“The days of build it and they will come… they are long gone”
How can you maximize acquisition and top of funnel for software companies (channels, tactics, team, consultants, CRO, etc) and not fall foul of the rapidly evolving battle between Google (and other search engines) and marketers? Rand Fishkin, Founder of Moz and Inbound.org shares great advice.
In such a fast moving area, it is no surprise that staying at the top requires focus and is exhausting. Rand also shares some of the personal issues that he has faced as a company founder – attachment to the business he founded, the problems he faced as CEO, how he came to terms with stepping down from the CEO role at the Moz.
Rand is one of the most interesting people in a fast moving industry. Bright, energetic, articulate, transparent and thoughtful, Rand is always entertaining, thought provoking and leaves people with great ideas that they can take and put to work in their own lives and businesses.
Slides, Video, AMA, Notes & Transcript below
Slide deck here if you are short of time.
Art Papas, CEO, Bullhorn
Tuesday 28 March 2017 at 17.00 GMT.
Don't Miss a Thing - Get BoS Updates
Want us to let you know about new talk videos, speaker AMAs, Business of Software Conference updates? Join the smart people who get BoS updates. Unsubscribe anytime. We will never sell your email address.
AMA with Rand Fishkin
Art Papas, CEO, Bullhorn
Tuesday 28 March 2017 at 17.00 GMT.
Don't Miss a Thing - Get BoS Updates
Want us to let you know about new talk videos, speaker AMAs, Business of Software Conference updates? Join the smart people who get BoS updates. Unsubscribe anytime. We will never sell your email address.
Rand: So, we have a tremendous amount to get through today. And I would like to leave some time for Q &A. So I’m going to dive right in. You are welcome and I encourage you to interrupt me with questions. That’s totally fine. You can raise your hand, shout stuff out. And we’ll hopefully have some time at the end.
So, this presentation this version of the presentation, you may have seen an earlier version of the presentation that I shared at some point but this version, is right online. So as I’m going quickly through slides and there’s URLs and stuff you want to reference or send to your marketing team, your CMO, you can do that easily here. And if you’re thinking, “Well this is probably going to be a crappy presentation”, don’t worry; this URL will be at the end in case you change your mind about that. By the way if you love my presentation, my name is Rand Fishkin on the form. If you don’t like it, I’m Chris Savage. Chris. Savage. Savage with an “S”. [laughter]
All right. So what is it that’s changed in the last few years around the SEO world? Well, actually we have had a dramatic amount of change. One of the big things is that you can see much more complexity in ranking algorithms. So things like: Advancements in user and usage data signal. Raise your hand if you’re familiar with ‘pogo sticking’. Do you know what ‘pogo sticking’ is? As it relates to search? Okay. Well let’s chat briefly about pogo sticking. So this is a behavior that we have all the time. We perform a search query, we enter our search into Google, into Bing, probably into Google [laughter], and we click on a result and the page starts loading and we decide, “This is a pile of crap. I am not interested in this.” And we click back to the search results and we click on another result. And the search engines, all of them, call this ‘pogo sticking’, they measure it carefully, and they take it as a very bad sign. If lots of users that they can authenticate as being real don’t try and create fake bot accounts for this, don’t worry, Dave and I have tried, it’s not a thing, it doesn’t work. I’m kidding by the way. Well I don’t know if you have; I haven’t. [laughter] We know people who have. If they see lots of people clicking on your site in lots of different results authenticated users that they know and trust – and clicking back, and then clicking on someone else’s result instead? And appearing to be satisfied with the query there? That number one ranking is not going to last for very long. That’s going to go away. That’s one of the user and usage data signals that they’ve got. Remember that over the last five or six years Chrome has become one of the most dominant browsers in the market, Android has become the dominant mobile operating system. So the amount of user and usage data signals that Google can collect today, compared to what they knew five or six years ago, is multiple orders of magnitude difference. And because of this, if you can’t serve up a great experience you are not going to perform well in search.
Google’s also gotten massively more sophisticated – and Bing has gotten somewhat more sophisticated – in spotting editorial versus manipulative links. So it absolutely is the case that links are still powering a lot of rankings out there. Anyone who says links are dead for SEO, I can promise you we have done tons of testing this year, just in the last few months, around the power of links; they are still incredibly valuable. Here’s an example. So we’ve done this experiment many, many times: we took just a few pages that were ranking in the 30s or 40s of results, pointed anchor text links from ten different domains at them. Boom, they’ll pop up to page one. Even for relatively competitive queries. We’re still talking about a very link-based search engine. Google actually said through one of their representatives early this year that they looked at removing links from their algorithm. They did not like the results. So they’re not doing that. What they have done – and what they’ve done very, very well – is get an ability to spot manipulative links. Links that are not editorial – that were paid for, that were cajoled in a way that Google doesn’t like. They’re not perfect at it. There’s still plenty of link spam out there, but they are aggressive and they’re getting better and better.
