Understanding SEO

Search Engine Optimisation, or SEO, is the hot topic of the moment because it promises higher rankings by search engines and therefore greater traffic. Unfortunately, there is more hype than substance in much of the advice about this technique. To cut through the optimistic claims, we need to look at how search engines work, where they get their data, and how they come up with their rankings. Without that understanding, we cannot focus our optimisation work in the right place.

Search engines such as Google take in a query through a web page on a web server and this is then forwarded to an index server. The index server is much like the index of a book but in this case it contains links to document servers that will contain links in turn to documents of interest. The query expression produces a focused group of links to document servers which in turn can provide the results of the query. The optimised structures of these servers enables the user to get a very fast response to the query.

Clearly then, if we want to come high up in the search results, we need to be very well matched to the query, and also have a presence on those servers, the index and document servers. Companies like Google obtain the data about web pages by means of crawlers, or spiders, computer programs that methodically traverse the web collecting relevant search engine data. It's that data that we want to be picked up from our site so we have to make sure it is there.

There is a certain amount of commercial secrecy about the nature of data obtained by spiders and the relative significance given to them but there are some common elements which are clearly important.

Keywords

The most common search terms for any particular subject are the keywords, words that appear most often in the searches, and are most specifically related to the target of the search. These vary in specificity from the very general to a specific make and model of a product. The site we want to be recognised by search engines must contain the relevant keywords. The question is where should those keywords appear. If they litter the site, the search engine may even reject the site altogether and not list it. This is in reaction to some site administrators who pack their text or code with keywords in an attempt to increase their ratings. Normally, the spider is looking for page titles, site name, article headings and text, but also in image tags and links. Main pages should use the keywords in their title rather than the name of the company, for example.

A decade ago, it was common to place a list of keywords in a section of the HTML page code in a Meta tag but web spiders are now much more sophisticated. It is still worthwhile doing this, but its importance is reduced. Similarly, the use of the keywords in the Meta description is still useful but less significant.

Inbound links

The search engines do not just consider the content of your site, but also the number of indexed links it can find which link to your site from outside. These are called inbound links and represent a measure of the site popularity, a kind of voting system. If the content of your site is worthy of a link from someone else, they think your site is worth reading, and therefore the search engine takes the site more seriously. However, search engines do not simply count the inbound links, but also make an effort to judge the relevance of the link, and the quality of the source. If you are linked to by high-ranking sites, that will count for more than those from low-ranking sites.

Listing by authoritative sites

There are many sites which act as web directories, listing sites according to some classification system. These sites vet the applications and check that the sites really do represent what they claim and this vetting process gives the directory site some status which search engines acknowledge. As a result, a listing from a reputable directory site such as DMOZ, or Lycos, or Business.com, will count higher than some others.

Internal links

These are less important than the inbound links but still enable the spider to find additional material to index. Internal links that also use keywords are useful but not all site structures lend themselves to this. Sites which are structured using template systems such as Joomla or Drupal, Wordpress of one of the many freely available content management systems, use a database to store content and when the internal links are created, these may not be friendly to spiders. Internal links may use coded names designed for efficiency when using the database and therefore will not match the keywords. Some content management systems, such as Joomla, contain configuration options which allow the administrator to turn on SEO support so that the URLs of the pages contain words which can be parsed by the spiders. This can help get the content of the site indexed correctly.

When images, audio and video are added to sites, there are additional tags which provide text explanations and these should contain the relevant keywords as spiders will check these too.

Summary

The search engines are trying to rank the content of web based on relevance and popularity and the best you can achieve by SEO is making the most of your content. But the content is the most critical factor because without excellent content, you will not get the inbound links. Without a vetted entry in an authoritative directory, you will not be seen by the search engine as a major player.

SEO is about optimising, not about fooling the search engines. It will take time for the techniques to make a difference and it is possible that they will even make little difference. Sometimes a company specialising in these techniques can add value because it can be very time consuming ensuring that the relevant keywords appear in the right quantity in the right places. We should though be cautious in accepting claims for dramatic improvements.

Search engine companies do not make public the algorithms they use and the SEO skills on sale are based on careful observation and conjecture. At least some of those conjectures will not be right. So the best advice is to concentrate on high quality content which attracts traffic and links, getting registered with the key industry directories, and treating SEO as an additional technique to make the site easier to index. Where the content is database driven, some care needs to taken over internal links.