INBOUND 2016: Matthew Barby - "How to Suck at SEO & Drive Your Business into the Ground"

We all having a good time at Inbound. We're pumped to hear about SEO. That one always gets the crowd going. Okay, so this is my third talk today. So if I like just pass out on the stage, just take me away. Follow the deck, it's easy to follow, it's mainly lies and just keep going with it. So I'm Matthew Barbee. For those of you that don't know me, I look after SEO globally at HubSpot amongst acquisition across our sales products. I'm going to talk to you today about something you don't often hear at conferences and this is how to drive your business full force into the ground using SEO as the vehicle to do that. This is how most people think SEO works. Lots of people come in, they're like, okay, organic search is going to be the main driver of growth for our business. The monies go up as the time goes on. Simple hockey stick growth, right Challenge that comes with that is just mainly counting the money as it comes in. You like paper cuts, you got to store it all. Decide what am I going to buy things with. It's problems now actually this is how it really works. You tend to have a bunch of uneven spikes in traffic. Maybe you had a successful piece of work nearly you've had a successful piece of content, bit of growth. Maybe you got a bunch of social shares that drove a bunch of traffic. And then the come down after is like, okay, we're back to our normal traffic again. It's like a little bit of growth underlying and this slowly over time starts to culminate into a nice steady bit of growth. Now obviously you can get back to rainmaking with this, but it's not something that you should necessarily plan for. So here's what I'm going to show you. I'm going to take you on a journey from Gates to Blockbuster. We're going to go and nose dive. Whether you choose to directly take my advice and crash your business or do a combination of maybe the complete opposite, it's your choice. I'm not going to tell you exactly what to do here. The goal for me is to get you to here. So get the notes out. I'll be checking. Following your progress, you're probably thinking, Matt, you know, kind of like having a job and a business, kind of got kids to feed, etc. All this other stuff. But let me tell you some benefits. Remember that guy you hired, right? There's just something about him that you're kind of like, I just don't like him. Fire him. He's gone. Dave is out of your life completely. And all his friends, they're gone. Next up, you're going to be the butt of everyone's jokes. But joke is on them. Because what's the secret? All PR is good pr. This is only going to elevate your profile. Don't forget that probably the most important side of this is the career development that you're going to get right. You're going to learn a whole bunch of new skills for all the job applications that you're probably going to need to put through. So let me get started. I'm sure you sold into the benefits of this. Now, number one on the path to destruction, only focus on content. Create great content, everybody. This is something that I hear so many conferences on, so many blog posts time and time again. 2.5 million blog posts. This is just WordPress actually are published every single day and haven't done the exact math on this, but I'm going to say roughly 2.45 million of those are complete shit. Open up your Twitter feed. Have a look at some of the stuff that's on there. These are the people you choose to follow. Half of the stuff on my Twitter feed is horrendous. Right? These people are the ones you choose not to follow. So all the aspirations of just creating great content and the results will come. Can't be working for everyone. Here's a few examples. Right, so this is a piece of content, actually what I would kind of like band in. It's more than just a screenshot. I just couldn't quite emulate the web browsing experience in the slide. This is a piece that I band into the great content category. It's a huge guide, actually. Puts together 33 different ways genuinely that you can go and save money on your groceries. And it's actually perfectly in line with their brand. It's really good. Long and short. His Results. A whopping 61 social shares. Four backlinks from two different websites. Not really groundbreaking. They didn't really share this much at all other than on their Facebook page. They shared it once. On their Twitter, they shared it once and that was it. Why spend all the time doing all this to get those results? I know, Kanye. It's bad. Here's another one. Big interactive green roofing guide. Woofing. No. So loads of bespoke photographies went into this. It's like a bunch of interactive elements. Clearly would be useful to the people that are going to be looking for something like this. And a lot of money's gone into it these guys took this to a new level. They didn't even share it on their Facebook page. They didn't even share on their own social media. They're like, okay, look, let's pile in the money, let's get the photographer, let's go create this big piece of web content. It's gonna cost a lot of money. Okay, Bob, could you just go like post this on Facebook? No, I've