PROMOTION OF WEBSITES THROUGH 'SEARCH ENGINE OPTIMISATION.
ABSTRACT
There has been a tremendous growth in the number of web pages since the Internets beginnings. Poor organisation and lack of standardisation of its materials make it difficult for potential visitors to find out about the existence of a business or its offerings. With an estimated 30 million commercial internet sites, attracting consumers to a web site is a challenging task. Search engines offer unrivalled exposure to the broader web audience. They also allow businesses to reach the most important audience, people actively looking for the products and services they offer. Since 7-out-of-ten users locate useful sites through a search engine, a successful site is one that is search engine friendly.
This paper presents a guide to how to create Web sites that rank at the top of search engines and have high volume traffic. It conducts a comprehensive evaluation of the web search engine, classifying them into several different types, along with assessing how important search engines are in delivering visitors to a web site. It argues search engines are the dominant method of attracting visitors to Web sites. It proposes that understanding how the search engine works, how the target audience searches, and how best to design a site from the onset can save web site owners considerable time and expenses
The author suggests that people seek information in stages but searching the web is distinctly different than traditional searches, so no effective comparisons can be made. Users have the option to engage in relatively complex web searches through search engines; however most searches of the web are short and simple.
Then, the detailed process of search engine optimisation by web site designers and marketers in order to attract more traffic to their sites is considered. Search engine optimization is a complex subject exasperated by the fast changing Internet environment. Search engines don't release data on such things as optimal keyword density or 'how to rank well' manuals and there are a considerable number of variables that have to be taken into consideration before submitting a site to a search engines. However there are some straightforward steps that should be taken to help a web site be indexed properly in search engines. This mostly involves Links, popularity, text, meta tags and keywords.
It concludes that there is no magic formula as regards search engine optimization. These techniques when used correctly will considerably improve a sites search engine ranking. But, the foundation of a successful search engine campaign should centre on a good, effective web site that delivers content for which site visitors are searching. Not quick fixes.
1. INTRODUCTION
For a business trying to attract visitors to its website, 'visibility' on the World Wide Web is an imperative (Green 2002). The majority of website designers and internet retailers have now discovered that the erstwhile philosophy of If you build it, they will come (Kinsella 1982) no longer rings true as regard online retailing [if it ever did]. Just three years ago, it was suggested that many online businesses gave little consideration to their website being listed with search engines (Thelwall 2000). Today, with an estimated 30 million commercial websites (Chaffey 2003) having a website ranked highly by a main internet search engine is generally considered of considerable importance [the relevance of search engines to the internet is introduced briefly in Appendix1]. For a site not listed with the major search engines will almost certainly miss out on a considerable amount of web traffic. Even if a site is listed, the marketing problem is apparent when a user gets thousands of hits by typing in any search word.
Submitting a website to search engines for listing is only half the battle. Most interested parties realise that successful exploitation of the online medium requires planning and strategy from the earliest stages of a websites development (Porter 2001). While the first step for some in a search engine optimization or visibility campaign is to register with as many search engines as possible, the idea of search engine visibility should not simply be about obtaining a top ranking in search results. The cleverer site designer and website marketer will design, write and create a website primarily for a site visitors [e.g. with a strong consumer value proposition]. And then help them find it via the major search engines. A search engine friendly website is a user-friendly web site that can be easily found in the search engines; as opposed to web site designed, written, and formatted to rank well in the search engines.
The contrasting view to this frequently exasperating struggle faced by the online retailer in running a successful optimization campaign, describes how many web users feel regarding searching for information on the web.; that the good stuff is hard to find. A few years ago, Furner (1997) noted the essential problems of the web as its size, heterogeneity and inconsistency. The same can still be argued today. In 1994, one of the first web search engines, the World Wide Web Worm had an index of 110,000 web pages (McBryan 1994). As of February 2004, Google claims 3.3 billion indexed web pages. While search engine technology has scaled dramatically to keep up with the growth in information, the quantity of information has also kept increasing. The web is still a vast collection of completely uncontrolled heterogeneous documents and users can become overwhelmed by the amount of information they find on It.; but unlike a library catalogue, the Web is lacking in a coherent systematic structure through which an information seeker can find desired information in the minimum of time.
