Blog

Wired Article: How Google’s Algorithm Rules the Web

Wired is one of two magazines I subscribe to, and at something like $8.00 for a year’s subscription, probably one of the best values I can think of. The March 2010 issue is of particular interest to me for its article on Google’s algorithm. Thankfully, Wired is about as open-source as they come, so you can read the full article yourself here. But before this becomes any more of a love-fest/free advertisement, I should probably move on.

The article contains no groundbreaking revelations about the constitution of Google’s ultra-secret algorithm (darn!) but it does give the reader a new appreciation for just what an incredible achievement the algorithm represents. As the writer illustrates, type “mike siwek laywer mi” into any other search engine, and you will receive random or nonsensical results. But type it into Google, and the algorithm instantaneously parses out the terms to understand that “Mike Siwek” is a person’s name, “lawyer” is his profession, and “MI” represents the state of Michigan. And voila! You receive the listing for Michigan attorney Mike Siwek.* In short, Google is paramount because it understands the meaning of your search. And so far, it’s the only search engine that does.

Of course, if this sort of thing interests you at all, you would be better served to read the full article than my impressions of it. My only disappointment with the article (besides it not revealing the algorithm’s darkest secrets) is that it touches briefly on the growing debate about whether Google’s dominance is anti-competitive, but fails to explore the topic fully. Perhaps in a future article, Wired? I’ll look for it in the same issue as the Great Algorithm Reveal.

*Fascinatingly, a Google search for “mike siwek lawyer mi” DID produce the attorney’s business listing…right up until Wired published this article. Since that time, the Wired article itself (as well as numerous pages referencing the article) have overtaken Mike Siwek’s listing in the search results! This is not a surprising consequence, given the PageRank and popularity of the Wired website as compared to Siwek’s own page…but I bet it’s one that Wired did not anticipate when it published the article. Nothing like a little irony for a Monday afternoon!

What is “me nofollow”?

I have heard several people asking lately,

“What is me nofollow?”

And that tells me that they have been trying to analyze the value of outgoing links on Twitter. If that describes you, then good for you! You are thinking about SEO the right way by examining Twitter’s source code to see if it allows search engines spiders to follow outbound links.

You probably already know that the dreaded “nofollow” appearing in the HTML of all of Twitter’s outbound links means that no, search engines will not count those links. Now, when someone posts a link to their own website in the bio section of their Twitter page (the area to the upper right that includes name, location, etc.) Twitter adds something extra to the no follow tag. The HTML view of a link contains this:

class="url" rel="me nofollow" target="_blank"

See that “me” appearing right before the “nofollow”? It’s just Twitter’s way of saying that the website being linked to is owned by the same person who owns this Twitter page. If I own the Twitter page, then the external website I am linking to in that area is another expression of me.

Here’s the skinny: The presence of that “me” doesn’t change anything in terms of SEO. The nofollow tag is still doing its job, same as it ever did. As far as I know, Twitter is the only website currently using this “me nofollow” tag. Perhaps its use will spread, and perhaps it won’t…but unless the nofollow goes away, all you need to know is that it’s not going to help your SEO campaign.

Horn Tootin’

You’ll have to pardon us momentarily as we unabashedly toot our own horn.

Web Design CASE award winnerLast year, we were thrilled to learn that the U.Va. Magazine monthly enewsletter (designed and developed each month by yours truly) won the CASE Circle of Excellence Silver Award in Electronic Communications.

The 2010 national awards have yet to be held, but we just got news that our work with U.Va. Magazine garnered no less than THREE awards in the CASE southeast regional awards:

  • We received the Award of Excellence, the highest honor, in the category of Overall Web Site Design and Implementation for uvamagazine.org
  • We received a Special Merit Award in the category of World Wide Web Home Page Design and Implementation
  • We received another Award of Excellence in the category of Electronic Newsletters, Blogs, and Tabloids

…an exciting start to 2010 for Refresh!