This is sort of interesting. So keyword matching. Let’s say I went to go see the lemurs movie in 3D at the IMAX. No, it wasn’t in 3D. Yes it was. The lemurs were like jumping out at you, right? Yeah. We went to go see it with Geraldine’s – my wife’s – cousins who were like eight and ten. And it was super adorable. The movie’s great but I went home and was like, “I want to go visit that park. What was that place?” I didn’t think it was a park so I searched for “national area with lemurs madagascar”. And you can see what Bing is doing. Bing is doing something that Google did for many, many years, until last fall when Hummingbird came out. They’re essentially keyword matching. They’re trying to find the pages on the Internet that have those words that are the most relevant, best match. What Google is doing is much more sophisticated. This is sort of topic modelling. Google is saying, “Uhh, national area with lemurs. We can interpret that query. We can interpret that query to mean something that we can then return the result ‘Ranomafana National Park’ for. “That’s not keyword matching, that’s intent matching. Google knows the intent of what I want. Go try searching for something like “movie with the dude”, “movie where guy offers a blue pill versus red pill”, “movie where two ladies drive off a cliff”. They will get them. They will return to you the results of ‘Big Lebowski’, ‘The Matrix’, ‘Thelma & Louise’. They’re on it! That’s intent matching. It’s a very different kind of keyword-based search and it means different things for us as well. I’ll talk about that a little bit.
There’s personalization. By device, by history, by location. And by Google+. Even if you think you don’t have a Google+ account, you almost certainly do. Even if you don’t use it, they’re using it. I can virtually promise that. And I’ll show you how that impacts things in SEO.
I think it’s kind of fascinating, too, that SEO…My water is gone. I’m going to take this one. I need to announce that to all of you in case you were to come up here and take that water later
Audience member: That was mine.
Rand: That was yours? [laughter]. That’s not going to go well for me in the ratings. Chris Savage, with an “S”. [laughter] So, this is an analysis of LinkedIn profiles in the U.S., and this is very different from five years ago, ten years ago. These are U.S. profiles with ‘SEO’ in the job description. These are profiles with ‘SEO’ in the job title. SEO is not really a job title anymore. It’s a job description. And I think that is because we as SEOs have found that doing just our job is not good enough. As the engines have become massively more sophisticated and searches have become more sophisticated, and the competition has heated up, you can’t just do keywords, links, crawl rankings. Which you could for about ten years. From about ’97 to about 2007. You could really do four things and be great at SEO. You can’t anymore. And so SEO’s need a lot more influence in the business, in the marketing team, in order to have success. So they are not calling themselves ‘SEOs’ anymore. They are ‘digital marketers’, ‘website marketing managers’, ‘directors of inbound marketing’ is a common one now.
Google has done some things that piss me off, frankly. The list is long. I won’t go through all of it. Web spam policing is something that we didn’t have in years past. In years past, essentially if spam pointed to your site, and you weren’t responsible for it, you had nothing to worry about. Essentially there was nothing to sweat. Today spam can hurt sites even if they are barely responsible – not really responsible for it. Like Google has essentially said, “You should go check out the links that are pointing to you on a regular basis and sort through ‘em, and we’re going to give you this interface in Webmaster Tools, which is very confusing, extremely hard to use, very hard to sort through. And then we’re going to make it your obligation to go check those out, because if you get bad links, we might hurt your site. For a very long time for it. “So this has made these non-editorial links, while they’re still potentially powerful, they’re also more dangerous than ever. So if you find one of these loopholes where you go, “Oh look, I know a guy”, and it’s the equivalent of the guy right off Broadway in New York City and he’s got the trench coat and he’s like, “You want tickets? I got some links for you right here.” True story, my grandparents actually still buy their Broadway tickets from a guy like that. He has a trench coat. With tickets in it. It’s unbelievable.
This means web spam is every site owner’s problem. If you are seeing that you’re getting a ton of links from sources that look very sketchy – that really can hurt you. And that is new and different. It’s also very sad because I think it disproportionately affects especially small and medium businesses both online and offline that aren’t sophisticated about SEO that get someone reaching out to them from the SEO world and then they buy their own links. We do see some efforts from what I would say more nefarious players in the search field to try and influence their competition negatively. However and it’s a strong however I would say 99.99% of those fail completely and some of them work the opposite of how they intend. You point a bunch of what you think are really bad links at someone, sometimes their rankings might rise, and Google might not think it was their fault, and you’re like, “Shit, I just helped my competition out..” It’s also expensive and time consuming going and trying to hurt one competitor versus trying to help yourself. There’s a ton of reasons not to do this. Some people do engage in it. I’ve seen enough credible one-off examples to believe that, yes, it does happen on occasion, but it’s super rare, very hard to do effectively. And Google is usually fairly forgiving once the site owner takes notice and sends in their reconsideration requests through Webmaster Tools. So I wouldn’t worry about this too much. But it is an unfortunate, and I think really frustrating, side effect of Google’s change in behavior. [pause]
This is an interesting one, and potentially a very frustrating one for content creators. I’ll show you what I mean. So, I do a search for ‘Boston population’. Huh, I don’t really need anything else, do I? This data is from a website that Google scraped. Google scraped Census data, they purchased Freebase, but they scraped a lot of the entity data you see over here, that’s actually from Wikipedia, who they also scraped, who they’re also not giving traffic to. Here are NFL scores. Like, “Screw you, NFL!” [laughter] I bet the NFL actually loved that traffic. I bet that was actually useful and valuable traffic for them. They could cookie those users and then they could retarget and remarket to the ones they wanted to, and they could give that data to the teams. That sucks for them. No traffic for you, City of Boston! Because look – I search for Boston art museums; I don’t need to visit any of these sites, including the City of Boston’s actual museums and galleries listings page, because Google has scraped that page, found all of these listings, and there they are. And if I click on any of those, guess where I’m taken? To their Google Maps page. There you go. Google sending traffic to Google for Google’s benefit.