done enough here. Right. I am work to the bone. This is just gonna work. It's great content, right? I can't believe it either. So I've been spending some time. I like to think of myself as a bit of a math man devising a formula. If any of you are not big on algebra, I'm going to break this down. Content minus promotion equals failure. Tweet that one out. You heard it here first. Yeah. Basically the long and short here is that actually you need to do a hell of a lot more than creating great content. No one ever goes out with the ambitions to create shit content. Like, I'm not, I'm not going to purposefully do that. Of course I'm going to go out with the purpose of creating great content. Doing it is another thing. And then promoting it is a whole other thing. And this is why the promotion's important. This is an analysis I did of all of the content across HubSpot, sales blog, marketing blog, around 16,000 blog posts. We want to understand, like, what is it that our top performing content has that makes it so good? Well, mainly backlinks. That's what makes it really good. Most of our top content, which is a very small subset that generates the large majority of all of our leads, usually has around at least 300 links. And that takes a hell of a lot of promotion to actually do. Backlinks matter. They matter so much in SEO. All of the blog posts that you see, it's like links are dead. Oh my God. It's all bullshit. This is like one of the most important parts of ranking in the search engines. This is just one part, by the way, that you need to be thinking about and actually like you. Yeah, getting backlinks is pretty difficult, but there's a few things on a basic level that you can be doing. You can be writing for industry publications and blogs. I'm not going to go into too much detail around this because it's almost like a separate talk in itself. Responding to press alerts from journalists. If this is something, if you create a bunch of content, especially if it's data backed stuff like that some quick wins that you can have is signing up to press alert services. So if you've ever heard of haro h a r o help a reporter out response source if you're Europe based SourceBottle, there's a bunch of these where you can get requests from journalists every day sent through to your inbox for sources for their stories. Great way of getting an intro to journalists and also some quick links and coverage and traffic, et cetera. Share content within online communities. Another like staple part of any kind of like online promotion, finding relevant communities to share content with. If there's no like dedicated communities within your industry, you can either one make one or go to a more general source. Reddit, Quora all of these places are great for promoting content. Second big important thing to do to destroy every last piece of your business is completely forget about the user. Right? I was actually talking about this earlier today and also this has been a big thread within the product keynotes yesterday is moving away from creating content for machines, for the search engines. The search engines know how to consume your content. They're way better at doing it than even humans are right now. But if you want to forget all that. Does this look familiar? Right? How many times have I pulled up a blog post or a services page? It's like we're an inbound marketing company that sells inbound marketing services to inbound marketing teams using our inbound marketing expertise. Inbound marketing, inbound marketing. Rank us, please. Right? And it's absolutely rubbish. And then all of a sudden you get hit, Bam. I'm going to give them my credit card details and you convert. Of course. It's a killer tactic, right? Yeah, no, that's not that great. Every, every search that's ever done starts with a need. Whether that need is specifically focused around like purchase intent. I want to buy this thing. Or I just want to understand some more information. I want to know where the best place to eat is, et cetera, et cetera. All starts with a need. After that need is there. There is a search that happens. And whenever you search for something as a searcher, you have a kind of expected experience that you're thinking you want to get from this. Right? Let me start with a quick example. So I'm not your typical manly man. I own a drill and that's about as far as I'll go. No idea how to really use it. But today I'm going to become man. I'm going to build a shed. It's about the manliest thing you could probably do. I Would say, can I create this shed? So I'm like, right, I need to figure out how to do this. So I search how to build a shed, come through into the search results and I'm like, yes, this is my ticket to manliness. I'm going to do this. I can see it. The fans are going to be cheering at me as I hammer in the last nail. Here we go. Then, bang. It's like, buy a shed right now. Like, no, that doesn't make me a man. That's what I do now. I just buy it now. And I don't want to. Not now, George. And you're all like, annoyed at this, right? Because why is that happening? And this is the kind of feeling that your visitors come into your blog when you're not matching user