Creating content that search engines love would be so easy if we had an exact formula to follow: x number of links, a certain length of the post, x number of images, and on and on. But the science of creating content that ranks well in search engines just isn’t that well laid out for us. In this month’s Google Hangout, John Doherty and Marcus Tober weigh in on the factors that can be seen to effect rankings, and what makes a content piece successful…or not.


Arnie: Hi everyone, I’m Arnie Kuenn with Vertical Measures and thanks for joining us for another edition of our monthly Google hang out. And today we’re going to talk about the content formula for ranking and search engines. And I think it’s kind of interesting, last April, we actually did an April Fool’s post about this. We just made up a bunch of metrics and said we had it all figured out, and it was reasonably well received even though some people believed it.

What’s happened since then is I think there’s been some really great research done and some excellent insights written about the content formula, if there is such a thing, and our two guests today, John and Marcus, have done a lot of that and I’m really happy to have them with us. And I think what I’ll do is start off and let John introduce himself and maybe talk just a little bit about some of the findings you have seen with content, and then we’ll go to Marcus and we’ll just go from there.

John: Thanks, Arnie, how is it going?

Arnie: Good thanks.

John: So, my name is John Doherty and I’m currently the head of Distilled New York City based in New York City, coming to you live from Brooklyn right now. I’m also the senior consultant there. So, I’ve been with the company for over two years at this point, working with–I currently lead a team, and I also have three clients of my own and oversee, and work with my team on their clients as well. So, I’ve done a lot of work around content. I’m obviously a blogger myself. I’m a content producer and Distilled is very heavy on content as well, putting out a lot of videos, lot of blog posts, that sort of thing. Do a lot of visualizations and all of that for our clients.

So I think the work that I’m most proud of when it comes to content was the, I published a blog post on, at the time, SEO Moz Blog a while ago. Actually, it was almost a year ago at this point I think. The title of it was, “What kind of content gets linked in 2012?” So from my perspective, there isn’t a specific formula that works, for content for everyone across the board, in order to rank in search engines because there are so many different websites, so many different kinds of companies, kinds of websites.

You can have a small 10-page website for your local chiropractor business or you can have a 50,000-page website for your hotel’s corporation. So, it’s going to depend on the website, which we can definitely dig into and talk about the different things that I have seen when it comes to ranking websites, when it comes to what kind of content do you need to have on the page, and talking about it both from a search engine perspective and a user perspective.

Arnie: Super, great. And Marcus, I’ve got to tell you before you introduce yourself. I read your full 70-page report that Search Metrics put out on my flight to Minneapolis last week. Found it really, really interesting, and I should also tell you that here at Vertical Measures, we were actually thinking about how could we actually conduct real research to discover some of the stuff you discovered and I’m so glad you did it instead of us. I’m sure it was quite a bear. So, go ahead and introduce yourself, and maybe you could tell us a little bit about that report.

Marcus: Absolutely, I have to. I’m Marcus Tober. I’m the CTO and the Founder of Search Metrics. I founded the company in 2007 and Search Metrics is a software as a service provider in mostly the search space, but we call us search and social analytics. We are providing the data mostly for enterprises and agencies.

And things that you mentioned, the ranking factor study, Arnie, because this is something that is really important for us, to provide deeper insights on how search engines are working. What do good ranking sites have in common and to show them some good sites. When it comes to the content and it’s really, really one of the most interesting topics. I think there is a content formula. Because the Internet is getting so much redundant information, Google has to do something.

The last number that I have in mind is that Google has about 30 trillion pages indexed last year and I guess when it comes to this year I think it is about 100 trillion and Google has to do something and content is still the key. And if you look back in the past, SEO was very technical. It was about having a keyword and a title tag. The keyword and the content you know. Many people, I think it’s five or 10 years ago people talked about keywords and all that stuff.

Google now tries to rank the content and put a relevance factor on the content, how relevant is the content. And with this relevance factor, and I’m pretty sure you have an in-depth formula, you have different parameters like is the user satisfied when it comes to social signals, when it comes to CTR, when it comes to search bounce rate, it comes to so many other parameters that are important.

And that’s why the ranking factor study that we made revealed that this year, compared to last year, content gets more importance. So, the number of words within a content, the correlation increased it also increased things like, we have a high correlation for the number of back links, but what was lower in correlation this year was very, very technical factors like keyword in the anchor text or what also increased was stop words in the anchor text, that means that you need more links that are more natural. And so, Google is looking for natural stuff and good content, and I’m happy to have this discussion with you, Arnie and John, because this is one of my favorite topics.