Rand: Yes. It doesn’t fall into the ads model. Excellent question. I’m about to answer it. I’m about to answer that. First I’m going to take another steaming crap on Google for a little while. [laughter] Here we go. That is extraction, right? They just scraped that off the page – that’s text – and they pulled the text in there, put it up there. It’s not even the right answer! I mean as frustrating as that is, we went and updated when we saw this was happening. It doesn’t send us traffic, so we have no way of noticing, but some folks tweeted at us and saw it, we went and changed it and they still haven’t changed it six months later.
Hubspot founders, no traffic for you, Hubspot! And this happens in some not so simple queries too. They try and extract out the answers from pages, and give those. Poor Expedia. Or Kayak. Yikes. This is sucking up a ton of their valuable traffic. This is why I think they’re doing it. I believe they are doing it because they need to drive up the volume of search queries that happen. And the way that they can do that is by addicting users to search. And the way to grow addiction is to provide faster and faster answers. So Google has talked many times about speed of search-query to answer-to-query as what drives people to come back again and again and make more and more searches. We’ve grown in the last few years from 3 billion searches a day 4 years ago to, today, more than 6 billion searches per day. And by the end of this year, Google says half of those will be on mobile. So when you’re on your mobile device, you are even less patient with clicking through and finding that listing for all the Boston art museums, finding whatever it is, going to Expedia and trying to use their interface, trying to get answers like this. And my supposition is that when Google does instant answers, what they’re trying to do is addict people to search. They want people to be so satisfied by Google that A) they never leave and that B) they come back and do more and more searches. And that’s what they’ve seen.
Audience member: Is it also part of their intent to try to drive away people from using apps?
Rand: I think they care a little less about. They do benefit a huge amount from the app world. But I would say one of the things we’ve thought for a long time was that apps were sort of going to eat the mobile web, and that has turned out not to be true. We still actually use mobile web very heavily and search is one of the biggest things – I think the biggest thing – we do on the mobile web. I don’t like the statistics that show mobile apps eating mobile web when the mobile apps are Angry Birds or Words with Friends. To me, that’s Game Boy usage, not mobile device usage. I think there’s a little confounding variables going on there when you see those kinds of things.
Essentially when you see this kind of data, what I believe is going on, is yes, Google is doing something that is potentially shitty to Expedia, or to Moz, or to Hubspot or whatever, and taking some of that traffic. But in the end, it might be a net benefit. Because it’s a smaller chunk of an overall bigger pie.
Audience member: A net benefit to…?
Rand: To content creators. To the web. Because we’re essentially now saying, look. A lot of those quickly answered, in some cases less valuable – not in Expedia’s case but in many cases – less valuable queries and searches that happen in the head of the demand curve are owned by Google now. But the chunky middle, and the long tail of search demand, is growing so dramatically, that content creators overall actually have a bigger opportunity in those areas.
Rand: I think probably indirectly it ties to that. Pogo sticking is a strong signal because Google knows how frustrated searchers get and how quickly they expect answers, especially nowadays. And so they’re trying to prevent that behavior entirely by just extracting out the answer for searchers as fast as they can.
Audience member: I don’t want to derail with a debate over whether this is cool or not, but as a user this is awesome.
Rand: Yes, and that’s exactly what Google is counting on. They’re counting on users to say, “This is awesome. I love Google because they answer my queries so quickly. Therefore I’m coming back again and again.” Therefore, hopefully, smaller chunk, for content creators, of bigger pie overall. Whether that’s true or not is very hard to say, that’s me speculating.
Audience member: I understand the narrative for Google wanting more personally, but if a disproportionately large percent of that.
Rand: I think they would argue the reverse. What they would say is the more information we have about people the better we’re going to do overall. So we don’t care about exposing people to ads nearly as much as we care about knowing everything about them. And then we can charge advertisers more based on that.
Rand: I think that’s why it’s so weird to me that they’re being so aggressive about this, because a lot of times they are showing the wrong answer, as I illustrated here. And so I think it’s odd.
Audience member: The larger pie argument. Is that because I’m so delighted by the results I get that I search more and get more brave by trying things I wouldn’t click on otherwise?