How to Use Twitter for SEO

It is inevitable. Any social media application that gets big enough will eventually become the target of eager marketing executives, convinced that there must be a way to “leverage” it (read: use it to make money). The poster children for this phenomenon are Facebook and Twitter, both of which present unique challenges to the social media-minded marketer.

It’s Twitter that I want to discuss today. Judging by the queries I have received recently, the burning question in the marketing world is…

How can I use Twitter to boost my website’s SEO?

If you think about it, this question presupposes that there IS some way to use Twitter to boost SEO. That in itself is quite the leap of faith and, for that reason, I would like to rephrase the question as:

Can I use Twitter to boost my website’s SEO?

As in seemingly all things related to SEO, this question generates both a short answer and a long answer. If you find yourself thinking that these answers are frustratingly contradictory…well, then, you are really starting to get the hang of this SEO business!

The Short Answer: No

Directly speaking, you cannot use Twitter to boost your website’s SEO. This is a product of the fact that Twitter “no follows” all outgoing links. Allow me to explain.

Let’s say you run a business selling tennis balls, and you start a Twitter page as a marketing tool for your business. Not only are you diligent about tweeting every day, but you tweet really compelling stuff. For example, tidbits about the latest technological developments in tennis ball manufacturing. In a very short time, you have hundreds (or thousands!) of followers and your Twitter page builds an exponentially growing network of incoming links. In fact, your Twitter page’s PageRank climbs to 5, meaning it has major authenticity in the eyes of Google.

Because you are a savvy marketer, you made sure to put a link back to your main company website right there on your Twitter page. Being on the receiving end of a link from such a high PR page as your Twitter page, your main company website should be getting a big-time SEO boost…right?

Sadly, no. Twitter automatically places the “no follow” tag inside the html of all outgoing links, which effectively tells all search engine spiders: “You see this link here? Totally ignore it.”

As a result, no matter how authoritative or illustrious your Twitter page grows to be, it’s not going to pass on any of its credibility to your company website. The value of your incoming links is the most important factor in your SEO success. Because Twitter is designed to prevent you from receiving any value from its links, I am here to say that you cannot use Twitter to boost your website’s SEO. Directly speaking.

The Long Answer: Maybe

The value of your incoming links is the most important factor in your SEO success. That said, every search engine algorithm comprises many additional factors. Placement of keywords on your pages, ease of site navigation, quality of your code, internal linking structure, domain age…the list goes on and on. While we don’t know the precise makeup of any search engine algorithm, we do have a good handle on most of the things that matter. Some of these components concern your site’s traffic.

How many visitors your site receives, how many pages your average visitor views, how long your average visitor spends on your site…these are all metrics that likely play a part in your site’s SEO success (and by the way, if you aren’t doing so already, you can track these metrics and more quite easily via Google Analytics). So, while Twitter can not directly boost your SEO, if your Twitter content were compelling enough to send visitors to your company website, and these visitors liked your site enough to spend time there (or even buy something!) then you would be very likely to experience an SEO boost as a result.

For SEO purposes, are any of these traffic-related metrics as individually important as the value of your incoming links? No. But taken together, they are a good indicator of how satisfying your visitors find your site’s content, and if these metrics are strong, the search engines will take notice. Of course, there are many variables here. Is your company site actually good? Do you have more reliable means of attracting high-quality visitors (those that stick around and/or convert) than Twitter? Are the visitors you could possibly receive from Twitter high-quality or not?

All of these things must be viewed in perspective but, under the right circumstances, it is possible that Twitter could indirectly boost your company website’s SEO if it sent high-quality visitors to your main page who would not have found your site otherwise.

Something else to keep in mind (and perhaps this is the more likely success scenario for you) is this: If you were able to use Twitter to make industry contacts, who then passed potential customers in your direction, you would at least be successfully “leveraging” Twitter to generate revenue. If those industry contacts could be convinced to link to you from their websites, you would even get your SEO boost as well.

In conclusion, short anwer: no. Long answer: maybe. Welcome to the wacky, wonderful world of SEO!

What is PageRank?

