Architecture is defined as a unifying or coherent form or structure.  Thus, it is fitting that site architecture is the manner in which the components of a computer or computer system are organized and integrated (Merriam-Webster, n.d.).  Simply, it is the process of organizing Web site content. 

Any successful Web site has been designed with integration as a foundation, as site architecture controls the flow and content of a Web site.  When you have quite a bit of information, you want to ensure you separate in an organized way that leaves the Web site visitor not feeling overwhelmed by the amount of information.  Additionally, not all of the information may be relevant to each visitor, so you will want to ensure they can quickly filter out the information they will not need.  Thus, an organized structure and flow are necessary for successful Web site navigation.

When evaluating the successfulness of a site's architecture, we must evaluate the following:

•·         Directory structures, which will include an overview of HTML, root directories and sub-directories.

•·         Navigation schemes, which will include text-based versus graphics-based schemes, multimedia effects, pull-down menus and hypertext links.

•·         URL structures, which will review domain names, file names and hyphenation.

•·         Web page types, which will explore the various types of pages, which can include Home page, services page, and search and search results page.

•·         Page layout, which covers how the different types of Web pages may need different layouts to be effective.

•·         Cross-linking, which will review vertical and horizontal cross-linking and then navigational linking to related pages.

It is also imperative that the site architecture flow with the business plan and Web design.  Also needing some exploration is why many search engine marketers have only specialized in advertising and have not focused on site architecture.  We will explore search engine optimization (SEO) as part of the overall site architecture. 

Directory Structures

To understand directory structure, we must first understand root directories.  The top directory in a file system, the root directory is provided by the operating system and has a special name.  For example, in DOS systems, the root directory is called \ (Webopedia, n.d.).    

Generally, your most important pages on your Web site are closest to the root directory.  Your root directory should include the home page (index.html) and the Robots Exclusion Protocol (robots.txt).  The robot exclusion standard, also known as the Robots Exclusion Protocol or robots.txt protocol, is a convention to prevent cooperating web spiders and other web robots from accessing all or part of a website which is otherwise publicly viewable (/ROBOTS.TXT, 2008).

By placing the most important Web site pages are at the root level, you are ensuring that the search engines and visitors to your site are viewing the important pages.  Generally, a Web site should have the following subdirectory structure:

/cgi-bin, /css, /images, index.html, /logos, /pdf, robots.txt, /scripts

HyperText Markup Language (HTML) is the main markup language for Web pages.  HTML provides a way of describing the structure of text-based information in a document.  It does this by expressing text as links, headings, paragraphs, lists, and so on.  An HTML file is a text file containing small markup tags that tell the Web browser how to display the page.  The HTML will supplement that text with interactive forms, embedded images, and other objects (XML Master, n.d.).

With the directory structure, the URL for a home page would be http://www.yourdomain.com/index.html.  For a B2B Web site, the URL for the services page would be http://www.yourdomain.com/services.html. 

Navigation Schemes

A Web spider, sometimes referred to as a Web crawler, is a type of bot that starts with a list of URLs to visit.  As the spider visits these URLs, it locates all of the hyperlinks on the page and adds them to a list of URLs to visit.  The Web spider from one machine sends HTTP requests for documents to other machines on the Internet (Blum, Keislar, Wheaton and Erling, 1998).  HyperText Transfer Protocol, or HTTP, is the protocol of the Web and determines what you can and cannot send or receive on the Web.  Many Web sites, including search engines, use Web spiders to provide current data.  Web spiders can create a copy of all the visited pages for the search engine to index as downloaded pages to provide faster searches.  Web spiders can also be used to automate certain tasks on a Web site, such as checking links or verifying HTML code.

A Web site's navigation scheme can be thought of as a map, depicting how the various pages relate to one another.  As with any map, it should show how visitors will travel through the site as they click on links or use the application interfaces.  The navigation scheme should be reflected in the navigation bars that are used throughout the various Web pages.

Now that we have defined Web spiders and given a general overview of navigation schemes, it is important to note that some navigation schemes are more spider-friendly than others.  Usability tests have shown that navigation bars and hypertext links are more spider-friendly than D-HTML pull-down menus and navigation buttons (Thurow, 2004). 

While you do want a Web site to be spider-friendly and to have a cohesive map, it is important to remember the visitor's view.  The Web site should be one in which visitors find visually appealing with ease of navigation throughout the site.  Also try to avoid application interfaces that require users to make additional downloads before they are able to proceed with viewing the site's content.  This could result in visitor frustration and cause potential customers to not return to the Web site.  Focus groups are a great way to determine what visitors find appealing in terms of site navigation schemes. 