Rand: It’s not just braver. All the research shows that essentially your neuron pathways rewrite themselves to where you no longer ask your friends, go to Twitter, look on Facebook, talk to human beings, like, you just search for everything. If any of you go out to dinner with your friends, how many times do you pick up that phone to do a search to answer a question about something? Like, you get addicted, addicted, addicted. ..And they know that addiction grows. So that’s what they’re counting on.
Audience member: Isn’t there may be a bigger picture going on, I mean Google just bought the A.I. engine to search for cats and here we are answering all the world’s problems, and here they are…
Rand: And they’ve said their dream is the Star Trek computer. It just knows what you want and gives it to you. It can predict what you want. Google now, they can predict that. So I think there’s probably a connection with that.
[Same audience member] So can content creators stay ahead of them?
Rand: I think content creators can benefit most by – I’m actually going to talk about that more later in the presentation – but yes, I think we can.
All right, I’m going to race ahead a little bit. Number five, social media. This is from 2006. This is Forrester. I used to stand on stages in 2006 and 2007 and 2008 and I would tell folks, “If you want to perform well in search and SEO”, and what was then called ‘social media’, Digg and reddit and that kind of stuff, Stumble Upon, “You need to talk to, and you need to be attractive to the creators.” That 13%, I would often call them ‘linkerati’. They’re the people who can influence. Today, because of the ubiquity of social media, that group has grown to be nearly three quarters of all adult Internet users. This is pretty phenomenal. And that number will only grow. So it used to be, I need to find the influencers, and reach out to them. Now it’s, we’re basically all influencers – or have the potential to be influencers.
And number six. A lot of SEO data has disappeared in the last few years. Keyword referral data: nearly gone, unless you are a giant site. This is again one of those things where I think Google is benefitting the top 1% of websites at the expense of the rest of us. So, ‘keyword not provided’ is now 95% of search referral. On average, I think it’s actually 87%… 12% is not good enough for any site with fewer than, say 200300 thousand visits a month from search, to really make any guesses on. But if you’re the New York Times, that 12% is plenty. You can make a lot of smart decisions based on the keyword referral data you still do get, based on that sampling. AdWords data. I don’t know if you guys saw, but a few weeks ago they said mid-September Exact Match will be gone. There will be no more ability to specifically say, “I want this ad to appear only when this precise keyword in this precise format shows up.” And for a lot of the advanced and power users of AdWords that sucks. And for a lot of keyword research, that also sucks. Google will now be using what the call ‘close variants’ in all Exact Match.
Groupon did an interesting experiment recently. They say it was an experiment – I’m pretty sure they messed something up. [laughter] They shut off their site to Google. They basically banned themselves from Google for about a day maybe a day and a half and lost all of their search traffic and 60% of direct traffic. As soon as the search traffic came back, so did the direct traffic. And in fact, they were so perfectly correlated with the pages receiving traffic that Groupon published an article about it on Search Engine Land that you can read saying, “We’re 99% sure that 60% of our direct traffic is actually search traffic that’s miscounted, that’s not carrying the refer string from Google.” And by the way, this was just Google, not Bing and Yahoo. So, very interesting to know, when you look at your direct traffic, some significant portion of that is probably not direct. A significant portion is likely search.
So what does this mean for us in the SEO world? For those of us who are pursuing search as a huge channel? Like I said before, great SEO is not just these four things anymore. Great SEO today means having an influence in designing user experience, in customer experience because of user and usage data. In new channels like email. In mobile. In social. In content, absolutely. In internationalization. In web speed. In localization. All of these things now fit into the influence of SEO, either directly or indirectly. And so we need some ability to influence them if we want to be as successful as possible in SEO. And when you look at companies that are doing a phenomenal job, they’re ones that integrate all of these things into their SEO strategy. Or integrate their SEO strategy into all of these things. And as marketers, our job is to choose those tactics that marry our strengths with our opportunities. And I’ll talk a little bit more about that. I’m of the belief that any of these tactics that I showed you might not be right for you. But I do think it’s really important that we’re all aware of the arsenal available to us.
Art Papas, CEO, Bullhorn
Tuesday 28 March 2017 at 17.00 GMT.
Don't Miss a Thing - Get BoS Updates
Some tactics to leave behind. Some to embrace.
When it comes to figuring out the things that your target market is searching for, for which they may find you, for which you may hope to be found… I hope you’re thinking broader than just traffic that converts. AdWords-only discovery was for a long time the standard. We’d go, we’d use AdWords, and we’d do our keyword research in there. AdWords is actually hiding a tremendous amount of keyword data. A remarkable amount. So when you start typing things into Google, they will show you a lot that doesn’t show up in AdWords unless you specifically search for it. In my opinion when I search for ‘bar inventory’, AdWords should say that ‘bar inventory app’ is a popular search term. A very popular search term. And yet it wasn’t showing for me. Unless I searched for ‘bar inventory app’ specifically in which case it would show me, “Oh yeah that does get 12 thousand searches”, I don’t know how they could have left that out before. [laughter] I really like this new tool; it’s from some folks out in Malaysia, keywordtool.io. It will do all this suggest stuff, sort of scrape all this suggest data for you. You can export it yourself. You can send it to your SEO folks. Hopefully your SEO folks are already using this. It’s become quite popular in the last 30 days.