As promised in my earlier entry about the importance of incoming links, I want to go into a little more detail about PageRank and what it means for your site’s ranking in the search engine results, specifically Google’s.

PageRank is a means of assigning a value (in terms of perceived authority) to any page on the Web. This method is exclusive to the Google algorithm, and if you’re interested in the history of that, you can learn a lot more on the Wikipedia entry for PageRank. But if you’re ready to jump straight to what it means for us, then read on.

In essence, Google assigns each page on the Internet as having a PageRank (PR) value between 0 and 10. This number is determined by the value of the page’s incoming links. Your web page’s PR matters because Google is wired to give extra weight to pages with high PR when it comes to displaying search results to its users. If your web page and a competitor’s web page have the exact same information but his PR is 3 while yours is 0, you can bet that he is going to rank higher in the Google search results every time.

Take a look at some sample PageRanks (as of January 4, 2010):

target.com: PR 7
digg.com: PR 8
amazon.com: PR 9
google.com: PR 10 (were your surprised?)

To understand how PageRank works, let’s keep things simple at first and pretend that every link on the Internet has the exact same value (that is, a link from your little brother’s blog is worth the same as a link from time.com). In this world, you create a web page. On its first day of existence, no one knows about your page and therefore, no one is linking to it. Fair enough, your PR is 0.

But after a month, there are 10 pages out there in cyberspace linking to your page. 10 incoming links of equal value = 1 PR for your page. Several months later, let’s say you have accumulated 100 incoming links. 100 incoming links of equal value = 2 PR for your page. It’s important to note that in this world where all links are created equal, each step up in PR requires a ten-fold jump in the number of your incoming links. That is, a PR of 3 would require 1,000 incoming links, a PR of 4 would require 10,000, and so on. One quickly realizes how exponentially difficult it becomes to climb each rung of the PR ladder.

However, we don’t live in that world. We live in Google’s world, where all links are NOT created equal. A link from the homepage of time.com (PR 8) is NOT the same as a link from your little brother’s blog (PR 0). That’s because Google lets PR “bleed” across websites via outgoing links. In short, when outside web pages link to you, then YOUR PR goes up as a function of THEIR PR. A link from a high PR site like time.com can very well carry the weight of hundreds of links from blogs like your little brother’s.

In order to prevent the exploitation of PageRank by those who would falsely boost their own websites, the exact algorithm behind it is kept a close-guarded secret. We understand that other elements probably factor into PageRank (the placement of your incoming links on the linking pages, for example) but for the most part, simply think of PageRank as a function of the quality of your incoming links.

The biggest take-away from all of this is the understanding that the quality of your incoming links matters more than the quantity. When pursuing incoming links for your site, remember that obtaining just a handful of incoming links from content-relevant, high PR web pages can easily outweigh the value of gobs and gobs of links from PR-deficient web pages. These high-quality links will be more difficult to come by, but they will prove a much better long-term investment of your time.

Is the HTML Address Tag Useful for SEO?

I was inspired to write this post because it’s a question I just recently asked myself. Having just started this website, I have been working a lot on the back-end, getting all of my SEO ducks in a row. Because I want our new website to rank well for local searches by potential clients, that means making sure our company address is present on every page of our site. My first thought was, “No problem! Stick it in the footer and call it a day!” But then I got to thinking…

Do the search engines care what format my business address is in? One would think that the search engines should be far more than capable of parsing out an address from a website, as long as it is listed in a half-way coherent fashion. Combine that effort with submissions to Yahoo Local and the Google Local Business Center, and it SEEMS like all your bases would be covered.

But then I thought about that often-forgotten HTML tag: the HTML address tag. The tag is simple enough, and designed to do exactly what you would think: identify a physical, geographic location. It looks something like this:

John Q. Search
123 Sesame St
Richmond, VA 23221
(555) 123-4444

The address tag is sort of a vestige from an earlier time, back in those bad old days before we had heard of Google (let alone Bing!) and we all thought animated GIFs on GeoCities pages were just about the best thing in the world. Shudder.