To view a site with poorly planned Web site navigation, please visit http://web.archive.org/web/20070124205212/http://www.deptofenergy.com/.  Once you click on the link, your browser will actually be re-sized, you will not be able to tell what the business is, and I believe the flashing is actually something that can lead to seizures for some visitors.  Because of the volume of complaints about the original site, whose site gave you no indication they were an energy company other than the URL, the company re-designed their site in 2007.  The new site can be viewed at http://www.deptofenergy.com/.  This new design uses navigation much more successfully.

URL Structure

Search engine spiders may not be able to reach pages not closely tied to the root directory, which is something to consider when developing sub-directories.  It may be helpful to use keywords in the URLs of sub-directories so that the search engines will recognize the page.  An example could be the difference between insurance-sales.html and sales-insurance.html.  If a search engine returned both of these, which one would direct viewers to the appropriate site?  It is critical to have a consistent formation of URLs so users are sure to know the Web site is specific for the information they are intending to locate.

Also, file names should include keywords and not generic terms.  As with our example above, if your company sells three types of health insurance, it would be a mistake to call your products something like this:

www.abcinsurancecompany/pdt1.html

www.abcinsurancecompany/pdt2.html

www.abcinsurancecompany/pdt3.html

This only tells a search engine that you offer something called pdt.  A better example would be something more along the lines of:

www.abcinsurancecompany/flexspendingaccount.html

www.abcinsurancecompany/healthsavingsaccount.html

www.abcinsurancecompany/medicareadvantage.html

You would also want to ensure that your home page links to each of these pages.

Another area of controversy is whether to use hyphenations in domain names and file names.  Hyphenations are recognized by most search engines as text dividers in URLs, while most search engines like Google ignores underscores (Faceries, 2004).  Using hyphenations can be appropriate if, for example, your company is named ABC Insurance Company and that domain name was already registered to another company.  If neither domain has been registered, purchasing both can be a good way to protect your brand.  One disadvantage is that visitors may not remember to use hyphenations and may have difficulty locating your business or be directed to another business site. 

Web Page Types

When searching for information on a site, each page visited must either provide the content being sought or take them closer to that content.  There is some debate over how many types of Web pages exist.  Regardless of the exact number, the main types of Web pages typically include the following:

•·         Home Page - Often referred to as a landing page, the home page is considered by some as the least important page on a site.  Its main purpose is to send visitors in the right direction.  Also, depending upon the size of the site, the home page can be a combination of other pages described below, which would increase its importance.

•·         Content Pages - The content pages should contain the content the visitors are looking for and are considered the most important Web pages.  The content pages allow users to gather information and visit more content pages to obtain additional information.

•·         Galleries - This page is the most important "link in the chain," as it is a listing of links to content pages.  Visitors will select their content page from within the gallery.

•·         Departments - For larger sites that have too many links to only have one gallery, a Department page is used to list all galleries.  Visitors should clearly be able to know that the department pages listing all galleries are different that the gallery pages listing all content pages.

•·         Search and search results page - For sites that have a great deal of content, a search feature allowing visitors to easily scan the contents of the site to easily find what they are looking for will make navigating the Web pages much more user-friendly.

•·         Product page - While this will not apply to all Web sites, the product page should list all products being offered by the company.  This can be done in a simple list, as hyperlinks to pages giving more detail or in a catalog format.

•·         Services page - The services page should cover all services being offered by the company hosting the site.  As with the product page, this may not apply to all Web sites.

•·         Forms page - The content on the forms page will give any forms the company is accepting from consumers and will not apply to all Web sites.  There are variations for the form page, as well.  Some company Web sites will choose to have forms available in HTML or PDF, while other sites will have forms that can be completed online and sent directly back to the company.

•·         News/media page - The news/media page will highlight any noteworthy news the company wants its visitors to be aware of, as well as highlighting media coverage and press releases for the business.

•·         Shopping cart pages - For companies who offer a product page, it may be more convenient to offer visitors the ability to place orders via the shopping cart pages.  While shopping, the consumer can place items they want "on hold" in a shopping cart and once they are finished shopping, proceed to their shopping cart to place their order.  This will usually include the ability to make a payment, which will require payment pages, as well as shipping and billing pages (Spool, 2005).

In visiting http://www.walmart.com/ , you can see that there are various page links vertically along the left side of the page, horizontal links at the top and bottom of the page, and image / text links throughout the center of the page.  The Shop pages are laid out in a different format than the Services and Help pages.  Likewise, the Online Directory page is formatted much differently than the Terms of Use page. 