Rand: Yeah I like it a little better than Übersuggest. Actually I like it a lot better.
Audience member: ?
Rand: Just UI export features, that shows up in Excel. I think most of the data in there, and the number of searches you have to run compared to Übersuggest. I also think it’s very important because suggest won’t even show you everything, to actually go out and research how your target market talks about the product or the problems that they have. So for Partender, which is this bartending app, which I’m using an example of, go to the reddit bartending forum, look at the questions people are asking, that is your opportunity to find new keywords, new content to potentially create.
Speaking of content creation, this has been ubiquitous recommendations in the SEO world for forever. You create ‘good unique content’ and the engines will do the rest. I think this is crap. I think it’s a terrible idea to make the bar for your SEO your content be ‘good, unique’ content. You know what’s embarrassing? Back when I was a consultant I actually did consulting for eHow and WikiHow and now they’ve just turned into the biggest spam farm. All right. So, this is a long, long page from Angie’s List that ranks number one plumbers in Medford, Oregon. You know what I see a ton of start-ups and businesses doing? They go, they look at this page – the page of results – they find whoever’s ranking first and they’re like, “Okay we need to make one of those.” Let’s go make one of those for all the potential areas we can target, all of the products we have. This is the only unique content on that whole long page I just showed you. Massively long, just goes on forever, this is all just these addresses, those ratings, and those two numbers: 50 and 36. That’s it. That’s the unique content. If you think that you can build pages like this, that you can just replicate Angie’s List’s strategy, and rank, you are dreaming! Granted, it’s a beautiful dream with like a sailboat and clouds, and there’s a whale with birds. It’s beautiful. But come on. This strategy will no longer work. And a lot of the time when you look at who’s ranking well, they are ranking well based on things that have happened in history, not things that you can simply replicate and get again. So I’m urging you to leave this behind, and to think about content with these lenses applied. First off: It does need to be one of a kind. You need content that doesn’t appear elsewhere on the web. That’s a very low bar. But it is a bar. It must be relevant. Meaning: content that the search engines can interpret as on the topic to the searchers – both their phrasing and their intent, what they are trying to accomplish. It should be helpful. It should actually help the searcher resolve their query. Not lead them down a hole, not sign them up for an email thing. If you are counting on, “Hey, I know that we’re going to rank for this and then only a low percentage of people are going to actually sign up for our email form because we don’t actually give them the answer but that low percent is good enough”? That will not maintain your ranking. That user experience is poor. Google will get rid of you. Uniquely valuable. This one is very challenging and often misunderstood. When I say uniquely valuable I don’t mean ‘one of a kind’. I don’t mean ‘unique’. I mean provides value in a way that no other result – or very few other results – in the searches can. I have a great example; I’ll show you in a sec. And finally, great user experience. It should be easy and pleasurable to consume on any device. These are the criteria for modern content. I think if you can’t hit this bar, you’re out of the game. I would worry about the relevance of the page to the search – to the searcher’s phrasing and intent. As you’re creating content, let’s say you have bird watching tours you offers all over the country, you want to use the phrases that your searchers are using. That will help you perform better in search. And it will help you perform better when the user lands on your page.
A fascinating study I didn’t get a chance to include, because it just came out, is one from a conversion rate optimization firm Optimizely, I think it was Optimizely, and they basically tested versions of an AdWords landing page that had exactly the title as it was in the search results, versus a slight modification – any modification – from the search results, meant more people bounced. More people didn’t stay. When you see ‘Bird watching in Portland, Oregon’ and you click that result, if that headline on the page, not the title of the page necessarily, but the headline of the page, if it doesn’t say ‘Bird watching in Portland, Oregon’, you’re going to bounce back. So, sort of interesting.
Here’s a great example of this content from genius.com, Rap Genius. Ironically genius.com was banned about nine months ago from Google for doing some super sketchy stuff. Can someone remind me, are there a bunch of people who work at Google who invested in Y Combinator and Rap Genius? Probably just coincidence. [laughter] So look at what genius.com is doing here: one click to play it on YouTube. That’s a great experience. Super great experience. Accurate, readable lyrics. Do how many lyrics sites absolutely, completely suck and they hide a bunch of them and the UI is just terrible? There data regarding the popularity and activity so you can see how popular this song is with all these people. There are uniquely popular annotations. For example: Did you know that when Bob Dylan was writing ‘Shelter From the Storm’ and he says ‘I came in from the wilderness, a creature void of form’? That’s actually that toil being reminiscent of Adam’s punishment in the Garden of Eden while being void of form is similar to the world on a first date. I mean Bob Dylan’s a genius. Like, we knew this. He can’t sing [laughter] but he clearly can write lyrics. I’m kidding. I love Dylan, but… [mumbles] I went to a Bob Dylan concert a few years ago and couldn’t understand a word he said had a great time. [laughter]
What about link building? When it comes to link building a lot of the old school, classic SEO was like, “Oh yeah, let’s get some directory link and then we’ll go do some link bait, some click-through bait.” This Boozin’ infographic was from a financial insurance company and that pointed back to them. And we’ll scale up our guest posting efforts; we’ll get a bunch of links that way because we can control the anchor text. The weird part about link building is while it’s still hugely critical to earn links, and to have links in order to rank, Google doesn’t actually want to count any links that you could build. If there is a place on the web where you can go and you can do, just something, and get a link back that you control how it looks and where it points to, they don’t want to count it. Google wants to count links that you editorially ‘earn’. Which makes sense. And so, my recommendations would be to build relationships. Relationships have a funny way of earning links. Over time, there’s a high correlation between relationships you build as a business, as a human being and people who point to your work and pay attention to you. Hence, relationship-building has become my primary link-earning tactic.
