Search Engine Optimization -- SEO

Website Promotion

Search Engine Optimization consists of four components:

Web Page Coding
Web pages must contain the phrases used in searches and these phrases must be appropriately placed on the Web page.
Search Engine Indexing and Directory Registration
Web pages must be found by the search engines, crawled and the indexed data must be stored in the database the search engine uses for searches. Relevant data must be subitted to worldwide directories and possibly local directories.
Search Engine Vigilance
Search phrases must be regularly tested to determine if they produce the required ranking on the major search engines.
Housekeeping
If a website isn't kept up-to-date and always online, a search engine may crawl the site, find errors, or, the site may not be online when it attempts the crawl. This can result in lower ranking and even deletion from the search engine's database.

Directories
Industry directories can be a good source for links and more importantly produce links that Google considers "relevant".

Local Directories
The importance of local directories is increasing for businesses that primarily serve a limited number of cities or regions.

Spamming
Search engines use the word "spamming" to describe gimmicks that attempt to improve ranking on search engines. Search engine spamming used to work, but not any more.

Web Page Coding

The algorithms (computer programs for finding Web pages in a search engine database) change and no one other than the programmers and system design people who work for the search engine company know precisely what these algorithms are. And, they aren't sharing their secrets.

However, there are companies that systematically test the effects that Web page coding practices have on search engine ranking and some of this information is shared and can be researched via the Web. EPCS does some independent analysis, but primarily relies on structured and statistically reliable research done by others.

Based on the published research and what we've observed on the Web pages we maintain, we make the following assumptions when coding Web pages.

The Web pages must contain the most likely words that may be used in searches. Search engines can't see or understand graphics, only text. If the words people are using to find your site don't exist on your site, we can guarantee that the page won't be found with these search words. Text challenged websites, sites that contain virtually all graphics (e.g. many Flash sites), are difficult to find via search engines.
The Web pages must contain enough text. Research shows that search engines raise ranking based on word count.
The most important search phrases must repeat, but not so often that the search engine considers the repeated phrases to be an attempt to influence ranking. How much is too much? That's a tough question. The problem is best solved by simply using normal essay like sentences and paragraphs to describe product and services.
In addition to text, we've also observed that placement of a search phrase inside the following tags will improve ranking for the phrase: <title>, <meta description>, <meta keywords>, header e.g. <h1> tags, bold <b> or <emphasis> tags and graphic alternative text tags <img alt>. But, don't overdue it; see Spamming below.
We've observed that recently modified sites generally rank higher on searches than our more static websites.

Search Engine Crawling and Indexing

A search engine will work like a spider as it crawls the Web. It will start with a Web page, extract data from the page for its database, and proceed to other pages referenced (linked) on the page. With the exception of directory services, which require manual submission of data, a search engine must crawl a website to obtain data for its database. It's the database that is subsequently used for searches, not the actual pages on the website.

Once initially crawled, the search engine will periodically go back to the site and re-crawl it.

How does a search engine initially find a website? There are two ways.

Search Engine Registration
You can request that a search engine crawl a page (there may be a fee).
On Its Own Via a Link
The search engine can find the site via a link from another page it is crawling.

Search Engine Registration

Some important search engines do not require a fee to register (suggest, submit) a website. An important search engine that does not require a fee is Google. Google has huge market share (see EPCS Server-Share side bar ). All websites should be registered with Google, MSN and Yahoo. The free registration process will eventually trigger a crawl of the website; however, it can take as long as eight weeks for the crawl to actually take place and our logs show that the crawler may not actually look at all pages on the website.

On Its Own Via a Link

Being initially found by a search engine via a link from another website is a method of being found. It's free and the data won't be deleted out of the search engine database because a renewal registration fee isn't paid on time. That's the good news. The bad news is that it may take a search engine a long time to find a site, or worse, it may never find the site.

Some search engines offer both a free submission and a paid submission option. The search engines that offer a free submission will take up to 8 weeks to crawl a site.

We encourage customers to request links to their website from other websites, particularly websites maintained by associations the business or organization may belong to, as well as supplier and and sometimes even customer sites. There are three benefits to doing this:

It increases the probability of being crawled by a search engine;
It increases ranking in Google and Yahoo; and,
Some customers will get to your site via a link on these other sites.

Directories can impact ranking on the major search engines, particularly Google. Directory registration remains an important part of a Search Engine Optimization strategy. See Directories below for more information.

Search Engine Vigilance

EPCS offers a service called Search Engine Vigilance. We systematically look for the website via the major search engines using an agreed upon search phrase. If we don't find the site on the first page of search results, we attempt to remedy the problem. There is a $10 monthly fee for each phrase used in weekly searches.

If a site isn't found, we'll submit (or re-submit) free registrations and paid registrations authorized in writing by the customer. We'll re-submit weekly until we can determine that the website is included in the search engine's primary database.
If the annual registration fee hasn't been paid, we'll request authorization from the customer and pay the fee.
We examine the website to determine if internal coding enhancements may improve ranking and either make the changes, or, if the changes significantly impact the look of the site, we'll request permission from the customer to make the changes.
If we believe a paid registration will improve ranking, we'll make recommendations to the customer regarding paid registrations.

Weekly Report

A weekly report is e-mailed to customers that shows the results of the vigilance process and what was done to resolve ranking and inclusion problems.

Referrer Logs

Referrer logs are maintained for all websites. These logs track where links to a Web page originate. The monthly and annual reports produced from these logs are available for all websites. Access to the reports requires a login. The reports include information regarding what search engines were used to link to your website and what search phrases were used for the link. The reports also show hit statistics and other data.