intent properly are having. And what you want to understand is a basic understanding of the types of searches that people are having. So the first one being informational searches. Someone in the example of how to build a shed is looking for information. At this point, they're not ready to convert right there and buy a shed. Maybe after they tried to build it and realize, shit, this is pretty tough, I'm just gonna buy it. Of course, then that's probably a point when you can capture them, but at this stage, they're just looking for information. Next up, we've got kind of like what we call, like navigational searches. People know what they're looking for and they're using the search engines to go specifically to it. But it's not necessarily stuff to do with like commercial things. So it's usually searching for a brand. Maybe someone's looking for like Urban Alphas or searching for HubSpot, for example, Right? Navigational stuff and then band in transactional queries, specific purchase intent. Usually these queries have things like buy or a price, maybe even a color if it's retail and it's next to a product. These kind of things is where actually you want to be showing. The kind of web page that I was just showing around buy sheds probably wouldn't go with that exact design. The Comic Sans may put people off, but that's the kind of general thing you want to be making sure you're matching against each other. Google really does care about this. This is not just me saying, hey, this is better for users, right? This is also better for the search engines and I can prove it. So I've been doing a lot of tests around how much does Google actually look at when people come through to your site and they have a negative Experience like, do they downgrade you? If I, like was searching for how to build a shed, I come through to this web page and I get the buy a shed now and I bounce back to the search engines. Does Google care and vice versa. If that happens the other way around, people are staying on my site. Is that a good thing? Right. So what I did is done a couple of experiments and here was one of them that we did. So we took a brand new page on our website and has no backlinks, nothing else associated to it, and we took it for a keyword that it was ranking like the middle of page two for document tracking. What I did is a highly mathematical way and scientific manner. Got 300 HubSpot employees to use their cell phones on their cellular data to go through, search for document tracking, find our result, click on it and browse the page. Like they would mimic a good user experience, maybe visit another page and then leave. But don't go back to Google. One hour later, we were at the top of page one. That was pretty conclusive. So I now have a team of 300 people. I'm joking. It's not quite that much. What this is telling us is that this really does matter to Google, right? As well as Google, this really matters to your users. So this is a conversion rate test, an ongoing project we do on an ongoing basis. And if any of you here in the room, I'm assuming a lot of you have content offers that you'll promote to capture email addresses. Right. What we did is we were saying, okay, some of our ebooks that we have, we have a lot of ebooks on our site, if you haven't noticed already. Some of them get downloaded way more than others. And actually it wasn't really to do with certain ones that were getting more downloads being down to the fact that they were better content. Some of them in our opinion, were way worse. So we were starting to dig into why is this. And here was an example. So this is an article that was on the blog called how to write a press Release. And you see in the bottom right corner, always wanted to use this laser, this little CTA here we had the newsworthy guide to inbound public relations, like press release, PR related. That should work, right? What we did is we had a little look using Google Search console at what were the most common keywords that people were coming through to this blog post. What were they searching for? The most common thing, press release template. So we're like, okay, well loads of people want a press release template come through to this thing. And it's kind of like teaching them how to build a press release template, but maybe people actually just want the template. So we create a press release template, of course, and we changed the title to completely match the thing that people were searching for most, matching this expectation completely. What we SAW was a 240% increase in conversion rate. And that 240% increase was an average increase across the first 80 blog posts that we did that with. It was by far one of the biggest content optimization plays that we did. And every single month we go through and have. Pam Vaughan, who sits on the optimization team, does this on an ongoing basis, reoptimizing content, squeezing more out of it. It's actually a really simple exercise that you can do across a bunch of your blog content that's already existing to squeeze more instead of having to create some new stuff. Third thing, probably the most important, do not test things. You do not want to know when things go