Arnie: I can imagine, yeah. And before I switched over to John, could you really quick describe what you mean by stop words?

Marcus: Sure. A stop word is something that’s pretty easy, findable like it’s here or there. A stop word is not like a noun or something like that, it’s something that we find is used in sentences. So, with a stop word you can find more natural language for example. And also the number of words within the anchor text increased, so this is more a sign that Google looks more for natural links.

Arnie: Right, got you. And John, so I read your Moz post back then and actually went back and refreshed, read it again, maybe two or three weeks ago when we were talking about putting this Hangout together. And I thought what was interesting was, and I read some other report, and I can’t remember who produced it, but I think what you were discovering was 2,000 kind of words and video and images and all that was directing the traffic and the links to those posts. And then if you look at the Search Metrics report, I think if I remember correctly, it was on the 700-some range like that. So do you feel like we’re getting any closer to–because that’s the question like, we’re a content marketing agency and I ask all the time, “How long should the content be?” So we feel like we’re getting anywhere near figuring out at least what Google might be looking for, if there is some magic formula?

John: I think Google is heading in that direction for sure. I mean think about in the last two weeks, they’ve launched an in-depth article section in the search results right and right now it’s the big publishers. Huffington Post and Washington Post and the Times and all of those guys, and you go to those articles and they are all long, they have a lot of images, they have other rich media. They might have slide shows, that sort of thing and obviously, Google actually released the formula, the factors that go into getting like to those in-depth articles. And I saw the guys at Verante actually released a WordPress plug-in to help you out with that, authorship and schema and that sort of thing.

I do think that Google is getting closer to that. In my opinion, in my mind, the longer the content is the better chance it has of actually being authoritative and actually being a well-researched, well-thought out, well-argued out, that sort of thing.

As far as, content on an e-commerce site or something like that, Matt Cutts has said that, anywhere from 300 words and up, like of original content on a page is enough. That’s the barrier, that’s the, that’s like the bottom line. If you do that it’s just like status quo. Honestly, in my mind, that’s easy marketing right, your range of words, write some like barebones marketing copy, it’s going to get there on the page, you might see some rankings improvement.

I did that on a site, a while ago I presented my findings at Search Fest six months ago in Portland, in fact it’s on SlideShare where we added content to a bunch of pages on a very authoritative website, between 500 and 800 words and we saw large improvements in rankings. It didn’t pop them onto the first page but, it might have taken it from the fourth page to the second page.

So I do think that longer content does rank better, but then the question comes in, why? And content with videos, with no images doesn’t rank better but again, the question is why? And then we kind of get into the psychological factors of that. I go to a site that is not well designed, and if I’m doing research online, I find a blog that’s ranking well, and I go and it’s just a wall of text, I’m gone right? I’m bouncing.

But if there are image is explaining, the idea is screenshots, maybe even an explainer video, that sort of thing, I’m going to stick on the page much longer and actually try to process what’s going on that page. I’m tired of going to these articles that are like, articles written by marketers that are like five point, super generic, five headings, and I’m like, “This does me no good. This isn’t actionable at all.” So I think, Google is actually moving in that way.

Also, with the Moz study remember that, I’ve found that there were articles with like 2000-plus words, that’s Moz-only articles. So Marcus’ study that he did which shows 700-plus words, honestly, I would go with that. But I think the lesson there is, if anybody was to say that no fewer than 500 words, I could go ahead and say like, “Write 700 words at least, just so you can rank better.” Then you have taken out the user, the audience, the relevance as Marcus was saying.

Arnie: Yup, yup. Okay Marcus, getting back to your research report, is there any kind of, like I kept trying, I actually got back into town and I had a bunch of pages marked and I recorded my thoughts on it, trying to come up with some summary to try to pass on to somebody here and maybe write a post about it, whatever but I’m sure you have thought about it 100 times more than me. What would you say is the summary of your research that people might be most interested in hearing about? And actually, maybe you should also tell the quantity of research that you did as well.

Marcus: So, in the beginning you mentioned okay, you appreciated that I did the study not you, but we worked on the study for three months, with some guys so it was really intensive. I think there is not only one conclusion but, let me have the conclusion for when it comes to the content. I think and you are completely right, John because if you have longer content, you can cover the topic better than just having short content. This is what the study also revealed that Google is looking, if you have longer content, you can cover the topic much better and this is something that satisfies the user right?

And now did you rank better, did you cover the topic, the user has a very long time on-site, low bounce rate. This is something that Google can also take into account to analyze is this the content that the user really wants? And for me one of the conclusions out of the study is that the very typical SEO-driven keyword optimization it’s gone, that’s in the past.

