4 Things I Learned at MozCon 2015: Day Three
by Anvil on July 31, 2015Uncategorized4 Things I Learned at MozCon 2015: Day Three
The only thing I would change about the third and final day of MozCon 2015 is the One Direction concert that was scheduled at CenturyLink Field on the same night and made us miss our train back to Portland. Seriously; we gave ourselves forty minutes to cover the 1.5-mile distance from our hotel to King St. Station in a taxi, and after thirty of those forty minutes had passed, we looked up and found we had traveled all of four blocks. Consequently, I’m now typing this from the passenger seat of a Kia Rio, rented at SeaTac and driven by my intrepid, entirely-too-tall-for-this-subcompact-iron-curtain-clown-mobile colleague Alex Gabriel, who volunteered his body for that task expressly so that I could carry out this one, so, thank you, Alex, and now, please enjoy our feature presentation.
That first paragraph was as far as I got in that passenger seat before carsickness took hold of me. I wrote the rest of this from my comfy, comfy office after our return, hence the delay in its publication, for wich I sincerely apologize. And now we will return to our feature presentation with no further interruptions.
- The best way to begin a keyword research process for today’s semantic web is to make an array of mind maps, each of which starts with a central keyword and branches out to propose related keywords. This one came courtesy of Gianluca Fiorelli, a particularly active member of the Moz community who nonetheless had never addressed MozCon until this year. His reputation as a keyword strategist is illustrious, and particularly impressive when you consider that English is his second (third? fourth?) language and still predictably commands the bulk of his work portfolio. His 15-minute talk covered a rather wide range of keyword strategy tips, but the one that stood out to me as most valuable was this mind maps recommendation, which reminded me that we all tend to reach for our tools (e.g. Keyword Planner, Ubersuggest, etc.) too early in the keyword research process (while also reminding me of how “brainstorming” was originally taught to me in the third grade). In this new environment in which simple string-matching is out the window and pages are now expected to explicate some topic in exhaustive depth, the better way to start the process is to pick one word or phrase that well represents the central premise of the page (or whole site, or whole brand), and do some simple free associating to map out the different words and phrases that naturally spring to mind when you think about that word. Granted, sometimes the industry into which we’re delving for a given keyword research project is esoteric and we’re just not conversant enough in the lexicon to know where to begin, but under most circumstances, this will work and will make for the strongest possible starting point. You will certainly be expected to use your tools eventually — there could easily be some more popular synonym for one of the words you thought of — but these tools are not powerful enough to do the real creation here. I’ve written before about how search is starting to operate like the human mind itself, so the act of trying to think like a search engine thinks is becoming quite a bit more intuitive. Just let your mind do its associative thing.
- People respond to emoji the way they respond to real human facial expressions. This eye-popping presentation from Courtney Seiter of Buffer, entitled “The Psychology of Social Media”, was chock full of bombshells about how our new favorite pastime (social media) is, to use her words, “changing our brains”, but this one was the most jarring to me: studies are showing that the sight of emoji stimulates activity in the same centers of the brain that the sight of human faces does. We react to a happy-face emoji like we would react to a smiling friend, and to a sad-face emoji like we would to a loved one in tears. This probably accounts for the explosive and growing popularity of these little guys, which every year take up more space in texts and social posts that might otherwise have been occupied by words (Seiter’s presentation cited the stunning statistics that 74% of social media denizens use emoji regularly, and that 6 billion emoji are transmitted every day). If I’m being honest, I think this is a good thing. Indulge me for a second. If you’ve ever been hanging out at a bar with me and one of us has received a confusing text message, you’ve probably heard me launch (unbidden, most likely) into a spiel about how the rise of text messaging and social media as our preferred communication media — and the attendant fall of face-to-face interaction and even the telephone —has been screwing us up emotionally. This claim is rooted in the oft-cited Albert Mehrabian 7% – 38% – 55% rule, which stipulates that in any face-to-face interaction, only 7% of a speaker’s message is conveyed by her words, while 38% is conveyed by her tone of voice, and a full 55% by nonverbal cues such as body language and facial expression. [This formula and the studies that gave rise to it have certainly been challenged over the years, but as a simple piece of anecdotal evidence to support the larger concept, recall any past experience in which you tried to use sarcasm in a text or a Facebook comment.] This is a tough time to be alive, psychologically, and I cannot help but think that one of the major reasons is because we are social creatures, and lately, our brains aren’t getting some our most traditionally dependable social signals the way they used to. For the first 99.99% of our species’ evolutionary history, face-to-face communication was the only kind possible. Then, about 120 years ago, we introduced the medium of the telephone, which robbed us of the 55%, the nonverbal cues we’d come to depend on so heavily, and then, only in the last five years or so — an eye-blink in evolutionary terms — we started moving to these text-only platforms and losing the 48% belonging to tone of voice as well. I submit that this forced collapse of human communication to 7% of its former spectrum is a major contributor to the larger feelings of disassociation, isolation, and bewilderment that characterize millennial life. If the use of emoji really is taking off like a rocket, and it is, couldn’t it well be because it’s giving us back at least a simulation of that 38%, thus making our textual communication not only clearer, but more emotionally nourishing? Neurological studies seem to be suggesting as much. [Now recall how much easier it became to use sarcasm in texts and on social media once you had the option of tacking on the winky-face at the end.]