This fascinating email from the place that cut my hair – you see this ridiculous hair cut? – it’s because I did that thing where you go in and the guy who cuts your hair has a super cool haircut and you’re like, “I want your haircut!” and then it looks way better on him afterwards. And you’re like, “Dammit!” So my haircut place did a partnership with a local sex toy shop called Babeland, and this relationship resulted not just in them sort of sharing customers and tips for haircutting – and other things – but also their websites linking. Beautiful, wonderful. [laughter] Partnering with sex toy shops might not be your preferred relationship-building system of choice, but that’s for you to decide.
Content drives a ton of editorial links. This is my wife’s blog. This is Geraldine; she went to Cambodia recently she went without me, I was very sad but I had to build this presentation among others – and wrote about it. This is run by The Wildlife Alliance out in Cambodia. Very, very cool. So she published this post and I can’t convince her to do SEO. It’s the weirdest, most ironic thing. There’s nothing I can do to get Geraldine to do any SEO, virtually any marketing of any kind. But, I’m on it. I’ll do it for her. So I tweet it, I just told the wildlife rescue, “Hey, y’all should check this out, ‘cause she wrote a great blog post about your facility outside of Phnom Penh.” And what do you know? Not only did they retweet it; they put it on their resources page – their ‘about’ page. So now they link over to her blog post. Relationship-yielding social sharing link.
These pages are great. If you’re thinking, “Oh, I want to get links from directories”, no, you want to get links from these kinds of pages: these resources pages, recommended stuff, blog rolls. That stuff still works wonderfully. And Google’s always going to want to count this. This is an editorial endorsement. You’re not paying to get a listing or applying for a listing or that kind of thing. If you’re having trouble scaling this out, you can buy it. You can spend money to get links, as long as it’s indirect. Google has no problem with you – through their ad network or someone else’s – buying exposure for your content that leads to links. The platforms Outbrain, Taboola, Zemanta…they all have pretty similar names, don’t they? That’s kind of funny. There’s like a cadence to it. And Facebook of course, exposing your content through these content advertising platforms is huge. And right now the prices are dirt cheap. Because all these people are relying on advertising revenue to pay them back they are not thinking, “Oh wait, I can rank higher with my content that will earn me people who convert.” And so this process of putting out great content that people are going to love and engage with, that will get shared and linked to, that will rank, sharing it on ad platforms and getting return from that – both directly and indirectly – very few people are doing this. You basically take a piece of content that you know is appealing to a certain audience, and is likely to earn their engagement – their social sharing, links from them – and you broadcast it through these content advertising platforms. That could be on Facebook, which kind of happens on and off Facebook if you want. Taboola and Outbrain are sort of the two most popular content advertising networks right now. So you go to menshealth.com and you scroll down to the bottom of an article and it’s like six articles about getting six-pack abs, and some articles featuring people wearing not enough clothing, and some celebrity thing. And then it’ll be like how SpaceX won the contract to deliver people to the whatever – and that is advertised by SpaceX, because they want to expose it to people. That stuff is content advertising.
So more link-building resources for those of you who are deeply interested and or want to send to your teams some tactics that are still good still work. Social and SEO is kind of interesting because in the past it was in the case that we used to be able to share socially and see that directly influence rankings. From 2010 to 2012 Google and Twitter actually had a partnership! Do you guys remember that? Where you’d do searches and stuff from Twitter would appear directly in there and you could connect up your Twitter. I thought it was awesome; I loved it. But the companies just couldn’t make the deal work – continue to work – and so it fell apart. So be very wary when you see stuff like this, “We prove that tweets do affect rankings.” I agree with this article in that tweets indirectly affect rankings, but I strongly disagree that when you look at correlation between tweets and rankings, that that’s actually meaningful. It could be completely the other way. It could be that because something ranks well, it gets lots of people clicking on it and then tweeting it. So I’d be cautious about that. Social is an indirect influencer. When you create content and share it with an audience who is likely – or potentially likely – to create links to your site, that’s when you have opportunity.