Because in the past, you took one keyword and you optimized the page or the URL for one keyword. Then you did another URL, another keyword, no demand media in the past, but they had a lot of phrases and for all these phrases, they created new content and that was redundant content. Google is now looking how the content belongs together, like you have an article for flu, you have the flu symptoms, you have it for flu whatsoever and if you have content that fits together for this topic, you are unique and that’s why you should rank.

That’s why I guess the longer articles with better estimations, rich media like images, videos, optimized on a single URL, that is huge and one of the conclusions out of the study. There are many other conclusions, like link builders, to social media guys, but I think this is the conclusion for the content.

Arnie: Yeah. Should we talk about the Google or the one pluses, correlation and causation that whole thing.

John: That’s quite the debate, wasn’t it?

Marcus: We can definitely repeat, but causation and correlation absolutely. I think, there are still many people outside that think okay, a ranking factor is a ranking factor and don’t understand what correlation means but, I think we really talked about the conclusions that you can prove. It can be proven. Like you said John, if you have longer articles and you would see an improvement in the ranking and that really makes sense. Not just blank text and longer many words but, we really see that covering the topic much better will help to improve the rankings.

Arnie: Yup.

Marcus: That’s good for content marketing, right?

Arnie: Yeah exactly. You know in one of my, I’m trying to instill in our company here with our clients, is a phrase, useful content. Jay Behr, a friend of mine wrote a book called “Utility,” and he is using the phrase useful content now. We all talk about epic content or engaging or awesome or whatever the words are, but I really, really think it does boil down to, what you are creating useful to the person that’s reading it and if you are starting to write fluff, or John said earlier, paragraph subheading, paragraphs subheading or a walk of text, I mean it’s not useful. But if it includes images and video and other helpful things, if it’s useful, it’s going to get shared most likely, not everything is meant to be shared, but I could see where that correlation certainly with Google Plus comes in. But it is really all about the quality of that hopefully useful content.

Marcus: But isn’t it how SEO worked in the past? You need a structure, okay and if the structure raise, okay I need 500 words at least or exactly 500 words, for headings and H one, that must be the keyword and the H two …that was SEO. Like with LinkedIn, how many links? What should be the structure for the links? Do we need the keyword in the anchor text? Then I have to do a lot of links. That was SEO in the past. That was the grain, just scaling everything.

John: Well, in my opinion it’s the lack of understanding the broader marketing ecosystem, right? People that are very SEO-focused, a lot of people in the industry still think like, plus ones drive rankings and drive traffic. It was like, they were just focused on this one avenue. What can I do to influence this one avenue versus, what can my avenue do to influence the other avenues?

So in Google’s mind, it totally makes sense. If the content is useful, then it’s going to be shared. You find out where it’s being shared and if it’s being shared, you know across the board and Bing actually released a very interesting study. It was a couple of years ago now, I represented in a couple of blog posts where basically they pulled a lot of data and they showed basically how they can read social signals and see where it’s coming from, and they can see where there are hotspots right? And they can actually pinpoint manipulative social signals. So it’s not just getting social signals, it’s where those social signals are coming from. This is two years ago and this is Bing. Google has had this for a long time, Google Has to have had this for a long time, if Bing has had it for over two years right?

Marcus: Hopefully Dwayne doesn’t you know see the Hangout.

Arnie: Well, hopefully he does.

John: Hey Dwayne.

Arnie: Oh darn, I’ve just lost my thought there with the whole Dwayne thing. The signals-wise you know, the co-relation and all that. I’ve actually been on a couple of panels and one with Dwayne where they were still saying, “No, it’s really not baked into the algorithm.” Because they still haven’t figured out how to filter out all the noise. I mean there’s just so many signals and so, I’m not convinced yet that it really is you know, certainly not a causation factor, in my opinion in any way.

Marcus: I think there’s a big difference from being to Google because Google owns the social graph data, so with one, in terms of services policy they can connect all the products together. So for me, I have my Gmail account for six or seven years now, so I have a history on Google. So, I’m writing mails, I’m using android, so they know me, what I’m doing on my phone. They know where I, that I’m really somebody who can’t cheat because I have a profile.

And if I share something, and this is what we proved many times, if I share something, this content will be indexed immediately. So, within minutes you can see it on the cache data. It could make sense that it’s not baked within the formula, that could make sense. So Google doesn’t maybe take into account for discovery, but it’s important for Google.

And if the SEO is now automatically creating millions of Google plus profiles, they have no history. This is how Google can filter out the noise and Bing can’t because Bing has a relationship to Facebook, but Facebook will never open the social [graph] for Bing.