2. SEARCH ENGINES AND OTHER METHODS.
There are many ways of attracting people to a website, but arguably the most significant and cost effective method is to get a good ranking with the principal search engines. Survey after survey substantiates this claim (Coopee 2000, Forrester 2001, Jupiter Media Metrix 2001, AMA 2002, Nielsen //Net Ratings12003, Overture 2003, Laundon and Traver 2004). The numbers of web users who select search engines as their method of choice generally ranges somewhere in the seventy to eighty percent region according to these surveys. Whether the actual percentage value is as high as this or lower, millions of searches are performed daily. Properly preparing a web site for search engine visibility increases the probability that searchers will visit a site.
Search engines are not the only way in which people discover web sites. There are a variety of different methods. Users may be made aware of a website through an email marketing campaign. Referral marketing techniques such as word of mouth and viral marketing are also popular methods of bringing visitors to web sites. In addition, people locate sites by clicking banner, affiliate and reciprocal links, commonly referred to a surfing the web. Web users may also find an address in an offline source such as print or television advertising. For example, many major online brands are spending money in the offline media with possibly good reason. Two high profile examples, Amazon.co.uk and EmpireDirect.co.uk have very high brand awareness amongst the online community, but a lower presence among the offline community. With more and more people coming online, it appears to this author that establishing their brand with new users is clearly a strategic aim.
Several authors over time have proposed that the above techniques are equally as important as any other in attracting people to websites (Parry 1998, Helm 2000, Godin 2001, Smith and Chaffey 2001, Doucleclick 2003). Regardless, such push methods appear to be waning in popularity, with more passive pull advertising [e.g. looking for and finding information through search engines] becoming if not the decisive, the dominant approach.
3. WHAT IS A SEARCH ENGINE?
As with many terms used on the internet this one is often misused. A search engine is a very broad term and conflicting definitions can be found. In everyday parlance:
"Search engines are computer programs that locate specific web pages, files or multimedia items stored on the internet" (Patel 2003, p51)
They can further be categorised into types of families or sources with several authors classifying them in a similar manner (Jenkins and others 1998, Rowley 2002, AMA 2002, Barker 2004): as spiders, meta-search engines, directories and portals. These terms are frequently confused with the main reason for doing so being understandable. Web users see a search button on a website and simply assume when they click that button they are using a search engine, when in fact they are using a directory. Anecdotal evidence from the authors personal experience also suggests users tend not to comprehend the fundamental difference between spiders and meta-based search engines. Perhaps a superior overall generic term to describe methods of searching the web in place of search engine could be search services. This might stop considerable confusion.
A very different question to what is a search engine? is what are the objectives of a search engine? The two should not be viewed as interchangeable, for the answer to the latter is not necessarily obvious. Many observers may suggest the objectives or goals of a search engine are to allow users to search as much of the internet as possible, perhaps stay up-to-date or even let users access information quickly. The more cynical may cite spinning off other services (shopping, advertisements, etc.) and of course, making money. Brin and Page, the inventors of Google suggested in 1998 that the goal of a search engine is to provide quality search results efficiently. Some years on, it would be interesting to know if they still held that view.
Notwithstanding this, as several authors ( Chaffey 2000, CIM2003, Suillivan1 2003 ) suggest, it is central to the implementation of a correct technical and linguistic search engine optimization strategy to appreciate the differences between search tools, as well as to have an understanding of how search engines work and what they deem important. .
3.1 SEARCH SERVICES
3.1.1 Spiders
In their truest sense, search engines are huge databases [called an indexer] that are dependent on automated search agents as opposed to human intervention. These search agents crawl the web on a regular basis around once every two months (Haryott 2003), keeping the search engine database of web pages up to date. The usual staring points for crawls are the heavily used servers of major internet providers and the most frequently visited websites (Thurow1 2003). In the act of crawling, they obtain new pages, update known pages and delete obsolete ones, building lists of words found on each page. Pages are then ranked by computer algorithm, as opposed to organising them by subject categorises as happens with directories. This type of search service is unsurprisingly called a crawler, or more commonly spider-based search engine. Google uses a spider based engine, as does AltaVista and Teoma. Most large search engines operate several spiders constantly. Even so, the web is that enormous that it can take several months for it to be fully crawled, resulting in a certain amount out-of-datedness [link rot] in all search engines.
These search engines enable end users to search for key words/phrases found in their databases. They try to match the word(s) typed in a search query with the web page that is most likely to have the information for which end users are searching. This part of the search engine is called the query processor.