Page Layout

Because there are many different types of Web pages, the layout for one type of page, such as a shopping cart page may not work for a different type of page, such as the forms page.  Visitors to your Web site would not expect to locate your product information on the News page or Home page. 

Search engine visibility can be improved with proper Web site page layout.  Google's PageRank was created just because page design and layout are so important. 

PageRank relies on the uniquely democratic nature of the web by using its vast link structure as an indicator of an individual page's value. In essence, Google interprets a link from page A to page B as a vote, by page A, for page B. But, Google looks at considerably more than the sheer volume of votes, or links a page receives; for example, it also analyzes the page that casts the vote. Votes cast by pages that are themselves "important" weigh more heavily and help to make other pages "important." Using these and other factors, Google provides its views on pages' relative importance (Google, 2008).

Another important factor for search engine visibility is download time.  Ideally, the Web site should have a Web server that gives the search engine spiders the Web pages as quickly as possible.  Search engines measure download time, as do visitors to your Web site, as well as cloaking, which is delivering one version of a page to one visitor and another version to others (Webmaster World, 2007).  If a cloaked page does not have the same file size as the actual page that visitors see, the search engines realize that the Web site owner is likely participating in Spam practices.

Search engines also prefer files that are 100k or less.  If an HTML file is larger than this, some spiders will not read past the first 100k of information.  PDF files are typically read up to 750K, as is the case for AltaVista.  Google actually indexes up to 2,000K for a PDF.  Therefore, the size of your files could have an impact based on the visitors search engine choice and how they are reaching your Web pages (Thurow, n.d.).

Cross-Linking

We can break down internal Web site cross-linking into two categories: vertical and navigational linking.  To understand both, a basic understanding of breadcrumbs is necessary.  Bread Crumbs are a series of links that leads to the current page. The visitors navigate the website with ease, because the visitor can go back and forth with ease (ABC Article Directory, 2008). 

Cross-linking, also described as internal link architecture, refers to how you inter-link various pages within your own Web site (The Online Marketing Guy, n.d.).  This is vital for allowing visitors to navigate through your Web site to various topics and to link to other areas of your site when items are related.  It is not necessary to link every page to every other page on your Web site.  This can be over-kill and could cause too many links to be displayed, causing frustration for your visitors. 

To see vertical cross-linking, you simply need to verify the breadcrumbs.  An example of this might be:

Home > Health Insurance > Health Savings Account > Compare Available Products

If visitors want to know more about health insurance, they can click on the "Health Insurance" text link.  If they want to find out more about Health Savings Accounts, they can click on the "Health Savings Account" text link.

Many Web sites, however, fail to group related pages through navigational linking, which can sometimes referred to as related or horizontal cross-linking.  Navigational linking helps organize the Web site content into logical groups for users.  The most easily used navigational links are text links, followed by linked images.  Text-based links can be used with most Web sites to pass value between the Web pages.  Linked images can be clicked on as a button to go to the next page.  Some Web sites have also been designed with Javascript-based navigational menus.  While these can appear like text links, they do not work with spiders and are typically not indexed by search engines because they are not seen by the spiders.  In these cases, bread-crumbs can be added to help the spiders find the information and ensure search engine optimization (The Online Marketing Guy, n.d.).  For example, once the visitor has found information about the Health Savings Account they are interested in, what do you want them to do?  From the "Product Information" page, are they able to link to the "Apply Online" page or the "Contact Us" page? 

Business Plan

The business planning process has several components, but the primary purpose is to identify business goals and measurable success factors.  The Web site architecture is the IT roadmap for implementing the business plan; it should provide the implementation details for the site.  Too often the business plan and Web site architecture are designed and implemented by different areas who are not communicating with one another.  Ultimately, the Web site should not be designed reactively to reach a new market not considered in your original business plan, but proactively as a way of completing your existing business plan.  The Web site architecture and business plan should be constantly reviewed together and adjusted cohesively to changing business requirements.

Additionally, a successfully designed Web site should be planned with future growth in mind.  The design should be scaleable to be useful and to grow with the business.  Too often, a Web site is implemented, only to be forced to have a major re-design every time a new product or category needs to be added.  When a visitor finds your Web site online and receives a "We apologize for any inconvenience, but this site is temporarily unavailable due to maintenance" or "Site down due to construction," this can lead to lost sales. 

Search Engine Optimization

Search engine optimization (SEO) requires that Web sites be well-designed and easy to navigate.  To put simply, SEO is just an extension of proper Web page design.  By making a Web site that is not too complex and readily accessible, the search engine robots have the easiest route to index the site.  This also creates a better experience for the visitors to the site. 

