total mentions and negative sentiment, it’s a larger discussion. The use of conversation monitoring tools can help you get an estimation to some degree. But when dealing with natural language and slang, its important to hand-sample and depend on automation.

Bustarde: If we frame Twitter (and social media in general) in the context of real world conversations, it becomes a little easier to understand how we should measure successes or failures. You want to know a few things about how people talk about you in the real world. Ideally, you want to know that they’re talking about you to begin with. Once you know that they’re talking about you, you want to know what they’re saying and how much they’re talking about you. If they’re getting some piece of information wrong about you, you want to make sure that they get it right. If they’re complimenting you, you want to thank them. If they want to ask you a question, you want to answer it. When we look at it that way, we realize that success can happen at a bunch of different levels. Do we want to know if we’re successful at getting people to talk about us? Go ahead and measure the number of mentions. Are we successful at curbing negative sentiment? Well, that’s a trickier question. In the context of a real world conversation, if we overhear somebody saying something negative about us, we don’t (typically) leave the room and go write a press release. We respond and have that conversation about what we did to upset the other person. What could we do better? Trying to “measure” that process almost wastes time that should be spent actually talking to our not-so-happy customer. With that understanding, in the face of negative sentiment, it is much more straightforward and useful to hold yourself accountable to a success measurement like “time to respond” instead of some difficult-to-measure metric like “sentiment.”

Are there certain approaches that seem to work better for different media? For instance, if you’re a new brand trying to create awareness at any cost, does getting Charlie Sheen to tweet for you make sense? But how do you go about it if you’re already well known and want to change the way people think about you?

Bustarde: First off, if you got Charlie Sheen to do anything for you, you are working in the realm of warlocks and tiger blood — stuff well beyond traditional analytics. That said, the marketing person in me likes the idea of doing something to get yourself noticed. The crisis management person of me cringes at the thought of “at any cost.” If we really buy into the idea that social media is just a series of conversations scaled up to an almost-ridiculous level, we can go about changing perceptions by understanding how people are talking about us now and how we’d like for them to talk about us. For some brands expanding within the same vertical — say, a clothing brand branching out into shoes — this might be just a new topic that our social media presence touches on. For a brand undergoing a more drastic change in terms of new vertical or shifting perception, we may consider starting a whole new social media persona complete with a new Twitter presence, Facebook presence and so on to really distinguish it from what people used to think.

Hofstetter: The approach to social media marketing is the same that it would be for any marketing, except in this case people talk back at scale. How you align your brand in social media — and who you align with — would go through the same considerations as you would for any other marketing mechanism.

Rosenberg: The potential speed of the content spread may be alluring to some less risk-averse marketers. Twitter has the power to ignite fast content sharing, while more closed networks may make the recommendation of content seem more trusted and therefore worth spending time with.

A lot of Facebook programs require you to like the brand, which seems a bit disingenuous. Would it be better if Facebook provided another way for consumers to note their interaction with the brand without “liking” it, or does “liking” work pretty well?

Rosenberg: It depends on the value exchange. Achieving a higher level of status or entitlements may be worth it for some consumers assuming those levels or opportunities are really perceived as valuable. Consumers always have a choice, so the pressure is really on the marketers and their brands to over deliver on the value exchange.

Bustarde: Even if there were a way for customers to note interaction with a brand, marketers would (and probably should) make the effort to get people to “like” the brand. The disingenuous part is for a marketer to really believe that these people truly “like” your brand. You can pay somebody to pretend to be your friend, but that’s not to say they’ll be there when you really need them.

Hofstetter: “Liking” is a relatively low barrier, so if you communicate value up front and make it clearbefore they become a fan, they’ll oblige. Getting them in the door is just the first step, though. Convincing them to stick around is harder. Getting them to engage and share is even harder. A recent study showed many reasons why people “unlike” brands, and it’s largely tied to overmessaging and lack of relevance. Twitter lets you create lists of people or brands without having to follow them. This way, you get information without expressing a public affinity.

Do clients generally come to you with stated goals for a given social media campaign or do you have to educate them? I know it probably varies, but what’s the primary response right now?

Hofstetter: Relative to other forms of media, social media is still new for brand marketers. They know it can help solve their business problems, but some still get distracted by the bright shiny objects — objectives tied to large fans, huge followings and viral apps — so it’s up to people like us to demonstrate how this medium can be used to help them based on their brand, their consumer, the competitive environment and the nature of the medium. Sometimes that is a large number of fans because it’s clearly aligned with their objectives and they have the right brands and programs for it, and we do that every day for many brands.

Rosenberg: They have goals in mind and it’s important that we demonstrate that there are real, meaningful connections between what we can achieve in social media and what they consider traditional goals. Making those attributions is critical so that you can set the stage for proper measurement and prove success when you achieve it.

Bustarde: The most common goal is probably the most honest goal as well: Use social media to drive awareness of my campaign. The part that needs education is the strategic question: How do we do that? The trick then is in demonstrating that social media is successful at building awareness. Do we just look at web analytics numbers like referrals from social media? Well that’s only part of the story. The rest of the story is where clients usually have more challenging questions.

How do you see social media measurement evolving? Will there be a standard way to measure it soon?

Hofstetter: Albert Einstein once said, “Not everything that can be counted counts, and not everything that counts can be counted.” There is tons of data in social media — and in digital in general — and it’s up to us to make sure that what we’re counting is aligned with solving the business challenge. The trick with social media measurement is not to get wrapped around the axle trying to compare social media reach with GRPs, but developing proxies for brand measurement that are more aligned with objectives and how these specific programs actually help the business. We’re breaking new ground daily, and benchmarking becomes a challenge when you have nothing to benchmark against because you’re officially the standard. Our job is to make sure that we’re setting KPIs that are aligned with objectives at the outset of the program and coming up with metrics that are aligned with those KPIs and tracking success from there.

Rosenberg: There are some tried and true standards that need not be thrown away. It’s possible to derive multiple attributions from the same metrics. For example, commenting on a Facebook post can be attributed to awareness as well as, potentially, purchase intent if the post was directly correlated to a product offering. There will be an introduction of standards over time, but for now making these attributions will be the right steps to marry some of the older models with the new ones.

Bustarde: In the relative short term, I see social media measurement evolving into more distinct practices. Depending on who you ask, social media measurement could mean anything from PR and reputation management to Twitter reports to broad “engagement” measurement to looking at Facebook Insights day to day. In truth, all of those things should inform a social media measurement program, but at the same time, if we’re talking standardization, it’s a lot of work to get it all under one umbrella.

At a much higher level, I see social media measurement (and web analytics in general) gravitating towards standards that businesses can really get behind. Things like CPM can be applied to social media, but it’s been the extra promise of being able to measure things like sentiment and share of voice that have kept the really smart people from integrating very basic marketing needs in a truly meaningful way. If my message goes out on a TV commercial, a radio spot and my social media presence, we currently have very rough ways of measuring success on two of the platforms (TV and radio) and a hundred and one ways of measuring success on social media. Still, TV and radio measurement tie smoothly to things like ROI and deciding on budgets for a marketing spend and for whatever reason the simpler (and less accurate) measurements end up being more trusted. Social media will get there soon enough.