Google has the data, that’s the big USP of Google that with the increase you know, we are using Hangout. So Google knows, these guys are doing something, make sense. And that’s why I think, Google has to take all the social graph data in the future because it’s less manipulative and better. It’s much better than.

Arnie: That’s well said, yup. Well guys, we’re going to have to head towards the wrap up. So what I would like to see if you could do, the question that we get asked is, what is the formula for content? How long should my content be and what it should include and I don’t know if each of you could just may be do your closing comment and see if you can concisely give to our audience what you think for most situations will be the ideal piece of content. So John, we’ll start with you and see if you can handle that.

John: Sure, I’ll give it a try. So in my mind, when I go to create content I think, my first question is, who is the user? Who is going to be reading this? Who do I want to share this? Right? And then I create a content based off of that. I also ask, “Where do I want it to be shared? Who do I want to be sharing it with?”

So, kind of the formula that I key to is–this is going to sound very esoteric, but it’s also going to be applicable is, is it thought leadership? Is it comprehensive? And is it interesting? So, a block of text is not interesting. I don’t want to read that. Is it comprehensive? Am I doing, is this going to be the thought leadership piece for this topic?

Basically, when I write something, I don’t want anyone to have to write something else on that topic. That is how you become the thought leader, that is how you get links, that’s how you establish yourself as the leader. So often, that’s going to mean, it’s going to mean more text, longer writing, it’s going to mean explaining your ideas out further. It may involve creating a video if you have that capability. It definitely going to involve images and screenshots and that sort of thing, I’m not talking and image slider along the bottom of like a Mashable article. I’m talking about screenshots that are explaining what you’re talking about.

So in my mind, that’s what drives links, that’s what drives traffic, that’s what drives social shares and then of course you know, having all the other leverage points in there as well. Like having social sharing buttons or encouraging people to share. I see people doing it on HackerNews all the time, like discussion going on over on how to HackerNews, when their article goes hot. So, encouraging people to also discuss it at other places because then, that spreads it across the Internet.

Also thinking about social seeding right, you thinking about who that user is, where that user is as well? So if I want to get something in front of Tim Farris, I’m going to go to people that, I’m going to write stuff that’s going to get in front of people that people on Twitter, on Google plus, that sort of thing. So, that’s kind of my formula for making content spread throughout the Internet.

Arnie: That’s super, great. All right Marcus.

Marcus: I can absolutely agree with John. So, but I have to add something to it.

Arnie: He tried to create that or say that so that no one could add to it.

Marcus: Yeah. Very comprehensive. So I think that especially in a very competitive market, where you have a lot of redundancy, you should think about first, what makes it different? If you have an e-commerce site you know, what makes it different from the others. If you just have the manufacturer description and you add some content, that’s not what the user needs. That’s why my content formula is that we have to think about what makes it different, so what’s the difference from the others? Why are you better than the others and do you cover the topic?

So, not just thinking about the keywords, please think about the topic and the topic consists of all the different keywords that are included in the topic. This is what I think makes sense for e-commerce or just content and everything. I can also recommend reading white paper from Search Metrics because we also have things, how the content must be structured, H1, H2, H3 and so forth. I think for me the content formula is all about the topic. If you can’t hit the topic, you will never rank.

Arnie: Yup, super. Well listen guys, I’ve really enjoyed this conversation and hopefully our viewers, the people who watched this from beginning to end I think should have gotten some awesome nuggets out of there, some great information. And I’ll thank John for giving up his lunch in New York and Marcus for giving up his dinner in Berlin. Thanks guys, and we’ll call it a wrap, and I want to thank everybody for joining us, and we’ll be back next month with another Hangout.

Marcus: No, we have to thank you. Thanks for having me, having us.

Arnie: Absolutely. Thanks Marcus.

John: Thanks Arnie, it’s been a pleasure.

Arnie: Thank you.

About the Presenters

John Doherty

John DohertyJohn Doherty is a senior consultant with Distilled in New York City. He’s worked on the Internet for years and focused on marketing since 2009. Previous to Distilled, John worked as an in-house SEO. Prior to that he ran a book publishing company from Switzerland.

Marcus Tober

Marcus ToberMarcus Tober brings pioneering experience in SEO to Searchmetrics. As chief technology officer, Marcus is responsible for advanced technology research, product development, and SEO consulting programs. While studying media design and computer science in Berlin, Marcus was already focused on search engine optimization, and he quickly became one of the leading SEO experts in Germany. Seeing the need for an integrated and professional SEO software solution led Marcus to found Searchmetrics GmbH jointly with Holtzbrinck eLab in 2007.