There are some obvious shortcomings with spider based search services. Timely re-crawls are required which is often not feasible for all websites. The contents of websites are also not evaluated for quality. Additionally, if a web page is not linked to another page, a search engine spider will never find it. The only way a brand new page [one that no other page has ever linked to] can get into this type of search service, is for its URL to be sent by a human to the search engine companies as a request that the new page be included (Sullivan1 2003).
Ultimately, the most serious short comings of these search services relate to matters of policy. Search engine proprietors decide which sites get crawled and updated more frequently in the database. They may also skew or censor results rankings (Schneider 2002). This can mean that search engines can in theory control search results more than the searcher themselves.
3.1.2. Meta search engines
Meta search engines utilise databases maintained by other search services. They look for keywords or phrases across a number of engines/directories at the same time. Most meta-search engines default to the major search services, such as Google, Lycos, Overture and Yahoo. Others will also include Usenet searches, and other specialized databases (Liu 1999). The most popular meta-search engine at present is Ask Jeeves (Nielsen//Net Ratings US November 2003), although there are numerous other popular examples [Ixquick, Dogpile and Profusion to name just three]. Most meta-search engines integrate search results, with duplicate entries removed to present a single listing. Some also rank the results according to certain criteria [such as Metacrawler and Ixquick] and others allow specific search services to be searched [such as Surfwax and Frazzle]. Unlike individual search engines they do not collect web pages and do not accept URL additions. Instead, they send queries simultaneously to multiple Web search engines and/or Web directories. Due to their plagiaristic nature, development of the majority of meta-search engines lags behind development of spider-based search engines (Barker 2004).
Technological advances have caused two types of meta-search engine to rise above the pack. These 2nd generation meta-search engines can be termed smart and deep-investigation, for want of better phrases. Smart meta-search engines search quality databases, accept more complex searches, integrate results well and offer extra features such as clustering by subjects within a search result. Examples of a meta-search engine that fulfil these smart criteria are Vivisimo and Dogpile. Other search services have also adopted technology that organises search results by peer ranking or clusters results by subject, as opposed to more long-standing relevancy ranking. Spider-based search engines that employ this alternative 2nd generation technology includes Google and Teoma [Teoma ranks by the number of linking pages on the same subject as your search]. Deep-investigation meta-search engines offer serious researchers the ability to powerfully probe a topic in depth for aspects such as relevancy and context; an example being Copernic.
Using a Meta search tool it is suggested is often more effective than using a single web search engine (Lawrence and Giles 1998) and they are obviously attractive as users do not have to visit multiple search engines. This author suggests that the idea of meta-searching is in fact much better than the reality in most cases. It depends on what they search and how the results are presented.
3.1.3. Directories
These sites are technically referred to as web directories although many people perceive them as search engines (Korper and Ellis 2001, Schneider 2002, AMA 2002). Search directories like Yahoo or the Open Directory Project are not constructed automatically by a spider but rather by some monitored human involvement. At the time of registration, sites are included in the directory if they are thought to be of a suitable standard and then assigned to an appropriate subject category. The rationale being that this adds more meaning to the indexing or cataloguing process. Directories tend to be smaller than search engine databases and the resouces they list are usually annotated. Other than Yahoo there are a variety of different directories varying greatly in specialty and size. For example, About.com claims to have over 1 million annotations; as compared with the Librarians Index [
www.lii.org] which lists just 12,000 or so high quality sites.
As with other types of search engine there are certain problems with using directories to find information on the Web. Directories never contain the full-text of web pages they link to, so you can only search what you can see [titles, descriptions etc.]. This means general terms have to be used. Secondly, while many profess to be carefully evaluated and annotated, it is often apparent that this is not always the case (Barker 2004). Often pages change and editors dont realise it. A directory may even point to a page that has been moved or that no longer exists.
The fundamental limitations of search engines and directories, irrelevance and limited listings, have meant that both have turned to the other to overcome these disadvantages. As Swartzendruber (2003, p16) suggests search engines and directories:
"...coexist in somewhat of an incestuous relationship, where directories and engines supplement each others listings."
The foremost examples of each, Yahoo and Google now offer both a web directory and search engine to their users. It is clear that the boundaries between engines and directories are blurring with many becoming hybrid in nature (Chaffey 2003).
3.1.4. Portals
The term portal only came to prominence in 1998 (Chaffey 2000, Thurow1 2003). These websites, such as MSN , Lycos and AOL act as hubs to information on the internet. They offer web search tools as well as an integrated package of other services, such as news, email and shopping all in one place. Many of the web's well known portals began life as search engines (Laudon and Traver 2004), evolving into more multifaceted sites which provide a variety of content.