wrong. It's pretty crap feeling. Ignore it. Just rely what you read online. Most things are probably credible, right? There's like 2.5 million people posting stuff. It's got to be credible. Why would people do it? There's a lot of content out there from credible sources that is pretty incredible. Some of the advice that's still online from blog posts that have been around for maybe like five, six years old are very outdated, of course. But a lot of people consume content on big sites like kissmetric saying, yeah, let's go after and just publish a load of stuff in general directories. This is a terrible idea. WikiHow, one of my personal favorites for terrible advice. If you ever need bad advice actually just go to wikiHow. Article submission is a little more time consuming, but it can help increase your website's page rank. It's all terrible advice, but so many people fall into this. And I think the fact that and complete disclaimer like there's probably bits on the HubSpot blog. Like I was saying on the marketing blog, we've got like 50, 15,000 blog posts. We don't maintain every single thing that we published from like 2008. Most important thing here is test whatever you think could be a good idea. Even the things I'm saying right now, do not trust me, please do not do that. Test things. And this is exactly what we do. It's like, okay, loads of people are saying longer content, long form content ranks better in the search engines. Well, does it for us? Does this do this for our content? Running some like basic analysis into what's performing across your site? Organic traffic versus word count. It's like rudimentary stuff. You can go way deeper than this, but at least you're testing this stuff out to see what you should be investing your resources into. Understanding which topics get us the most links so that we can prioritize what we're doing and start shifting towards the topics that are going to get us most bang for our buck. If backlinks is what we want, we do the same thing with leads. We do the same thing with social shares. What words do we tend to find that are in our blog post titles and our Twitter shares tend to result in the most tweets. So now we just put template in every single thing? No, we don't do that. Test things or don't. It's up to you. And the fourth and final piece. Blindly use keyword rankings and keyword rankings alone to measure your success. So many people rely on keyword rankings. I look at keyword rankings, but I do not rely on them. Just knowing that you rank number one for a keyword does not really tell you anything. Doesn't tell you how much traffic you're getting. Doesn't tell you how much engagement you're getting, how many conversions you're getting. Tells you nothing about your funnel and how things are working. All it says is, we rank number one. Cool. This is one of my favorite quotes from the philosopher dog from a while ago, and this resonates a lot with me. We rank number one, so we'll make ROI regardless of whether we track it or not. This is like genuine quotes that I used to hear all the time when I was working agency side with different customers that we were speaking to. And this is not a great approach to have. And there's one huge reason for this. This is the traditional view of keyword rankings, right? We rank number one. We get around about 30% of all the monthly search volume in terms of people that click on us. Number two is roughly 16%. Go down 10% to number three. And yeah, we can just model off that and make predictions, but actually the search results page has changed a huge amount and is changing constantly. So we've got like an image result section, we've got knowledge graph results section, You've got the featured snippet results section, you've got the space cat section, right? It's like everything has changed. It's all different. It's hard to rank there. But there are other ways. Don't worry. You should be looking at basic stuff, right? Conversion data to measure roi, you don't need to necessarily also be looking at this on a granular level. I was talking about this earlier around evaluating the performance of content in topic clusters, so grouping large volumes of content together that are topically relevant and analyzing them in clusters so that you can understand the true ROI of a campaign as opposed to an individual blog post, which you'll never really be able to do. Have a Look at Branded Searches how many times are people searching for your brand on an ongoing basis? You can check this in Google Search console for how much this is increasing over time to have a little look at what kind of the impact on the lift of the brand. A lot of the stuff you're having done keyword rankings should be supplementing a lot of these metrics. Organic traffic leads customers and then basically the way I use keyword rankings is to just flag up any big fluctuations that are happening. So just tie this all together. We want to go from Big Bill down to the little minnow of Blockbuster. Only focus on content, build content for the search engines alone, never test it's bullshit, and only use keyword rankings to measure success. It's all from me. Thanks very much. Any questions? Happy to answer.