Search engines change their search criteria based on their company, their version, etc.  Some search engines only specialize in advertising or search engine optimization.  The search engine companies typically do not create Web sites, nor do they perform usability testing on page layout, design and navigation schemes.  An ethical search engine marketer will follow the terms and guidelines set forth by the search engines.  Search engine rankings tend to rely on three components:

1.      Keyword rich text - A page must contain the words and/or phrases that people type into search queries.

2.      Search engines must be given easy access to that text.

3.      Other people must consider a page's content to be valuable. If a person thinks content is valuable, that person tends to link to the page that contains the content (Thurow, 2004).  

Of course, no search engine measures these components in the same manner.  Rather than attempting to keep up with each search engine's criteria, it is better to focus on good text and good links.  This will ensure that the information on your site is reliable, accurate and easily found by the search engine robots. 

It is also important to note that search engine ranking is not the same as search engine indexing.  Search engine ranking determines how to rank the content, which is a technical task.  Search engine indexing is getting search engines inside your site to understand the content, which is part of a business plan or strategy.

According to OneNaught.com (Shah, 2007), some steps for improving indexing include:

•·         Sufficiently clean URLs to avoid page weight dilution

•·         Good use of the <title> element

•·         Proper use of redirects and other HTTP status codes to help search engines

•·         Sitemaps

Typically, there is very little a developer can do to improve ranking, as this is mostly determined by a site's popularity.  As part of the business strategy, a business should determine how to have content good enough to cause other businesses to want to link to the site.   This will ultimately lead to higher ranking from search engines.

Conclusions

Web site architecture development and maintenance is not an easy process.  The site's directory structure, navigation schemes, URL structures, page types, page layout and cross-linking may work on well on some pages but not with others.  It is important to develop clear and logical layout and design to ensure overall functionality, scalability and usability for a Web site.

It is important to develop site architecture as part of the business plan to prepare for the future of the business.  Additionally, search engine optimization should be determined by the usability and friendliness of the site.  Being ranked by a search engine through advertising is not the same as being indexed by a search engine. 

References

Architecture. N.d. Merriam-Webster Online. Retrieved May 5, 2008 from http://www.merriam-webster.com/dictionary/architecture.

Webopedia. N.d. Root Directory.  Retrieved May 7, 2008 from http://www.webopedia.com/TERM/r/root_directory.htm.

About robots. (2008, April). /ROBOTS.TXT.  Retrieved May 7, 2008 from http://www.robotstxt.org/robotstxt.html.

Introduction to HTML. XML Master. N.d. Retrieved May 8, 2008 from http://www.w3schools.com/html/html_intro.asp.

Blum, Thom; Keislar, Doug; Wheaton, Jim; and Wold, Erling.  (1998, January). Writing a Web Crawler in the Java Programming Language. Retrieved May 8, 2008 from http://java.sun.com/developer/technicalArticles/ThirdParty/WebCrawler/.

Thurow, Shari. (2004, November). SEO and Successful Site Architecture. Retrieved May 10, 2008 from http://www.interactivereturn.com/successful-site-architecture.htm.

Faceries, Fabien. (2004, August). Google test: hyphen and underscore. Retrieved May 12, 2008 from http://www.prweaver.com/blog/2004/08/26/2-hyphen-and-underscore.

Spool, Jared. (2005, November). The 8 Types of Navigation Pages. Retrieved May 14, 2008 from http://www.uie.com/brainsparks/2005/11/28/the-8-types-of-navigation-pages/.

Google searches more sites more quickly, delivering the most relevant results. (2008). Google. Retrieved May 15, 2008 from http://www.google.com/technology/index.html.

Cloaking. (2007). Webmaster World. Retrieved May 18 from http://www.webmasterworld.com/cloaking/.

Thurow, Shari. N.d. Download time and search engine visibility. Retrieved May 20, 2008 from http://www.searchenginesbook.com/downloadtime.html.

Ezine Ready Article.  (2008). ABC Article Directory. Retrieved May 20, 2008 from http://www.abcarticledirectory.com/ezineready.php?id=7279.

Cross Linking and Internal Link Architecture. N.d. The Online Marketing Guy. Retrieved May 22, 2008 from http://www.theonlinemarketingguy.com/cross-linking/.

Thurow, Shari. (2004, November). Search Engine Spider, Index and Ranking. Retrieved May 24, 2008 from http://www.webpronews.com/topnews/2004/11/29/search-engine-spider-index-and-ranking.

Shah, Anup. (2007, September). Explaining Natural SEO: Search Engine Ranking vs. Indexing. Retrieved June 5, 2008 from http://www.onenaught.com/posts/30/explaining-natural-seo-search-engine-ranking-vs-indexing.