Top portals are the most frequently visited sites on the web with hundreds of millions of unique visitors worldwide each month (Nielsen //Net Ratings 2003). This is because they are more typically the first page to which users point their browser on start up, generally by default. In all probability few web users use Google as their home page, but it has still become the worlds top search engine and by starting to act in a similar way to portals [as a gateway to the web] search services such as Google are increasingly being perceived as rivals to the portals . For example, Google has recently established Google News (Waters 2003), Orkut [a social networking service] and the Google Desk bar (Notess 2003), further encroaching on traditional portal territory. Consequently, there is the beginning of a battle starting to be fought between the two to be the Webs main focal point (Hinton1 2003). In response search engine encroaches, many portals appear to be adopting the unusual strategy of sleeping with the enemy, by attempting to establish direct links with search engines. For example, AOL and BT OpenWorld offer the Google search facility on their homepage, with MSN getting its results from the LookSmart database and Inktomi.
4. TRADITONAL AND WEB SEARCHES
4.1 Search Strategies
Before looking in some depth at how web sites can be designed to be optimized for search, it may prove useful to breakdown the process of searching. Firstly, into consideration of why people use search facilities and then of how. This can then be contrasted with the online search mechanism. Choo and Marton (2003, p269) described searching as
"...a deliberate or planned effort to obtain specific information or types of information about a particular issue.....to provide a basis for developing a decision..
Searching can be undertaken for a variety of different reasons. For example, attempting to locate something known to exist or seeing if something does exist. It can also be done to get a feel for the extent of a subject. Searching for information is an important part of the consumer buying decision process. Several authors (Dibb and Others 1994, Hoffman and Novak 1996, Peter and Olsen 1999) identify common stages of the buying process that a customer is likely to go through. Searching is uniformly found after the recognition of a problem or need and before evaluation of alternatives or choice.
Putting aside the rationale behind a query, Hepworth (2002) maintains searches generally follow a set pattern. Kuhlthau (1993) suggested that people seek information in stages. Ellis and Haugan (1997) developed a model as regards the stages of the information search process, albeit within an academic context. They designated six specific categories of information seeking behaviours: starting, chaining, browsing, differentiating, monitoring and extracting.
It may not be appropriate to simply extrapolate Ellis and Haugans findings from the more traditional context specifically to the web based information seeking environment. Searching for information on the web is quite different from searching in a more traditional environment such as a Library (Spink et al 2001). Indeed Jansen and Pooch (2001, p235) described it as a
"...whole new searching environment".
From a web search perspective, numerous best practice approaches have been suggested [Green (2000), Iivonen and White (2001), Barlow (2002), Spink and Ozmutlu (2002)]. Of particular interest was Ackermann and Hartmans (2000) approach. Their 10 step method included identifying the key concepts of the search, choosing keywords that describe said concepts, determining which features such as truncation may apply, selecting a search engine and trying the same query in different search engines.
These search strategies represents the ideal. However, it could be suggested from more anecdotal personal experience, web searchers follow a less complicated method: -
Analyse a topic to decide where to begin, looking at the nature and extent of the information needed. Formulate a question generally following the 'SOAP' method. Words that identify the 'subject' or 'object' [the nouns] and 'actions' or 'processes' [the verbs]. Searchers usually select broad or specific words (Wenyu et al 2001) which become the electronic key search words (Bellizzi 2000).
Pick a starting place (search service). Execute search in a chosen search service and review results.
Based on the results, the user decides what to do. To explore the results, refine the search [if as is often the case if too many results are produced by a vague query] or give up. While this may appear odd, Wang et al (2000) concluded that searches of the Web are not particularly successful, with more than 30% of searchers not finding the correct answer to a factual query. From an e-commerce perspective there still remains considerable price dispersion on the web. The prices on books and CDs can vary significantly. While some businesses have found new ways to differentiate their products, an element of this price differentiation could be attributable to poor search strategies by consumers [of course this assumes price is a major buying criteria, which isnt always the case].
Most users will learn as they go along and vary their approach. For example, if specific keywords were unsuccessful many users will try synonyms and variant word forms (Barker 2004).
