Ingredient Three: Links
You should now understand the meta page title, but how does it fit within Google's algorithm? Well, if I were an oversimplifying kind of person, I would tell you the following:
- When someone does a search on Google, the first thing Google does is locate every site whose meta page title contains the words in that search. It then looks at how many trustworthy links each web page in that result set has and puts the ones with the most trustworthy links on the first page and the ones with the least trustworthy links at the very end.
So once you've got the right page title, it's all about links.
It amazes me how much I still hear that SEO is about the things that are on your website. In fact, the main service most so-called SEO companies sell is one where they work on your website to cause it to attract search engine traffic. This boggles my mind because, other than the meta page title, what's on your website barely even matters to Google! It's all about links.
Links are to Google what grades and SAT scores are to a college admission officer. Does the admission officer care what you look like on your interview? Sure. But he wouldn't have invited you for the interview in the first place if you didn't have high enough grades and SAT scores. Similarly, does Google care about how user-friendly your website is, what's written on it, and how fast it loads? Absolutely. But they won't even look at it if you don't have the right quality and quantity of links.
That being said, if you have great links but maintain a poorly coded, slow website with nonsense copy written on it, you won't have much luck ranking on Google. Just as you wouldn't have much luck getting into college if you have perfect grades and SATs but show up late to your interview wearing sloppy clothes and making wisecracks at your admission officer.
The point I'm trying to make is this: Links matter far more than any other factor. And if you get the link component and the meta page title component right, you've got 85% of the job done right there.
In Chapter 1, "Trust: The Currency Of Google," I went over how to determine a high TrustRank link versus a low TrustRank link. And in the next chapter, I go over how to acquire links. So for now, let me just explain a bit more clearly what a link is and what makes it valuable so that you know exactly what you're hunting for.
For the purposes of this book, a link is anything on a website that, when clicked, brings you to another web page. The granddaddy of links, the one that was there in 1994 when the Internet was still something only nerds cared about, is the text link (or, as I like to call it, the "blue underlined link"). Figure 2.7 shows a snippet from my bio on evanbailyn.com that contains two text links.
Figure 2.7 An example of the way links are used inside text: they guide visitors to other web pages to supplement the information on the current page.
I had my web designer add these two text links into my bio so that people could see my web properties with a simple click. That is generally the purpose of links—to allow people to discover new web pages with a quick mouse click.
Text links are usually blue and underlined, but there are exceptions. Sometimes you'll catch webmasters being creative with their text links. Also the color of a text link usually changes when you click it; this is to remind you that you've already visited the site that the link references. A clicked-on link will typically be purple, but again, that is up to the web designer. No matter what color it is or what it looks like, if it takes you to another page, it's a link.
The other type of link is the image link. As the name implies, this kind of link is in the form of an image but functions exactly like any other link in that it brings you to another web page when you click it. We've all seen plenty of image links; every advertisement on every website is an image link.
You know by now that Google gives your website credit every time another website links to your website. However, certain types of links are more valuable than others. Text links are the most valuable types of links because Google can easily read the words in and around the text link to get a sense of what the link is referring to. An image link has the same magnitude of value as a text link, but without all the description. In other words, an image link is like an overall vote for a political candidate, one that says "I like this candidate." A text link is like a detailed vote for a political candidate, the equivalent of "I like this candidate because of his stance on health care and education." Text links are valuable because they tell Google what a site is about. This information allows Google to decide which specific keywords to rank that website for.
As I explained in Chapter 1, the reason links are so important is because they pass TrustRank. Text links pass highly detailed TrustRank. Image links pass general TrustRank. When asking other websites for links, you should request a text link, but if you absolutely can't get one (for instance, if text links don't fit into the site's aesthetic), you can settle for an image link. Ultimately, your site will rank higher with a mix of both, because Google's engineers have determined that most sites would attract both types of links in a world without SEO.
When requesting a text link from another site, you should usually ask for one that has your keywords inside the linked text. For instance, if I print business cards for a living, I want my link to appear in underlined text reading "business cards." These links could stand on their own or appear in the context of a sentence or paragraph. You do not want a link that simply has your website's address in the linked text. So to be clear...
The three ways you want your link to appear on another website:
Print your business cards quickly and inexpensively.
Nowadays there are many ways to send people your contact details. A simple email signature can suffice. You could share your Twitter username. And of course, we must not underestimate the old-fashioned convenience of business cards.
The two ways you do not want your link to appear on another website:
I would be finished writing this section if it were 2008 right now. For the vast majority of Google's existence, simple text links with keywords inside of them have been the greatest weapon in any search engine optimizer's arsenal. But alas, in mid-2009 Google slammed the SEO world with an algorithm update that punished sites that have too many same-keyword text links pointing to them. In other words, if I had been telling webmasters for years to only link to me like this— business cards —my site would have dropped way down in the rankings. I wondered why it had taken the arsenal of highly paid Stanford and MIT graduates at Google so long to realize that keyword-rich text links were overused by optimizers and were unnatural looking.
