SEO stands for "search engine optimization". The term may sound like optimizing an actual search engine such as Google, however this professional service focuses on tweaking a website to do well – to appear among the top listings on search engine results pages (SERPs). SERPs are web pages returned by search engines like Google or Yahoo! after a user performs a search. These pages contain links to web sites and documents that the search engine deems relevant to the word or phrase. These words or phrases are also known as "keywords".
Using complex and proprietary algorithms, search engines consider hundreds of factors when gauging the relevance of a certain site or webpage. Search engines are constantly changing their proprietary algorithms – sometimes significantly – in an attempt to list only the most relevant results.
Major search engines vary their results by geographic region and language. For example, Google's German page will place more emphasis on German websites. Search engines use various pieces of information; among these is language, the physical location of a website's server, etc. Thus, the results from Google.de may differ from the results of Google.com.
SEO is "keyword centric". A major emphasis for professionals is to figure out what words or phrases a website's target audience is likely to use when searching for a site's content. Using this information, they strive to have the website appear within the first page of search results. SEO is strongly related to search engine marketing (SEM), which focuses on delivering advertisements that are relevant to an executed search. It is critical to the success of any SEO or website marketing campaign to have specific keywords in mind from the onset of the project.
A website's search engine "rank" refers to its position in the search engine's results. There are factors that influence the rank of a website on SERPs that website administrators can control; others they cannot. Controllable factors include page title tags, page content, the website's architecture, and the ease at which a web "spider" can examine a site. A web spider is a website discovery program deployed by search engines that scans the Internet looking for new pages and content changes on pages it has already discovered. There is little control over what competing websites can do to increase their search engine ranking; these actions may result in the down-ranking of other sites. Furthermore, website administrators sometimes can influence which other websites will link to their own site and how they will do so.
At the heart of successful SEO is determining what keyword or keywords to use when optimizing a website. The prevalence of a specific word or phrase on a webpage is critical to influencing SERP ranking. In most cases websites should be optimized for multiple keywords, and a simple way to do this is to optimize different sections or pages with different keywords. Of course, there are exceptions to this approach.
Keyword usage in <TITLE> tags is perhaps the most important factor that a website administrator can control when it comes to SEO. The search engines weigh this factor heavily. It is very important to place keywords in title tags!
Search engines only examine a certain amount of data, and when it comes to title tags it is our experience that spiders will probably only check the first 60 to 70 characters – including spaces. It is therefore best to keep title tags short and specific.
It is helpful to have unique title tags for each page on a site. This is also very useful when a site wants to do well in searches for multiple keywords.
There is some debate within the SEO community about whether it is better to put a website's name, organization name, or product name before or after a target keyword. Consider for example "Nike – Running Shoes" versus "Running Shoes – Nike." The important thing to consider is keeping the tags short and ensuring that the keyword for which the page is optimized is included in the web page's title.
If a keyword is in a web page's title tag, make sure that the page is actually about the keyword. Search engines will check the content of the page and expect it to match the <TITLE> tags.
When separating terms within a tag, avoiding symbols such as "&" and other code-like characters is advisable. The presence of such characters may confuse the search engine into thinking that it is not looking at code instead of a keyword.
The page title is what search engine presents people in results pages. That's another reason why it is important to make these tags descriptive. Does the tag accurately describe the page?
In addition to a <TITLE> tag, web pages can also contain a <META> description tag. While the description tag is not as heavily weighted as the <TITLE> tag, search engines sometimes use the content of this tag to provide a summary of that webpage on results pages. Other times they'll just pull some text from the webpage that seems relevant to the search term and provide that to users as a description on the result page. It is important to use the description tag to clearly explain the purpose of the page or site and invite a person to literally "click here."
Web site developers can also use <META> keyword tags within the code of a web page. However, these aren't considered to be very important anymore as overuse and abuse over the years has led search engines to simply ignore this tag for the most part.
Search engines obviously place a lot emphasis on a site's content when assessing its relevance to a particular topic. There are many ways that they evaluate the content. Many years ago, search engines figured if a word was mentioned a lot on a page or site, then that site must be relevant to that topic. However, it became far too common for people to stuff a site full of keywords; many times the wording was unnatural, and people hid text by making it color of the background, etc. Search engines realized that this did not help provide their users with truly relevant results and have since lessened the emphasis they place on keyword density. While it is important to use a target keyword on a page or site, the search engines are aiming for relevance, and density isn't the only measure. Therefore, it is very important to use target keyword(s) found in the target audience's natural vocabulary.
When looking at content, a small way that search engines determine the importance of a word is its characteristics on the web page. Is it linked to another page? Is it in bold? These are hints that search engines can use since it makes sense that a word that is somehow made to stand out has some importance.
A common content issue for websites is duplicate content (both on the same site and amongst many websites). There are many reasons why content may be duplicated throughout a site (like browser and printer friendly versions of the same page). Search engines feel it is important to direct their users to the most relevant and original version of the content. Further, it can confuse the engines when identical content appears on a site in more than one place. There are a few ways to deal with this situation. One is redirecting from one version of the page to the main version of the page using what site administrators refer to as a 301 redirect. The other is to use the robots.txt file, a simple text file housed in the root directory of a website's file structure that instructs web spiders to ignore a particular web page.
