SEO Best Practices

3.1 Index Card Concept

Search Engines use the term, “index a site.” The schema comes from the library index cards invented by Carl Linnaeus, the pioneer of information retrieval. Melvil Dewey further improved on this introducing the Dewey decimal system. The schema for a web page is based on this indexing system, with all the accompanying elements.

Plate 1: Information Search the old fashioned way

The reason this is mentioned is that page design should follow the same pattern as an index card since it is schema that was used in its initial conception. As in the case of index card information retrieval, key elements in the cards helped researchers determine a book’s relevance. By the same token, Search Engine determination was originally based on the same key concepts.

The more the keyword phrase appeared in the core index card elements, the more relevant the book was. Conversely, the more the keyword phrase appears in the above mentioned web page elements, the more relevant the page is.

When creating a web page, these structural indexing elements should be representative of the content of the page. For example, putting words and phrases in the meta keywords that do not appear on the page are seen as spammy. The same goes for the meta description, it should, like the library index card, be an accurate compendium of the page. This does not mean that the meta description should be a scholarly summary, but one that concisely describes the page while still marketing of the product or service.

3.1.1 Weighted Relevance

As in the index card system, when one perused the files, book titles carried more weight than just a chapter. Keywords and summaries carried less weight than book titles. Main chapter listings carried more weight than sub-chapter listings. And so on.

The same goes for a web page. Different page structure elements will weight differently for relevance. Proper optimization can be achieved by making sure the targeted keyword phrase appears in all the core elements.

Note: in the past, webmasters began stuffing keywords in the meta keywords section. Many Search Engines ignore the meta keywords. However, this does not mean that it isn’t important and can be ignored.

Note: The meta description is an important element in that it is usually the summary listing first shown in the SERPs when a new page is freshly indexed. Making good use of this by utilizing descriptive text dialog structured to encourage buying can make a difference in a page’s click rate.

3.1.2 Key Elements

As much as we would like to include every possible search phrase in the core structural elements, there are limits to the number of characters these elements can hold. Below is a suggested table of these limits.

*Open to discussion, some claim 874 characters

Other key elements are the heading tags H1 to H6. Note that the lower numbered heading tags have higher relevance weights. In addition, key elements include; Alternative (ALT) text in images and TITLE tag attributes in anchor text.

Bold and italicized text also carry relevance weights, however they should be used sparingly and not abusively throughout the page.

3.2 Search Results Order

As described above, weighted relevance plays a big role in SERPs. As one drills down the SERPs, you will notice that less and less of the keywords appear in the listings. The order of the results, excluding those skewed by popularity metrics, is usually:

1. Exact phrase
2. All words in the phrase
3. Any Words in the phrase

Consequently, page optimization should follow this pattern. When attempting to rank a page for a target phrase, the exact phrase should appear in all the core page structure elements and should meet content densities in the body.

3.3 Densities

Relevance is viewed in terms of densities; the higher the density, the higher the relevance. However, there is a tipping point wherein the text gets over optimized. Think of a fish bowl. The more fish in the bowl the denser they are. Too many fish and they die due to lack of oxygen. The same goes for a web page. Hitting the right densities is goal in optimizing a page. Note that in calculating content density, Head elements such as Title and Meta tags are not counted.

There are two (2) types of densities. Phrase Density where the instances of the phrase are counted and Aggregate Density where the sum of the individual words in the target keyword phrase is counted with respect to the total content.

Based historical data, the sweet spot for page word count has diminished from 700 to 900 words, favoring less content of 400 to 500 words. Note that all words on a page are counted toward density; this includes text in banner, navigation and footer sections. When creating a webpage please take in to account the whole page in the body section of the HTML document.

Note: Page densities are inversely proportional to word count. The higher the word count, the lower the density and vice-versa. The lower the word count the higher the densities can be.

3.4 Uniqueness

Search Engines have improved in their ability to deliver “good” informational results. Meaning, you don’t get the same repetitive content listing after listing. In the last few years, major Search Engines have instituted as system by which, in cases of duplicate content the first page indexed will be listed and all others discarded as plagiarism. It is important to maintain unique content and not replicate content even if it is from the same site.

3.4.1 Title Tags, Meta Keywords and Meta Descriptions

Uniqueness extends to Title Tags, Meta Descriptions, and Meta Keywords. These elements should be unique throughout the site with no duplication. This means is exact duplicates. Going back to the Index Card schema, these elements should reflect the content of the page.

3.4.2 Content

It is common for large sites to have different URLs pulling up the same content. Search Engines store URLs as database indexes for content. Having duplicate content makes it difficult when working on Off-Site Optimization (aka Popularity via links) because it is not always certain which URL was initially indexed.

The common solution to duplicate content issues is to redirect the offending URLs via 301 Permanent move.