At the time the keyword-rich text link party was ending and hundreds of search engine optimizers were in their home offices crying, I remained relatively calm. I had outsmarted them before and was determined to do it again. How should I react to Google's rebellion against unnatural-looking link patterns? By adopting natural-looking link patterns, of course. I started linking to my clients in a smattering of different ways. Some links had the keyword a few words away in a paragraph, some had the keyword as only part of the text link, some had the keyword not mentioned at all, and some had the keyword inside the text link exactly the way I did before. Following is my recipe for the perfect "natural" linking pattern for your site. In other words, this is the formula for how other websites should link to your website under ideal circumstances. I have used "business cards" as the example keyword that I am aiming to optimize for.
- 20% links in text around your keyword but not on your keyword (for example, "When I need business cards, I contact this company.")
- 30% links that include your keyword but also include other words in a sentence (for example, "I ran out of my brand new business cards today.")
- 30% links that include only your keyword (for example, "I love my business cards .")
- 20% image links
In other words, I am saying that, if your site had 10 total inbound links, each link from a page on a different website, it would be ideal if
- Two of those pages contained your keyword but linked to your website from text near that keyword, not from the keyword itself.
- Three of those pages contained your keyword and linked to your website from your keyword in addition to the word or two that is next to it.
- Three of those pages contained your keyword and linked to your website directly from it.
- Two of those pages contained images and linked to your website from those images.
For the professional search engine optimizers among you, this recipe might seem like the cream of this entire book. Why not wait until nobody's looking, rip it out of the book, and stick it in your wallet, right? I must warn you that although this recipe works very well at the time of this printing, remember that Google is always shifting their algorithm. Your best shot to always fall ahead of the curve is to obtain as many natural-looking links as possible in a variety of formats.
Before moving on to the next ingredient, I must warn you about a certain type of link that passes no TrustRank at all and must be avoided at all costs. I call these types of links dummy links because they look just like real links but contain none of the substance that helps your website to rank. You should keep dummy links in mind because there is nothing worse than finding out that some of the links you have worked hard to earn are not actually passing any TrustRank. Here are the two types of dummy links:
- Redirect links— If a webmaster is trying to prevent TrustRank from being passed via the links on his site, he can have them coded in such a way that when someone clicks a link to an outside site, that person is first sent to a page on his own site before arriving at his final destination website. This is called a redirect. In other words, there is an intermediate page that the visitors hit before they get to the page they intended to visit. Usually the visitor is on the intermediate page just for a split second so that they never even notice they visited an intermediate page. This is a sneaky way of withholding the site's vote—that is, their TrustRank—from ever being cast. Google gives all the TrustRank they would have given to the outside site to the intermediate page, which is just some random page on the originating site. You can identify this kind of dummy link by hovering your mouse over any link on another site that seems to be going to your site and then looking at the bottom left corner of your screen where the URL that the link is pointing to is displayed. As long as it reads http://www.yoursite.com, you're good. If it reads something like http://www.othersite.com/redirect.php?url=www.yoursite.com, it is a dummy link.
No-follow links— The most sinister type of dummy link is the no-follow link, simply because you can't tell that it isn't passing TrustRank without looking at the HTML code of the web page. No-follow links are normal links that have been intentionally crippled via a short bit of code to prevent TrustRank from leaving a web page. They are an invention of Google, created to give webmasters a way to link to advertisers without "voting" for advertisers. For a couple of years, the no-follow link has been a subject of debate in the SEO world. Google holds that they are merely trying to figure out which links are editorial mentions and which are paid mentions. (Google only wants to give ranking credibility to the former.) Many webmasters think that they should not have to change the way they link to advertisers just to please an outside company, even one as important as Google. Naturally, advertisers don't like being labeled with a no-follow tag because that TrustRank is one of the chief things that they are paying for.
When another site links to your site as part of an arrangement you made, ask your webmaster to check to see whether it has a "no-follow tag" on it. If you catch a site linking to you with a no-follow tag, you have license to be very upset with them. The only good use of a no-follow tag, in my opinion, is in the comments of a blog because there are so many spammers who leave comments with links in them just to steal TrustRank from your site. Many sites, includes the New York Times website, place a no-follow tag on all comments. If you'd like to check a link yourself for a no-follow tag, find the View Source button on your web browser's menu (often you can access the code by right-clicking the page as well) and do a search for your site's URL. If you see rel=nofollow next to your link, you know that the webmaster is withholding TrustRank from being passed to your site.