The above method of searching might seem crude; however recent studies suggest that the main body of web users adopt a 'simple' search strategy (Choo and Marton 2003). Nielsen (2000) also reports that the greater part of search queries are one or two words as opposed to search statements. Studies by Jansen and Pooch (2001, p236) found that Boolean and more advanced operators such as word truncation and wildcards are virtually non-existent at between 2-8%, stating that:
"...the vast majority of Web searches... do not use complex query syntax, and view no more than 10 documents from the results list"
Green (2002) goes further suggesting that ranking position in search results becomes the be-all-and-end-all, even taking precedence over the description of the result. From a personal viewpoint it is difficult to believe that searchers dont adopt some degree of personal filtering, as opposed to passing responsibility for the task wholly to the search engine.
4.2 Not searching the web at all?
At this point it may be useful to point out that in the act of 'searching', the user is not searching the web directly at all but merely the search tools database created at an earlier date. This is a relatively small subset of the billions of web pages residing on servers. It is not possible to search the WWW directly as a Web users computer cannot find pages unless the address [URL] is known. Rather, the user is able to access them through an intermediate search tool which provides hypertext links with URLs to pages.
While it is difficult to assess what the actual percentage of the web covered is, it is certain that search engines only index a fraction of the total number of documents (Schneider 2002). Bergman (2001) notes certain types of web pages are excluded from many search engines by policy, being too resource intensive or simply because they cannot be accessed due to technical barriers such as firewall or passwords. Often referred to as the 'Invisible or Deep Web' [as opposed to the visible web-what you can see in results from general web search engines], this is estimated to be 500 times larger than traditional search engines can index (Bergman 2001).
5. OPTIMISATION
5.1 Principles of website optimisation
From the earliest days of the web a considerable amount of research has gone into analysing search engines and the methods they adopt for classifying web pages and sites (Brin and page 1998, Chun 1998, Kirsch 1998, Spink et al 1999). Notwithstanding this, one of the difficulties associated with the evaluation of search engines is that they are constantly refining their search mechanisms and user interface. In fact, search engines change their relevancy algorithms every day (Thurow2 2004) and exactly how individual engines work is a proprietary secret, as each engine does things a little differently. It is pointless therefore to build recipes or such like for the perfectly optimized web page. However, there are certain basic principles of website design and search engine friendly copywriting that should be adhered to in order to improve visibility to search engines and have a site ranked highly:
5.1.1 Rules of web design
To get the best search engine visibility, web designers should follow the basic rules of web design (Nielsen 2000). A site should be easy to read, easy to navigate [a breadcrumbs text-based system-including a site map- as opposed to image-based system (Hill 2003)], consistent in layout and design, quick to download and easy to find. It is interesting to note that both search engines and users are looking for similar characteristics [e.g. search engine optimisation and good website design practice go hand-in-hand]. For example, good website design is to make any page on a site accessible within 3 clicks (Nielsen 2000). This is important from an optimisation viewpoint as the majority of search engines only spider three pages deep on a site. In a similar way, users are looking for quality content and so are search engines. Search engines give a higher ranking to Websites that offer many pages filled with high quality text content.
This author would also add one additional rule to Nielsens list. A website should have its own domain name. Users dont always remember to bookmark sites and long domain names are easily forgotten. More importantly, serach engines and especially directories are more likely to index the root domain than subdirectories (Sullivan2 2003).
It is only worth driving traffic to a website if the consumer experience is going to be a positive one. It is relatively pointless being ranked at no.1 postion with one of the major search engines, if the site in question is labourisly slow to load. Most visitors will not wait for the page to download. The basic web site design rules still have to be obeyed, as driving traffic to a site where the expereince is poor will mean those visitors are lost forever. For a website to have any real chance of success, repeat visits are essential.
5.2.1. Main parameters
There are several main parameters on which search engines base the order of their ranking. These are mostly concerned with links and popularity, meta-tags, text and keywords.
5.2.1.1 Links and popularity
The way web pages are linked to each other affects a sites search engine visibility. The assumption being the more links there are too a site, the more useful the site will be. Attaining an optimal link popularity component is not as simple as obtaining as many links as possible as search engines measure both link and click-through popularity of websites. While there should be a link development campaign (Chaffey 2003), link popularity is increased not just by the number but the quality of links pointing to a page (Laundon and Traver 2004). Having links to a site from relevant, high-ranking pages [such as Yahoo] can greatly increase sites link popularity (Sullivan2 2001). Marketleap offers a link popularity analysis tool.
