FEATURE: 2009: The Year in Preview

by Bruce Clay, January 15, 2009

As 2009 dawns, change is a given. Actually, in the world of search, as any day dawns change is certain. The new year will bring substantial changes to our search engine optimisation world worth calling out, and this is our moment to dust off the crystal ball and expound on the yearly trends we expect to see.

2008 was a year of giant strides by the search engines in two key directions: customization of search results for the individual user based upon behavior and preference, and blended (or Universal) search results. Together, these have improved the overall search experience for users by making results more relevant and more engaging. And the SEO industry has so far taken these changes in stride.

I predict that in the first quarter of 2009 that there will be an increase in do-it-yourself PPC spending on advertising, thus overall advertising spend will go up, but this is not going to be spend placed through agencies. As companies have their Yellow Page ads expiring, a fair share of companies will opt for smaller ads and will switch spending to online. This shift in spend results in increased revenues for search engines, but little for our industry. Spending on SEO where results are not immediate will generally result in a delay in those projects, companies will be opting for the reduced risk of PPC out of a need to know that each dollar spent is on near-term revenues. With 28% of all businesses at risk of closure, only the companies with a war chest can afford the necessary investment to get ranked organically. As such, I am predicting that in the first quarter of 2009 agencies will see decreased services spend per client, tougher collections, and in general that this will be a tight 3 months.

I predict that in the second quarter those companies that have survived will recognize that online is a growing revenue source, and is quite frankly cheaper to operate than storefronts. We have businesses that are getting a larger share of their business from the Web and are closing storefront operations. The toy industry is a tiny example of that. Unfortunately this is a two-edged sword because as storefronts close there will be fewer wage earners, and they are the consumers online. Those businesses with vision and funds will increase their online spending to grab market share and position. The second quarter will see a general upturn in online projects and spending comparable to 2008.

June is a month in transition. Summer brings vacations, and along with this is the beginning of panic. The storefronts are diminished and your well-heeled competitors are getting ready for the 2009 holiday season. You snooze, you lose.

The latter half of 2009 will see chaos in the online marketing area. Everyone will need to spend online, meaning that PPC fees will be too high for those that have not tuned their conversion funnel to maintain. As such, the traditional last minute rush to the congested SEO project space will be much higher than normal. As companies consider that the major chains are online, that manufacturers are online and competing against their own retail channels, and that the lessons of 2008 where Amazon posted record sales is understood by the small proprietor, everyone will want to be ranked for their target market right then. Unfortunately this SEO-feeding-frenzy for the few top spots in an ever increasing competitive landscape will not be enough for some businesses. Search engine marketing emerges to be a competitive sport, and agencies with professional SEO athletes will win and amateurs will lose. Unfortunately, by the end of 2009 many companies will find themselves nowhere online, wishing they had started their race in March instead of July. The poor business decision of hiring cheap amateurs will cause businesses to fail. Those that understand that hiring the best is always a smart move, especially in 2009, will be winners.

During this period we will see the resurfacing of local search as a power. As geo-targeted businesses go online there will be an opportunity for competition at a level never before seen. The trained and competent semi-pro search engine optimisation athlete will emerge as a dominant player within a city or vertical. Not all businesses are national, and consumers will see an advantage to online purchases from local merchants, especially if there is an internet sales tax (and there will need to be if stores close).

By the end of 2009 we will see SEO as a mandatory defensive move, part of every business and absolutely a discussion at the Board of Directors level. Quality agencies will thrive and amateurs will revert to selling local projects. Online web sites will be a massive growth opportunity for Web designers, and online starts to be a measurement of economic health just as housing starts are today. Where online goes so goes the economy.

A massive growth in competent SEO training courses will be necessary to educate the marketing world to how Internet marketing works making 2010 the year of education.

That is how I see the industry (other than search engine technology) shifting.

From a technology standpoint I predict that search engines will greatly expand behavior-centric, intent-centric, and even geo-centric customization in 2009, and all SEO professionals will need new ways to get around it. Even the level of customization already implemented caused me to say in an interview at the PubCon and SES Chicago conference that ranking is dead as the main measurement of SEO success. Driven by behavior, personalization, and intent, two people can no longer run the same search and come up with the same results. As a result, ranking alone is no longer a reliable measurement of SEO success. You must use your Web site analytics to measure traffic, bounce rates and conversions to determine whether your search engine optimisation efforts are delivering as hoped. The days of "how high do I rank" being the measure of doing SEO right are essentially gone in 2009.

