Lost in Translation
Robert Gourley, Creative Director at Mojave Interactive, recently released a “social media report card” that gauged 4 big North American bike brands (Trek, Cannondale, Cervelo, and the one I work for) on their “social media performance” during the Tour de France. I appreciate what he’s trying to do – I really do – and in the comments he astutely points out that they’re not out to make sweeping proclamations, since he’s doing this as a (totally valid) way of promoting his own Social Media Analytics services (and they’re probably very good). But this post is written in a way that is more likely to attract non-cycling customers by wowing them with the cool of cycling and the promise of social media services. As a cyclist and cycling marketing guy, a few minutes of critical thought started to quickly unravel the specifics of his report.
As just one example, his assertion (made in the comments) that his report leads one to be able to see the relationship between paid media spending vs social media conversations is simply false: nowhere in the report do they discuss paid media spending (nor is this data published anywhere else). Further, they then go on to rank the four brands with their own proprietary “social media index” which seems a bit watery, based on their descriptions. Granted, this is a service they’re trying to sell, so I guess it makes sense that they’re not about to give away all their measurement secrets. Call me crazy for being skeptical though.
Overall, I think the report is interesting, but oversimplified to the point of being irreparably flawed (at least in so far as any conclusions may be drawn from it). Subsequent interpretations that have been drawn from it suffer similarly, maybe more so. Let’s recognize what this is: a demo sales pitch for a media company, not a decision making tool for the bike industry.
This oversimplification is mostly because of two simple points: attribution, and context.
Attribution As just one example: the vapid “mentions in blogosphere” graph offers neither source information, nor the methods by which they filtered specific terms like Cervelo and Cannondale against more common synonyms such as Trek, or the even more google-unfriendly term Specialized. These are notoriously difficult terms to search against. Plus, the scale on the y-axis of the “mentions in blogosphere” graph references a peak of 150 – if that’s the absolute number of mentions they found, this graph is nothing short of violently under-indexed. Either that, or this entire report was merely indexing my own personal twitter feed and nothing else. Further down the report, their 5-point x-axis score on the Performance Summary graph approaches poetry in it’s inspiration. Why 5? Why not 10? or 12? What are the units? I suspect it rhymes with marbitrary.
Context As the often-clever Josh Kadis pointed out in the comments of this article: a brand like Cervelo rises to the top of search terms because they’re title sponsor of their own team (which is, for sure, an enviable position to be in!). Further to this, if you’re relying on search terms to index comparisons here among brands, it will fail: too many of the most important conversations happen without mentioning the brand directly at all. Consider interviews with Andy Schleck, or photos of Pozzato, or retweets of @lancearmstrong. All entirely valid conversations, many entire devoid of branded search terms like Trek or Cannondale. To put it another way, do you imagine you’d find yourself more often saying casually to your friends “I like my car” versus saying “I like my 2008 Ford Focus SVT”? Measuring social media relies on understanding how people actually talk to each other, rather than how brands sometimes wish they would. Sadly, this makes it very, very hard.
So, enough with the whining – I hate whining, especially that sound of it echoing inside my own head after I say something whiney. What *would* I like to see here, as a guy who works in marketing at a company that is experimentally active in a wide array of social media outlets? Here are three things I can think of that would help (ie we’d consider paying for):
- A way to monitor the social reactions to specific single actions: if a single tweet earns us (or costs us) 35 followers on twitter, we should know what that was so we can do more of that (or avoid it).
- A way to measure the effects of our corporate social media efforts at a retail sales level: blog comments and tweets are great, and a good source of feedback, but are we doing anything more than conversing with the converted? How are our ‘corporate’ social media efforts helping our dealers, if at all? And what specifically should we be doing more of to incresase this positive effect?
- A way to measure social media efforts of all our retailers, versus all the retailers of our competitors (or retailers in competing industries). This is where rubber meets pavement (or dirt), and I strongly believe that social media offers far more promise at a widely dispersed micro-regional level (ie the collective power of 2500 different bike shops) than it ever will offer to the global brands that support those same dealers.
Despite the fact that our myriad efforts at Specialized sometimes get us accused all kinds of things (we’ll never escape criticism, ever), we are still exploring this space because we believe it has real, demonstrable value. Even more: it’s fun, and it’s honest, and it’s authentic to the cycling culture that we’re all part of. Cycling is social, and so for us, social media is something we see as having the potential to do something that’s different than the print ads, Pro Tour sponsorship, and other legacy marketing programs that the industry relies upon. And it’s NEW. It still has that smell of new vinyl and those plastic static-clingy screen covers on everything. So we’re still experimenting, and doing our best to pay close attention. And if we make a mistake, we find out about it pretty darn fast, cuz we’ve got thousands of fans/followers/friends out there that tell us. And we appreciate that!
As one final thought: why is the assumption with social media that it’s necessary to “win” – given that conversation topics change daily, or hourly, does it really matter if Cervelo scores 4.5 bizwits versus Cannondale scoring 3.75 zeehas? Social media, if it is really engaged discussions, can probably help the entire industry make the sport cooler and more approachable. What if we all just looked at this phenomenon as a rising tide, and realized that all of a sudden it’s not the crochety mechanic at the local shop with the only opinion that gets heard? In the long run, this is probably good for the sport more than it’s good for any brand in particular.
So, I offer every respect to Robert Gourley, and I mean that; his blog post has sparked lots of interesting discussion, and I suspect that his company does fine work. But at the risk of sounding arrogant, I’m not going to pretend that his blog post taught me anything about the cycling world. It probably didn’t teach you anything either. Now, as my old friend Eric G used to say: get back to work.