Housekeeping

Broken Links

Links on your site to other websites don't improve search engine ranking, but broken links can degrade ranking. It's important to quickly find and fix broken links on websites. EPCS runs a weekly process that automatically examines websites for broken links. We'll attempt to figure out the correct link, fix the link if we can, and, ask the customer what to do next if we can't fix the link.

Server Up-Time

Search engines will sometimes drop sites from their databases or change the ranking for a site. Why? We usually don't know. However, we can control one factor related to dropping website data: not being available for a crawl.

Server up time is a significant factor in SEO. If a site can't be found by a search engine when it attempts to crawl the site, the search engine may simply delete the site from it's database. To mitigate this problem, we use a Web server that is optimized for up-time and Internet connectivity. The Web server, located in a secure data center near Chicago, uses automatic processes to backup and recover data. It has multiple high speed Internet tier 1 backbone connections and will automatically switch to an available connection if it loses a connection. It's backed up with both battery and natural gas generator power. It's regularly updated with the latest virus and intrusion defenses and has operators on site 24 hours a day, 7 days a week. The result of all this redundancy and precaution is near zero down time for websites.

People Errors

The EPCS Search Engine Vigilance process includes a step where our searcher must actually view each customer's Web page each week. This step is included to catch mistakes that may have been made by people. For example, some of our customers directly maintain their own domain name registrations and/or DNS updates. Our weekly search engine vigilance process has uncovered errors and omissions in this area that result in the website not being accessible via the domain name for the website. It's also uncovered some mistakes made by programmers and system administrators at EPCS (a rare event, but it's happened). Good housekeeping mitigates these problems by having a process in place to systematically find them.

Directories

In the beginning, there was but one search engine, Yahoo, and it was a directory. The importance of the Yahoo Directory was immense, since at one point they had the entire search market. Then the first crawler, AltaVista, came online and the importance of directories started to wane. Yahoo currently maintains a directory service, but nearly all Yahoo searches are done via the Yahoo database populated with crawled data, not directory data.

Our referrer logs (see EPCS Server-Share side-bar) show that relatively few searchers use directories. But, we also know that even though directories are not directly used much by people, search engines use these directories as a supplement for their primary databases. An important example of this is Google's use of Open Directory

A major component in Google ranking is the number of other websites that link to your website. If the link is an authoritative site for your industry then Google considers the link to be more relevant than other links to your site. Relevant links to your site will increase your Google ranking more than non-relevant links. Directories can be an excellent source of "relevant" links and indirectly improve your Google ranking.

Inclusion in directories is still important, not because people use them, but because inclusion in directories can improve ranking in Google and other heavily used search engines.

Yahoo Directory Fee

Yahoo charges $299 annually to be evaluated for inclusion in their directory. Yahoo doesn't guarantee inclusion for this price, but we've never submitted a site that wasn't included.

Is inclusion in the Yahoo Directory worth the price?

Maybe!

New sites registered with the Yahoo Directory appear on the Yahoo New Additions pages; and, we've long suspected that search engines that crawl the Web use the Yahoo New Additions pages to find new quality sites to crawl. We've read claims that inclusion in the Yahoo Directory may improve ranking on Yahoo searches. Google gives us a big hint regarding Yahoo Directory's influence on Google ranking: Google Webmaster Guidelines suggest, "Submit your site to relevant directories such as the Open Directory Project and Yahoo!, as well as to other industry-specific expert sites."

Local Directories

The importance of local directories is increasing for businesses that primarily serve a limited number of cities or regions. There are several important local directories:

EPCS offers a service that will register and maintain relevant information on these directories for a $300 annual fee.

If your local chamber of commerce has an online member directory that provides a link to your site, then inclusion in this directory can be useful. Google and other search engines may consider the link to be of high quality and the chamber link may improve ranking. The downside of chamber of commerce links is that they usually don't script your e-mail address and spammers may harvest your e-mail address from the chamber site. For more on this see the EPCS Avoiding Spam page.

Industry Directories

There are many specialized directories on the net. If the specialized directory relates directly to your industry, then the particular directory is worth a look. One of the most well known directories for manufactures is Thomas Register. Industry related associations often provide specialized directories that are worth the fee for membership and inclusion. Links from these types of sites may produce click-throughs but may be more valuable as highly relevant links that can raise Google and Yahoo ranking.

Industry directories are sometimes labeled "vortals" (Vertical Industry Portals). Examples of copper tube manufacturing vortals include:

Spam Directories

Some directories provide only one thing, a link to your website, and promote the service as a means to improve ranking on Google. It's our opinion that these types of links actually reduce ranking on Google, since the directory is viewed as a spam site. Google calls these sites, "bad neighborhoods".

E-mail Promotions

You may receive e-mail spam messages advertising website promotion. These messages promise to get a website included on 1,000's of directories and search engines for a one-time fee. None of these services are productive. The service won't include any directories with any appreciable market share, and never the Yahoo directory. The service won't include any crawler based search engines that require a fee.

Spamming

Spamming no longer works. Once upon a time you could load up the internal <title>, <meta description> and <meta keywords> tags with keywords. You could hide repeated keywords with invisible text and with text the same color as the page background. You could create many doorway pages optimized for a particular search phrase. In an attempt to raise ranking on Google, you could cross-link unrelated websites in a fashion where the links could be found by search engines even though these links were not displayed on the website. These and several other spamming techniques are no longer productive. Why? Because the algorithms (computer programs) the search engines now use can detect all these gimmicks and worse, once detected, the search engine may either delete the offending site from it's database or lower the ranking of the site.