I predict that user behavior and related community tracking will be expanded (collectively "personalization") and increasingly applied Web-wide to all search results, paid and organic, regardless of whether the user is signed in to Google or not. By looking at the user's recent searches, visited sites, bookmarks, communities (common behavior groupings) and other online choices, the search engines adapt the results to what will be most relevant to that individual. So if a user searching for "java" had earlier searched for programming-related terms, or visited a programmer's forum, the results would be personalized to show sites related to Java the programming language, rather than coffee or travel. As for community, search engines observe patterns of search behavior across like groups of people, and then predict what results a similar searcher would be interested in. The similarity could be as small as all users that search for a single keyword, or it could combine multiple factors. Google understands what communities, loosely speaking, a searcher is part of. They're getting better at predicting what else the person is interested in.

I predict Google will implement Universal Search-driven algorithm changes most likely during the first quarter of 2009. Engagement Objects™ such as images, videos, Flash objects, audio, maps, news, books and blogs will weigh much more heavily than ever before as ranking factors. It will take a while to balance out the weighted averages in the algorithm, and for sites to imbed Engagement Objects into their sites, but it will clearly emerge as an influence in the first part of 2009. One advantage of this is that it would make it economically difficult for spammers to SEO 500 sites by adding Engagement Objects to all of them. Affiliate and lead generation sites will be collateral damage.

Many indicators point to this coming shift. Soon after launching Universal Search, Google claimed that its algorithm contained "over 200" ranking factors, compared to the previous 128-ish. Though Google keeps its algorithm largely a mystery, it's logical to conclude that the additional 70-ish ranking factors would in considerable part be related to Universal Search elements — things like the number and size of images on the page, whether there's an embedded video, and so on. So far, in part due to the need to change the spider to identify such factors, search engine optimisation and rankings have not changed to the extent you would expect from such a large expansion of the algorithm. 2009 will tell another story.

Google's ability to "read" and index rich media has definitely leapt forward this year. Googlebot is now able to spider non-moving text within Flash files, and has begun to apply voice recognition to soundtracks of videos and MP3 files, converting them to text that can be indexed. There's a lot of effort being expended to get the meaning out of Engagement Objects.

Another factor at play is intent-based search. The search engines, after resolving personalization factors, will determine if the intent of the query is research, or if the intent is shopping. This was easily seen in Yahoo Mindset before it was withdrawn, and clearly showed that the search engines will present different pages after the query based upon the search intent. Obviously the page and site content and presentation contribute to understanding the proper pages to be in the search results. Added to this is the automatic inclusion of geo-targeted bias. If you are a shopper, then local shopping-oriented sites become much more prominent in the results, whereas the researcher simply wants the best answer without localization concerns. We have seen this many times, and know it is currently in production. Results localized by searcher IP are already common.

In 2009 we will see these factors cause increasing volatility in search engine rankings in turn causing SEOs to wrestle with the new algorithm factors. With Web pages jumping up or down in the rankings, only to resume their original positions 20 minutes later, the game is going to be much more complex than most can handle. This volatility while Google is testing the new algorithm and gathering data by turning one variable at a time on and off will make the marketplace question whether SEO is a viable program compared to PPC. Those with vision will stick it out; those SEOs who understand this shift will win, and those slow to adapt will suffer. Yes, 2009 will be an interesting and fun year for the search engine marketing industry.

I predict that pay per click bids will go up sharply as Web site rankings destabilize. However, PPC ads will also be influenced by behavioral search, causing them to appear to a better-targeted audience. This improved targeting should reduce click fraud, since competitors will have a hard time seeing an ad if they don't personally fit the targeted profile. It will also become more difficult or expensive to buy keywords that don't naturally fit with a particular behavioral demographic. PPC advertisers will experience higher conversion rates, since searchers clicking their ads will be more likely to want what the site offers, and Google will benefit by increased profit.