But is it useful today? As in all things SEO related, the definitive answer is buried within secret search engine algorithms, held close to the vest by cryptic search engineers. However, I did some anecdotal evidence gathering of my own, and came to a conclusion fairly quickly. I’ll cut to the chase:

In short, there is no evidence that the HTML Address Tag is useful for SEO. The few sites that I can find using it are showing no apparent advantage for doing so, and the tag is noticeably absent from the vast majority of sites with enviable PageRank.

Furthermore, there IS evidence that the presence of the tag slows down your site’s loading time. I’m not just picking on the address tag here: any HTML tags are going to slow down your site’s load time. But unnecessary tags are going to slow it down unnecessarily, and that’s the point. Why should we care?

What a timely question! This month Google announced that it is adding page load speed into its algorithm for page ranking consideration. Which means that, along with about 200 other factors or so, the speed with which your site loads will now impact your placement in the Google search results.

Conclusion: With no evidence for the HTML address tag doing anything for your SEO efforts, and clear evidence that its presence will negatively impact your Google ranking (even if only slightly), I must advise you to toss this ancient tag out the window. The dog had its day, but is now little more than a leech.

How important are incoming links?

Alternatively, this post could have been titled, “Do I REALLY need to conduct a link building campaign?”

I think everyone asks this question because they are hoping to find the answer they WANT to hear. Something like, “A link building campaign? Nahhh, you don’t need one of those!” Unfortunately, it’s my job to bear the bad news (which you probably know deep down already!)

Here’s the short answer: Incoming links are VERY important, and YES, you need to conduct a link building campaign. Strategic on-page optimization is the first step of an SEO campaign, but incoming links are necessary for genuine success.

All search engines are not created equal, and it’s really Google that values incoming links more than anyone else. So if you are only interested in your site ranking well in Yahoo, then perhaps you can discount my advice and do just fine without rabidly seeking out links. But how many of us does this really describe? For better or for worse, Google is the 800 pound gorilla of search, so if you want SEO success, you gotta play the game Google’s way.

Once you’re on board with this cold, cold truth, there are some things you should know about acquiring those precious incoming links…

All Links Are Not Created Equal

If you only take away one lesson from this post, let it be this one (which I promise to describe in deeper detail in a future post all about PageRank). Google evaluates every web page in terms of its perceived “authority” on the Internet, and assigns each page a PageRank, from 0 (the lowest) to 10 (the highest). The higher your PR, the more likely that Google is going to rank you highly for the keywords associated with your site. This PR score is determined entirely by the quality (and quantity) of the page’s incoming links. When a page with high PR links to your page, some of its PR “bleeds” over onto your page, giving a boost to your own PR score. Score!

Since your own PR is determined by the PR of the sites linking to you, the quality of your incoming links is far more important than the quantity. While the exact calculations are buried inside Google’s secret algorithm, understand than ONE incoming link from a site with a PR of 1 may be worth more than TEN incoming links from sites with 0 PR. It is, of course, more difficult to procure incoming links from high PR sites, but when time is money, I want you to invest yours wisely.

Also note that the type of site linking to you matters as well. That is, all things being equal, an incoming link from a .gov or .org site is worth more than a link from a .com site. Again, I will go into more detail in a future blog post.

One-way Links vs. Reciprocal Link Exchanges

Exchanging links with your business partners is a good way to scratch each other’s backs, especially if you each have decent PR to pass on to each other. However, the folks who build the search engines understand these reciprocal link exchanges as well as you and me, and that they are more of a business transaction than a true marker of website quality. For that reason, search engines are built to recognize such link exchanges, and devalue them. Not totally devalue them, mind you, just…not hold them in the same regard as a one-way incoming link.

You see, when someone links to you without you linking to them, it’s an indicator that your website must really have something worth reading. After all, there is nothing in it for that website owner. Because of their perceived purity, one-way links are more valuable for your SEO campaign than reciprocal link exchanges.