UPDATE: within seconds of posting this, saw that Josh also posted another interesting series of points around this topic. Read that too.
9 Comments »
RSS feed for comments on this post. TrackBack URL
mmm… Nerds…
There’s also the question of influence. Some blogospheric mentions are more important than others.
From my perspective, Specialized (i.e. you) seems to be doing a better job at social media interaction. FWIW.
Does Cervelo (the bike company) even have a social media presence?
Any “report card” or briefing is an attempt to drive business to one’s agency . . . so that’s generally a non-starter — if anything trust a report from Forrester or Gartner as that’s all they do. Nielen does them as well for Interbike.
Aside from the validity of Social Media Experts, a larger issue in bike blogs and social media is whose influential, doing it for reals v. to get free bike parts and hookups. Doesn’t take much to figure out “hey if we post on X it goes up in Google because Y company’s website sucks and Google can’t index it.” Bicycling.com is a case study in how blogs beat old websites. That’s not just blog love, but simply how Google works.
Bike blogs are amazingly vibrant online and so are other outdoor enthusiats blogs (baseball), but that’s driven by the bloggers themselves and not the companies whom we all know aren’t spending much, if at all. Chris Mathews and I debated last year Specialized’s Walled Garden approach and we’re always saying do we need yet another bike community?
How is Greenlight Ride from Civia, Cateye Commuter, and Pedro’s Plus 3 doing? An agency can pitch a branded community much easier than a strategy across social networks where the conversations are. Because the social strategy is much harder to measure. No one has ever been fired for running an ad in a magazine vs. a risky branding, product placement effort.
It’s very encouraging of late, about 3 years in, for the bike companies to take note. My guess is like other industries, a maverick in the community started reading blogs themselves or publishing them. Our concern now, just like with recent FCC issues regarding mommy bloggers is pay to play and influence.
PR companies and the clients they serve have readily figured out it doesn’t take much to impress a blogger and gain a favorable review.
Twitter as a disrpuptive technology is certainly going to push companies when they see their athletes on there and they response they get, let alone themsleves. Lance is above 3M followers.
Cervelo likely suffers from what any other company does in Social Media — it’s a lot of work, sustained work. You can’t just set it and forget it, no matter how big you are. Take their flickr photostream as an example — they just toggled it to bloggable and if they opened the license, you’d see it take off. Or actually responded to comments.
One other hole I can see in this report is the simple fact that you guys make an effort on the social media front, whereas I can’t even get permission to start a Facebook page, or a Twitter account to represent my job.
-CG
Whoa: skinny margins made my comment look massive! One correction, I meant SRAM on Flickr . .. Cervelo is not on Flickr.
Nice, Chris. Good to see this in-depth discussion happening in a bike industry forum and not just on a marketing site. The marketing world has been struggling to determine the metrics of social media (and, frankly, internet advertising in general) success and it’s good for the bike industry to be a part of the discussion.
I would agree with Josh on the importance of influence and also the concept of engagement (much more significant than how Mojave uses it in their report).
Very well written Chris; as usual. I don’t share your pessimism though. As you point out, more or less, you don’t have access to the model, inputs, sources, weights, philosophy, purpose, audience, or other relevant factors (e.g. marketing spend figures, conversion metrics, etc.) which I think have to be considered if critiquing at your level. To me, at face value, the Mojave information looks solid. It’s pretty…
“Awww, come on guys, it’s so simple. Maybe you need a refresher course. Hey! It’s all [ceramic] ball bearings nowadays.” – Fletch (1985)
And as Tom DeMarco said, “You can’t control what you can’t measure”. So it has to be done. Has to. Only way out is to get in there and make (or buy) even better methods (weighted for Influence as well as Engagement, and obviously to make Specialized look a little better, or possibly even to induce integrated marketing campaigns to perform better).
I’m reminded of my old college book “How to Lie with Statistics”. … Hum? So all this ambiguity is really GREAT for marketing! This is the golden era. You can measure social media as you like, and the evil dark side of the insatiable marketing machine can rise to full force. Mwaaaahahahah!
On behalf of all consumers, we don’t want your push-marketing infecting our social media. And if you irk us, we will change our habits, and drop you like a … Rockhopper.
“…but what I won’t do is play ball.” – Mr. Pink, Reservoir Dogs (1992)
And as a side note, if social media marketing ever wants to be taken seriously, you need to somehow preclude spam categorically (and particularly porn spam) from within any given channel. Otherwise anything proactive (vs. just measurement) is just as much noise.
I wonder if Heisenberg’s uncertainty principle comes to bear in social media marketing analysis?
And what’s up with Craig Hummer mentioning Twitter like 10 times during the tour? Who paid for that? Pretty soon announcers will be just as desperate for viewers to go Twitter something as they are now to drive people to the VS. website and type in special words to PROVE they were driven there – to be converted – to drive revenue.
All good points, thanks to all for the comments, and thanks too for the compliments – though I assure you I’m not doing this alone! lots of people at Specialized are active within social media, they’re just doing it organically (which is kinda how and why this stuff works). As things progress, we’ll show each of them a little spotlight love when & where it makes sense.
For sure, the value of social media editorial is not static – different sites carry different levels of influence, and have influence in different areas (some are big, some are specific, some are both). No different than magazines in that regard. That was Josh’s first point, and it’s a good one.
Compelling and succinct social media info – http://tinyurl.com/n9av66