We see and believe that this is the way 2009 will happen enough to include appropriate features into our new tools. 2009 will see the launch of our new SEOToolSet® which will address many of the changes to technology outlined above. Our new tools have many updates to the current presentation and content functionality, a more efficient database and an easy-to-navigate interface. Our first new tool is the Search Engine Optimisation/KSP tool, available free on our SEO methodology page.

A new SEMToolBar component will be available free. The toolbar enhances search results pages with keyword and page statistics for each result. As the user works on the Web, the PageRank, backlinks count, pages indexed, and other information displays in the toolbar for convenient reference. The new integrated tools will equip SEOs to deal with the changing search engine methodology and the increased globalization of Internet marketing. Over twenty different languages are supported, so you can run a search and view it in one language while your client simultaneously views the same report in another. For those selecting this option, the proxy search feature will soon overcome localization issues by letting you run a search as though you were in another city or country. Using the toolbar, you could be in California and run a proxy search as if you were in New York City or Australia or any other location you choose. You can also soon specify which search engine to query, picking from a growing list of up to 200 engines from around the world. So if you're working to optimise a UK Web site for users in Thailand, proxy search gives you a way to run those searches through the search engine of your choice, from anywhere. The new toolset works on the Microsoft Internet Explorer browser. We're working on a version for Firefox due out second quarter.

For permission to reprint or reuse any materials, please contact us. To learn more about our authors, please visit the Bruce Clay Authors page. Copyright 2009 Bruce Clay, Inc.

BACK TO BASICS: Evaluating Inbound Link Relevancy

By Fernando Chavez, January 15, 2009

How relevant are your site's inbound links?

Everyone knows that having external inbound links is a crucial component of SEO. With Google in particular, it is extremely difficult to rank well without high-quality inbound links to your Web site. But what is a high-quality inbound link?

In the past, people have generally considered a good link to be a link from a Web page with high PageRank (PR). Back before I started doing SEO, that may have been the case. However, Google got wise and has since enhanced their ranking algorithm to prevent inbound links from having the desired effect on rankings if they're coming from high PageRank pages with no keyword relevance. Although you will be getting more PageRank if you obtain links from high PR sites, the links will have little to no value if there is no keyword relevance outside of the anchor text to your site. Not surprisingly, purchased links now have much less influence on rankings than they had in the past.

So what has Google specifically done to combat paid links? In my opinion, they have drastically increased the importance of linking page relevance outside of the anchor text that links to your site. With that knowledge, you should be able to increase the SEO value of your existing links as well as know which sites you should try obtain links from.

PageRank = Link Importance

We have all heard about PageRank, and there are hundreds of sites out there that detail how it works on a more technical level. I’ve found a much simpler explanation that I find very interesting. On their Corporate Information Technology Overview page (http://www.google.com/corporate/tech.html), Google describes PageRank as follows:

• PageRank Technology: PageRank reflects our view of the importance of web pages by considering more than 500 million variables and 2 billion terms. Pages that we believe are important pages receive a higher PageRank and are more likely to appear at the top of the search results.
  
  PageRank also considers the importance of each page that casts a vote, as votes from some pages are considered to have greater value, thus giving the linked page greater value. We have always taken a pragmatic approach to help improve search quality and create useful products, and our technology uses the collective intelligence of the web to determine a page's importance.

In essence, Google is trying to determine your relative importance on the Web. And that's all the PageRank algorithm is — a method for valuing that importance using links alone.

Keyword Usage and Google Rankings

On the Corporate Information Technology Overview page referenced above, Google also states the following:

The software behind our search technology conducts a series of simultaneous calculations requiring only a fraction of a second. Traditional search engines rely heavily on how often a word appears on a web page. We use more than 200 signals, including our patented PageRank™ algorithm, to examine the entire link structure of the web and determine which pages are most important. We then conduct hypertext-matching analysis to determine which pages are relevant to the specific search being conducted. By combining overall importance and query-specific relevance, we're able to put the most relevant and reliable results first.