One exception to this rule: and that is Google+, which we talked about in the introduction. So this is Mark Traphagen, he’s a genius guy – I think he’s based here in based here in Boston actually, works at one Stone Temple, one of the best SEO consulting firms in the area. And because I follow Mark on Google+, whenever he shares or ‘plus ones’ content, that content ranks higher for me. Interesting side note: even if I didn’t follow him – because we email so much, through my Google account, and through his, Google might indeed actually bias to show me his content anyway. Even if I’d never clicked that ‘follow’ button. Because that email connection had created something. This is powerful. If you were to tell me, “Hey Rand, I could spend a bunch of time and energy doing all sorts of good things for SEO, or, I have this alternative, I could press this button and get all my potential customers to follow me on Google+”, I’d be like, “Just do that!” Because right now if you share something on Google+ and it has the words – just the words that the searcher is looking for in the element of the bold item text of the item that’s shared on Google+. It’ll just rank. Google’s biasing to Google+ content if you’re logged in as a searcher which something over 60% of searchers are logged in, to a Google+ connected account tis cheating! It’s cheating at SEO but, like, state-sanctioned cheating. Google-sanctioned cheating. It’s just weird. Incredibly powerful.
So historically we’d be like, “Okay, I need to rank for stuff, I just got to publish, publish, publish, publish, put out content.” That’s junk. I actually really hate the…. I get the idea of a content calendar; what frustrates me behind content calendars and editorial calendars is this idea that I have to publish every day, or three times a week. If you don’t have something great, why bother? Because links and shares basically looks like an income distribution chart. [laughter] Here’s the top 1% and they’re getting a ton of that, and the top 5%, and 10%. If you’re in the bottom 75% of content that’s created on the web, it’s just not even worth it. Not even worth it to publish in my opinion. I don’t think you get anything out of hitting that button.
If you’re going to make investments in content marketing, and in a content strategy, then I think you need to answer these: The content you publish must be strategic and relevant – it actually ties to the business goals and it fits with your branding. It targets likely amplifiers. Granted, this is a much bigger audience than it has been historically but you have to have an obvious answer to the question ‘who will help spread this content and why?’ This is true where you’re doing SEO or not. If you are doing any type of content you should have a great answer to that question, and if you don’t, you should re-work it until you do or you shouldn’t bother. And content isn’t a fire & forget process either. Many, many folks have their content team, or they have a person who does content for them, they hit publish and then they’re done. They think that’s the end. But it is not. It is not. Long investments and a long period of time. Virtually everyone who I’ve ever talked to who’s done phenomenal things with content. I mean, Joel, who long at Buffer before you guys felt like, “Hey we’re good at this content stuff!”? A couple of years. Moz is the same way.
Yeah. Go back and look at like, the first year of Whiteboard Friday. It’s embarrassing. That guy is terrible at it. It’s awful. The camera’s all shaky – it’s like handheld and it’s grainy. I’m literally holding a whiteboard, like one of those ones you can just buy for ten bucks, and trying to scribble on it. It’s awful. But you get good at it. Eventually you get good at it. If you’re looking for inspiration for creating great content, or for seeing what’s been successful in your field, check out Buzzsumo. You can do a search for virtually anything and you see the most shared content on the web in the last year or six months, or three months, or month – that has that title in there. So I looked for ‘waste treatment’ and I can see sites that have actually gotten thousands, hundreds of shares on content around waste treatment. This is page two, in fact. So you can browse through, you can find stuff that works. I love that tool.
All right, the last thing I’m going to talk about. So before 2012, I could look at my traffic and my performance like, “How is search performing for me? What is the value of SEO to my business?” I could really look at that by looking at my keyword reports. But post-2012, 94%, whatever, 87% are not provided – this really doesn’t help. I can’t do it anymore. And so instead, the alternative is to measure the value of the search traffic to you by page – by the page people are landing on. How well does this page perform in converting search traffic or in bringing people back or in being somewhere in my conversion funnel? I can look at the pages that receive search visits to get that data. You can also use estimates that exist inside some of the SEO software packages: Search metrics, if you’re more enterprise, Bright Edge and Conductor. And this screenshot is from Moz’s own software where all of us will try and predict, using our own internal algorithms, which keywords are sending traffic to which pages.
Most companies are still using this attribution model, and I think this is quite dangerous especially in the start-up world because it biases you to do dumb, dumb things. I will give you an example: I’ve talked to several start-ups where they say, “We used to invest a lot in social media – we had a Twitter and a Facebook page, we were getting Pinterest because we were seeing some traction there – but we looked and actually social is almost never in our conversion funnel. It’s never last-click attribution. And sometimes it doesn’t even fit in any of these. And so because of that, we stopped investing in social.” That is a conversation I have had many times. Social is just one example by the way. Content can be in there; blogging can be in there. What happens is you forget you’re stopping filling the top of the funnel. It makes sense for a model like that to get much more credit, but in fact, the channels that are not frequently part of your conversion funnel are often how people find you initially – how they get branded to you – and branding means familiarity, means a higher likelihood of conversion. This is just a long process. It can take six months, a year, for someone to go from ‘yes I’ve heard of them’ to ‘now I’m going to subscribe to them’. It’s been my experience too that – and this is so weird to say as someone who’s been obsessed with measurable marketing for basically my whole career that the best return on investment is often coming from non-measureable serendipitous marketing. Otherwise why would Wistia have – what’s that van that’d driving around out front with your logo on it? Like a big billboard? You will not be able to measure that. There’s no way. By the way, we use Wistia for our video hosting. Especially since this terrible presentation was given by me, Chris Savage, founder of Wistia. [laughter] It’s definitely the case…
Audience member: I was just wondering how many people are not customers of Wistia?