- Facebook reviews show in Google rich snippets. This will seem like an awfully trivial little factoid after that last one, but at least it’s about marketing. Apparently, Facebook reviews are now being interpreted as structured data and output as review rich snippets in Google search, just as Google reviews and Yelp reviews are. I had no idea about this, and now I do, thanks to Moz’s own local search expert David Mihm, who shared this knowledge as part of a larger presentation about the multitude of ways in which one can use Facebook to succeed in local SEO. Drive those reviews, people. It’ll do a lot more for you than soliciting page Likes.
- Google is harvesting all sorts of post-click user behavior data, and using them to influence rankings in all sorts of ways. Moz’s founder Rand Fishkin traditionally closes out the proceedings with a nuts-and-bolts talk that distills the major findings of his own independent research since the previous MozCon, and most of his talk this year was about how strikingly sophisticated Google’s apparent grasp of post-click user behavior has become. The fact that they’re collecting these data should surprise exactly no one — what exactly did you think Chrome was for? — but what is surprising is the breadth of behaviors that Rand has reason to believe they are bringing to bear on rankings. Here are a few examples:
- Relative CTR. I’ve been musing on organic click-through rate as a ranking factor since it placed #1 in SearchMetrics’ study last year, perfectly content to tell myself, and clients, that if a page ranked #3 has an “exceptionally good” CTR, it might get bumped up to #2 or #1 at some point for that reason alone, while also admitting that I don’t have the foggiest idea what constitutes an exceptionally good CTR. Thankfully, Rand shone some light on the basis of comparison that Google uses for these kinds of evaluations. Apparently, in a scenario like the above, Google isn’t comparing the CTR of the page ranked #3 to the CTRs of the pages ranked #1 and #2 for the same query; it’s comparing the CTR of the page ranked #3 to the average CTR of a typical page ranked #3 for a typical query (obviously there’s a data set of finite size that they’re using here, and obviously we don’t know anything about it). So if you’re ranked #3, your goal is to get more clicks than the average #3 result. Granted, this knowledge doesn’t necessarily change anything about how any of us would approach optimizing for CTR (write really good metadata! no, like, really really good!), but anytime you glean a little more insight into how Google thinks, you’re better off. [This claim cited a research project by Moz’s Philip Petrescu as evidence.]
- Query terminus potential. Even though I just made that phrase up and decided to use it because it sounds cool, it still refers to something real, I promise. What it refers to is the likelihood that a page might satisfy a searcher so completely as to mark the end of her quest. How many times have you Googled something, clicked through to whichever result grabbed your eye first, felt disappointed in the page within one second, and hit your back button within two? How many times have you had to try three or four search results before you found the page that fully satisfied you? Maybe it was ranked #6 and therefore took a while for you to find your way to it, but once you did, your search was through. Well, if 10,000 people searching the same query have the same experience of having been flummoxed until they hit result #6, and they’re all using Chrome, that page isn’t going to stay at #6 for much longer. Google wants to serve results that help searchers fulfill their tasks in full. Take your search audience all the way to the finish line and you’ll be rewarded for it.
And with that, two weeks following its conclusion (again, do forgive me), MozCon 2015 is officially behind me. Thanks for sharing the adventure with me, and let’s wait and see what next year has in store!