We have bolded the most important points. In a nutshell, what Google is saying is that they utilize their PageRank technology to determine which pages are the "most important" on the Web. They then combine that PageRank algorithm with hypertext-matching analysis to determine rankings for a particular keyword. Presumably, the sites that are determined to be the "most important" and the most keyword-relevant will rank highest.

The description of Google's Hypertext-Matching Analysis is not specific and doesn't cover anything outside of what is standard SEO practice. However, the following sentence illustrates the importance of links as it relates to keyword usage:

We also analyze the content of neighboring web pages to ensure the results returned are the most relevant to a user's query.

"Neighboring web pages" is another way of saying inbound linking pages. A page that links to your site is a neighbor in the way that Google is using the term. Although internal pages also qualify as neighboring pages, we are focusing on external pages because they have a much larger effect on rankings.

This sentence has remained virtually unchanged since February 2002 (still love the Archive.org Wayback Machine, by the way), which means that Google has evaluated off-page keyword usage factors since the beginning. But based on recent changes that we have seen in rankings for various clients, I suspect that Google has expanded upon or changed the way they evaluate neighboring pages. The end result is that you need much more than optimised anchor text on the pages that are linking to your site if you want to achieve optimal rankings.

What’s more important for a link — keyword relevancy or PageRank?

All other things being equal, the more PR a linking page has, the more value your landing page will receive from an inbound link. However, it is worth noting that the PR factor itself is not tied to a particular keyword. PageRank is an algorithm that is based solely on where a link originates and where a link goes. A link from a page with high PR is not necessarily going to improve your landing page's rankings for every keyword phrase that the page is optimised for. If the keyword phrase you are trying to optimise for is not used anywhere on the high PR page, then the PR you receive will not have the result you might expect. In fact, it is often more valuable to obtain a link from a lower PR page that actually uses the keywords you are optimising your page for.

Google has made this change over the last several years to prevent Web site owners from manipulating search results with purchased links from unrelated sites. Before these modifications, it was very easy for site owners, webmasters and SEOs to pay for links from completely unrelated, high PR sites with keywords in the anchor text and see huge improvements in Google rankings. Nowadays, I would argue that relevancy is far more important than PageRank. So much so that I would rather get a link from a PR2 or PR3 site than a PR6 or higher site if the lower PR page had good relevance for the exact keyword phrase I am trying to target.

If you want to receive significant benefit from any inbound link to one of your landing pages, the linking page should be very similar in theme to the landing page. Ideally, the page will directly use the keyword phrase you are targeting. While links from similarly themed pages are more valuable than links from pages that are completely off topic, links from pages that use the exact keyword phrase you are trying to optimise for are far more valuable. The next-best case in terms of a link's keyword relevance is a page that uses some words from the exact phrase for which you are optimising. For example, a link from a site that uses the one-word phrase "marketing" would be more helpful than a link from a site about "advertising" if I was trying to optimise for the phrase "search engine marketing".

Increasing Value of Current Links

The best way to evaluate your existing inbound links is to use two tools: Google Webmaster Tools and Yahoo! Site Explorer. You will have to verify ownership of the domain you are optimising in order to have access to Google’s information. Please note that we did not suggest using a Google link: command. This reports such a small percentage of the inbound links to a site that it is not worth your time. I have been asked many questions about this particular command, and I would strongly suggest not obsessing over it since there is much better information available to you.

In order to evaluate your external links, you will need to know which third-party pages are linking to the page that you are optimising. Google Webmaster Tools provides you with this information in the Links section, which you can find by clicking on "Links" in the left navigation once you've logged in and verified your site. Here is a screenshot of the main Links page from the Webmaster Tools for BruceClay.com:

After you reach this page you will want to click on "Pages with external links". You will then find yourself on a page similar to the one below:

At the bottom of this page, there's a link that says "Download all external links". Download the spreadsheet and you will have a list that shows the pages on your site with external links. The columns have the page being linked to, the linking page and when Google last found the link. You can then visit all of the pages that are currently linking to your site and evaluate their relevancy.

Obviously, if your site has thousands of inbound links, it will be difficult to evaluate the relevancy of each page individually. If you have a good programmer at your disposal, you may want to create a program that visits the pages in the downloaded spreadsheet and captures every page's Title tag. It will then be easy to search and sort the data by keywords that you are trying to optimise your site for.