Anchor Text

You already know what anchor text is, even if you don’t know the name for it. That changes now. When I post a link, say to Refresh Web Design & Internet Marketing, the anchor text is the text (“Refresh Web Design & Internet Marketing”) that is serving as the link. Using anchor text like this is much prettier than seeing “html://refreshperspective.com” stuck right in the middle of a paragraph. It’s usually colored or underlined text, so you can easily recognize it as a link.

When it comes to your incoming links, the anchor text matters. A lot. The search engines read this anchor text, looking for keywords to associate with your site. Let’s say you sell apples through your business named “John’s Market.” If people are linking to you with anchor text that simply says, “John’s Market”, then you’re missing out on an opportunity.

If, however, their anchor text read, “The Best Apples in California” and that text was a direct link to your website, then the search engines will associate that key phrase with your site. Now, the next time someone uses Google searching for the best apples in California, Google is much more likely to put your website at the top of the search results. Of course, things like PR and on-page optimization are still in play as much as ever, but the anchor text of your incoming links is a real factor in your SEO success.

Since incoming links are, by their very nature, not on your own site, you do not often have control over what they say. But in those instances where you can control this text (reciprocal link exchanges, business directory listings) be sure to get your most valuable keywords into the anchor text.

The HTML Matters: Follow vs. Nofollow

When working to secure an incoming link from a particular web page, one of the first things you should check is the page’s HTML code to find out whether they are designating their outgoing links as “follow” or “nofollow”. Let’s take an example of a blog. Perhaps you want to post in the comment section of the blog, and leave a link back to your own web page. As long as you aren’t being Sir Spam-a-Lot, this can be a great strategy. But if the owner of that blog has set blog comment links to “nofollow”, you are wasting your time.

By default, search engine spiders automatically follow any and all links they find on web pages, and add the information gained to their indices. In order for Google to give your PR credit for an incoming link, it needs to actually follow the link going to your site. HOWEVER, if the blog owner has set the link to “nofollow”, they are blocking the search engine spiders from following the link, and therefore preventing any PR from being passed through!

View the page’s source code and check any outgoing links already in the comments section. If any of those links include the text “rel=nofollow”, then give it up. The search engine spiders can’t follow the link, and as a result, your site will get no credit for it.

To be continued…

I hope this address the question of the importance of incoming links, and gives you some idea about how this whole link business works. I’m looking forward to going into more detail about the value of different links in a future blog post.

The Google Ad During Super Bowl XLIV: How Hell Froze Over

In the days leading up to February 7, 2010, there was much online speculation about a rumor that Google would air an advertisement during Super Bowl XLIV. Certain people claiming to be “in the know” were certain that it would happen, but many found it hard to believe. That skepticism had a lot to do with Google’s long-standing attitude toward brand advertising, which CEO Eric Schmidt denounced as “the last bastion of unaccountable spending in corporate America” back in 2006.

But the rumors were confirmed the day before the Super Bowl, when Schmidt himself posted on Twitter:

alt text

Of course, as anyone who watched the Super Bowl or visits Digg or has heard of YouTube knows by now, when Google finally decided to advertise, they came out of the gates strong. It’s my favorite Super Bowl XLIV ad by a longshot, and I don’t think that’s just because I am in SEO.

https://youtube.com/watch?v=nnsSUqgkDwU%26hl%3Den_US%26fs%3D1%26

So it’s true: hell has indeed frozen over. But what could so fundamentally change the direction of a company that always seems to know exactly where it’s going?

The same reason any company feels the need to invest in brand advertising, I would imagine: fear of the competition. The timing of Google’s commercial seems to confirm this assumption. Just two months ago, on Dec. 4, 2009, Microsoft and Yahoo! finalized their merger agreement. And while Google’s 65% share of the U.S. search market still dominates the landscape, the combined force that is Bing+Yahoo effectively holds 30%, representing (by far) the largest competitor Google has faced in years.

So Google’s surprising decision to enter the arena of brand advertising is perhaps not so surprising. But by acknowledging Bing as a competitor worthy of a 180-degree policy change, Google not only reminds the SEO industry of the importance of the “other” search engine; it looks, for the first time in nearly a decade, a little more human as well.