It is also important to note that search engine spiders cannot read all types of content. They cannot read the text within Flash; they cannot interpret the words in an audio file; and they are unable to determine the words used within a video. However, site designers can use tags (such as <ALT> tags with images) to help the web spiders know what a piece of media is about. The Flash, audio, or video can also be surrounded by regular HTML text content that is descriptive of the content. Using <ALT> tags is only really crucial when the majority of a website's content is unreadable to web spiders.
Structure is an important aspect of a website's search engine ranking. Search engines aim to find sites that are logically constructed since they want to provide their users with relevant sources of information, and it makes sense to expect a valuable source of information to have a logical structure.
Ideally, any visitor to a website should be able to access every page on the site within three clicks from the homepage. Search engine spiders don't usually burrow deep into a site and instead try to capture main sections. Obviously, not all sites are able to position pages that close to the homepage, but there are some strategies to help spiders see more of a large site. One solution is to create a site map containing links to all the pages on the site. Another strategy is intra-site linking; for instance, link from a major page to a deeper page. Keep in mind that it is far better to use a keyword that relates to the destination page than something generic like "click here." Search engine spiders can then use that link to find the deeper page.
The site's directory URL structure is also important. While placing keywords in URLs is not essential, it's helpful. One way to get keywords into the URL – even if www.[desiredkeyword].com is not available – is naming site directories using the keywords. If a section of the site is about a particular aspect of the overall site topic, use a keyword to name the section. Using a keyword to name a specific page (topic.html) is also helpful. Search engines do use directory and pages names when determining a site's relevance.
A great way to see this principle in action is to examine a search friendly URL like http://www.cnn.com/2008/TECH/ptech/06/02/blueray.sales.ap/index.html. Search friendly URLs contain information – including relevant keywords – that can help both search engines and people understand what a page is about by simply looking at them. See Figure 6 on the next page to see how the URL above is search friendly as one can see that the webpage is about technology and Blu-ray technology. It further also reveals when this news story was published.
Links help people navigate the web, and search engine spiders use them to determine how the Internet is tied together.
Search engines weigh the prevalence of links a particular website has pointing to it quite significantly. Websites with lots of links pointing to it are considered important sources of reference, and will therefore be ranked higher in search results pages than websites with fewer in-bound links.
Beyond using links to navigate a site and the Internet, links are valuable to search engines as they can help the engines understand what a site is about. The text in these links is a great clue to the search engines in determining what a site is about. They reckon if a there's a link with "whales" in the text points to a site that probably has something to do about whales (imagine how many sites are about "click here…"). Site owners can use this to their advantage by placing relevant keywords in the text of links within a site. Doing this helps search engines determine what various pages on the site are about. When doing intra-site linking, it is helpful to use words for the link that explain what the destination is about.
A site owner can also place a disclaimer for search engine spiders on a link. For instance, if a site feels it is important to link somewhere but doesn't want the link to count as a vote as to the quality of the destination, there is a "nofollow" syntax that can be used for this purpose. This practice is useful when linking to a competitor; this is a way to do without boosting their ranking. Using "nofollow" is useful for intra-site links to pages that are not important to a site's ranking. For instance, the site probably does not need to have its contact or privacy policy page highly ranked, and this will help preserve the intra-site linking value to pages that really should get ranked.
There is some conjecture that if a site links to a "bad" site (for instance, a spam site set up to lure people to click on text ads), that search engines will penalize a website for such a link. That's one reason to use "nofollow," but it is better to stay away from those "bad" sites altogether while at the same time avoiding linking to websites that don't appear useful to the website's target audience. The external websites that the links on a website point to can help reveal what a site is about. For example, if a site about animals has links to a zoo, that can help search engines determine that the linking site covers a similar topic as the zoo site.
Another interesting idea to consider when creating links on a website is using universal links ("http://www…") instead of relative links ("../topic/blah.html"). If universal links are used in text that is scraped by a spammer, the link will point back to the scraped site. So at the very least, the site gets another link (likely not a very valuable one), and anyone who comes across the link may in this way find your website.
It is very important to note that there are many strategies to collect links (which is commonly known as "link building") from other sites. Since this document focuses on what site administrators can control, this topic is not covered here. Please see SEOmoz's article "Beginner's Guide to Search Engine Optimization" for more information about this topic.
When optimizing a site, it is important to ultimately focus on the end user's needs and not the needs of web spiders. While the search engines play a large role in determining a site's relevance, it is important to remember that Google and Yahoo!'s main goal is to provide their users with quality results. So, a truly optimized site should aim to accomodate a site's target audience, not the search engines. One cannot ignore how a search engine spider views the world, but websites are not targeted toward web spiders. True optimization is trying to make a site as relevant and easy to navigate as possible to a specific target audience – that's what both humans and spiders are looking for.