Rand: How many are NOT customers of Wistia? [laughter] Well, there you go. Thanks to this audience, it’s now measurable. So I think a lot of people who are watching this presentation might go, “Shit, this SEO stuff is hard. That’s going to be a deep investment if we make it.” Very, very challenging. The nice part about that difficulty is that is creates a barrier to entry. If you get good at search – at SEO, at organic traffic – we’re talking about incredibly low customer acquisition cost, a very scalable channel – one that Google is growing for you, because they’re growing the volume of searches, unless they’re taking your stuff away – and if you can succeed, it’s a way to run away from the competition.
So with that, you can download the presentation there. And I hope you have some more questions, we have a little bit more time. Thank you very much.
Art Papas, CEO, Bullhorn
Tuesday 28 March 2017 at 17.00 GMT.
Don't Miss a Thing - Get BoS Updates
Mark: Rand, if I may? I’ll ask the first question. And it’s less about link building and SEO, and more about you as a founder and a CEO and a SEO. Incredible journey. How does it feel?
Rand: Uhhh. Mediocre. I cannot escape my own head, and that’s not a fun place to be. Yeah, I recognize that externally a lot of what many of us do seems like it should feel incredible, but in point fact it just feels like… not good enough. So, I hope in years to come that I’ll find more happiness around it. But I really geek out when I –
Mark: You absolutely will. I know lots of people in this room have been in that situation where they’ve set something up and gone off and done stuff and it feels less attached. Who’s been through that?
Rand: I know, I know. And Moz is, it’s doing well, it’s gotten a lot better, but it’s frustrating. Yeah – in the back?
Audience member: This is not related to your presentation, but years ago you published your ‘Adventures in Venture Capital Land’.
Rand: Yes, yes that’s right. A couple of times.
Audience member: It was a wonderful thing. It showed that one paying customer had to support 39 freeloaders, basically. What was your motivation for doing that?
Rand: So this is actually a funny thing about all of my sharing. I think a lot of folks look at me, or look at folks like Joel or Chris, and see essentially a content marketing strategy. But in point of fact that’s not the case. You probably can’t see it, but I have a giant hole in my chest. And that hole can only be filled temporarily for a few seconds at a time – by lots and lots of external praise. Hence, I am absolutely addicted to blogging and social media sharing and chronicling our journey. And I have to transparent – it’s in my DNA, I can’t disassociate myself from that. If I have something that goes well, I have to share it; if I have something that goes poorly, I need to share that, too. And so it started as much less of a content strategy and much more personal sharing that I have to do. It turned into – and it turned out to be – a phenomenal way to get people aware of Moz and to build a customer base. I think if you search for ‘VC funding’ we still rank on page one. That was one of my things; I’m like, “Yeah none of you fuckers invested in me. I’m taking your rankings.” [laughter]
Audience member: You had a slide about the components of SEO in 2014 – the complex matrix of things that make it up – I think there’s one missing .The ‘https’ which has recently been added as a component and it will probably give you a little boost in the search rankings but if you use extended validation SSL search on your site, it will give a little perception of improved trust in your brand and I think there’s some components and…
Rand: There are. Google has said it’s a very, very tiny ranking factor. So we’ve actually – not me personally, but many folks in the SEO community – have been looking into how much does ‘https’ really affect SEO? And a lot of people have said and seen the same thing, which is, “We’ve seen more people mess up the redirection and lose traffic from getting it wrong through the migration from http to https than we have people benefiting from the switch. And so if you’re going to do it, you have to make that migration right. And that’s pretty challenging. But yes, https can provide a small boost. Okay – whoever has a microphone?
Audience member: Hi, Evan from Point-of-Rental. What are your thoughts on PR Web for SEO purposes? And then for general brand-building purposes?
Rand: Poor in both cases. The only thing I would say about press releases is, like anything else, you need to be able to stand out from a crowd. PR Web is the way to stand inside of the biggest crowd of crap you can stand inside of. And so I would never recommend that. I just don’t think reporters pay any attention to those. And having an article published on PR Web or places like it, recently Google has said those may be actually considered manipulative links that can hurt you. So if they see that thousands of your links are coming from press release sites that you pay for, and no reporters or journals or anybody who’s picking those up, they could actually hurt your site because of that. So I’d warn you to stay away from those.