Yahoo! Site Explorer provides similar information to Google, but includes more information by default about the links you find with the tool. However, the data is less exhaustive. Typically, you will only be able to see the first 1000 results from an Inlinks search.

To reach Yahoo! Site Explorer Inlinks results, you can either query Yahoo!’s Web search with link:http://www.domain.com/ for the domain you want to research, or put the URL you want to research into the "Explore URL" search box on http://siteexplorer.search.yahoo.com/.

You will want to click the Inlinks tab on the results page if you did not automatically arrive on that tab. In order to eliminate internal links, select the "Except from this domain" option in the Show Inlinks pull-down menu. Here are the Yahoo! Site Explorer results for the Bruce Clay® home page:

You will find that one of the advantages of using Yahoo! Site Explorer is that the Title tags of the linking pages appear in the search results (the Title tags also appear in the exported TSV results). Title tags are an effective way to determine the topic of the linking page, which should help you quickly eyeball the most relevant inbound links to your site. If you have the highlighting feature enabled in the Google Toolbar, then you can enter your keywords in the search box and highlight the Yahoo! Site Explorer page.

One other advantage of using Yahoo! Site Explorer is that they seem to rank their Inlinks results. The top-ranking results generally have higher PageRank, which means that pages are probably ranked based on their inbound links. Since Yahoo! has their own algorithm for measuring link popularity, it's fair to assume that they order their Site Explorer results based on that calculation. This means that you should tend to focus on sites that are higher in the results if the keyword relevancy is roughly equivalent.

Once you find the most relevant pages that are linking to your site, you should visit the pages and determine what the anchor text of the link is. If the anchor text is not ideal from a relevant page, then you should try to influence it if at all possible. You may find that the external links are the result of a relationship or affiliation with another site, and that a change of the anchor text is possible. You should absolutely take the time to contact all sites who might be willing to update their links.

It is worth noting that Yahoo! might include links with a rel="nofollow" attribute. For example, the link to the BruceClay.com home page from TopSEOs.com has a rel="nofollow" attribute (unfortunately for us), yet it is still included in Site Explorer results. This does not make sense to me as an SEO, but that is how Yahoo! has chosen to build the tool. When evaluating pages linking to your site, you should also make sure that the links do not have a rel="nofollow".

Potential Link Evaluation

While it is useful to evaluate your existing links, new links are more likely to have a significant impact on your site's rankings. There are several ways to develop a list of potential sites.

We suggest putting your top competitors into Yahoo! Site Explorer and scanning the results for relevant pages. You may find that some sites that are linking to your competitors would also be willing to link to you. Directories and industry-related blogs are the most common sites that you will end up finding because those sites are generally a little easier to obtain links from.

Once you have a list of sites to contact, you must have a way of evaluating their potential value. Assume you're trying to improve the rankings for the phrase "LA Lakers gear". The most valuable inbound links for that phrase would be from pages with the following characteristics:

• "LA Lakers gear" is used in the anchor text of the link.
• "LA Lakers gear" is used in the linking page’s Title tag.
• "LA Lakers gear" is used at least twice in regularly formatted text (i.e., not part of anchor text) on the linking page, ideally in paragraph content.

The more of these characteristics that the linking page has, the more valuable the link will be if you are trying to improve your site’s rankings for “whatever you’d like writing team”. In general, the ideal link for a particular keyword phrase would be from a page that contains the phrase in each of the page elements listed above and has the highest PageRank possible.

Remember that some relevance is better than none. Sites that share some of the words from your targeted phrase in the page elements listed above are better than sites that have no relevance. In fact, you may have to rely on semi-relevant sites such as these, because sites that use the exact phrase you are targeting are competitors more often than not.

Going through this evaluation early in the link development process will allow you to easily filter out non-relevant sites and thus prevent time wasting. Link building is a painstakingly laborious process, as many of us know. You do not want to invest several hours of time developing a link that has no direct effect on your site for any of your top keywords.

For permission to reprint or reuse any materials, please contact us. To learn more about our authors, please visit the Bruce Clay Authors page. Copyright 2009 Bruce Clay, Inc.