Alta Vista uses link analysis and ranks sites based on 'good' link popularity. Google uses weighted link popularity and analyses link content [ Googles Pagrank algorithm] to determine site rankings. Inktomi uses link popularity as the main ranking criteria (Sullivan3 2003).
The way web pages are linked to each other also affects a sites search engine visibility. If search engine spiders can find a site quickly and easily, it has a much better chance of appearing at the top of search results.
5.2.1.2 Title and Meta-tags
All pages should have the best possible title and meta description tags (Alimohammadi 2003), which should always be placed at the head of the document (Bradley 2002). A meta-tag is a HTML tag that gives information about the contents of a web page, such as the HTML specification a web page follows. For search engine visibility, the most common use of meta tags are the keyword, description and robots exclusion attributes.
One of the most widespread beliefs about search engine optimisation is that meta tags are the secret ingredient to obtaining a top ranking. In fact, only some major search engines use meta tag content for relevancy (Thurow1 2003). Others, including almost all directories do not use meta tag content at all.
5.2.1.3 Text and Keywords
Web pages that contain the words that a websites target audience is typing into search queries generally have a greater search engine visibility than pages that contain little or no keywords. The first 2 or 3 keywords listed should be the words they focused on as a site is optimized. Keywords should be placed high up in pages of well written content (Spencer 2002). Search engines should be checked to determine the most relevant and popular keywords that the target audience will use (Ostler 2001), as opposed to how site marketers would like them too search. Overture offers a free search suggestion tool (Swartzendruber 2003) which has information on how many times a particular keyword or phrase was searched for during the last month.
An interesting exercise is to investigate competitors sites. By performing a search on a sites main keywords
the user will presented with a list of sites with which youre competing for the keywords you want to use. By looking at their source file a user can clearly see a compettiors keywords.
5.1.2.4 Worst practice and bells and whistles
Worst practice should be avoided. Some website designers repeatedly use a word or a meta tag on a page. Known as cloaking, this can also be done by typing text in the same colour as the background in a tiny font (Korper and Ellis 2001). Search engines 'punish' or even ban altogether sites that attempt practices such as 'keyword stuffing', 'spider traps' or doorway pages. They analyse pages to determine whether a word is used too frequently in natural language. The current keyword density considered safe is between 6-12% (Swartzendruber 2003).
Many sites are penalized in search engine performance by the technological advances that enhance the experience for users (Ostler 2002). A search engine only sees the skeleton of a site, its source code. It doesnt see beautiful graphics or fancy
design. So dynamic content delivery using Macromedia Flash, and excessive use of Frames or JavaScript hinders performance in search engines and can make a site more or less invisible to search engine spiders (CIM 2002). Search engines consider home pages as the most important page (Spencer 2002). Having an opening Splash page is a wasted opportunity.
5.1.2.5 Search engine submission
It is important to make the distinction between search engine submission/registration and search engine optimisation. These phrases are often used interchangeably to discuss different efforts to promote sites on search engines. Submission refers to the act of getting your site listed with search engines. This does not necessarily mean that the site will rank well, just that the search engines are aware it exists. The American Marketing Association (2002 ,
www.marketingpower.com/live/content)
describes optimisation and phrases such as search engine marketing or placement as:
.the practice of guiding the development of a website so that it will naturally attract visitors by winning top ranking on the major search engines for selected keyword phrases
Notwithstanding this, planning effective search engine and directory submission campaigns is important to a sites search engine visibility. The guidelines for submitting to search engines are completely different from the guidelines for submitting to directories. Directories usually give web site owners one chance to submit correctly. Also, because they are the staring point for many search engine spiders, submitting a site to directories is the first step in effective search engine submission. All the major crawler-based engines [Google, Teomi, Fast Search] measure popularity. The quickest way to get site popularity is to get it listed in the major directories [Yahoo, Open Directory].
This author believes there is an argument that the 80:20 rule applies as regard search engines. That 80% of the searches are conducted on 20% of search engines. Certainly the bigger search services generate the greatest number of visitors for a website. As the top 10 search services account for the majority of the traffic it is prudent to concentrate most effort in this direction. It could be suggested that in designing a site Google-friendly site, it will become friendlier for other crawler based search engines too.
5.2.2.5 Ongoing process
Improving the ranking of a site in a search engine is an ongoing process. Green (2002) suggests that companies should allow one year for evaluating success in terms of increased sales. Search engines should be monitored on a regular basis, with ranking checked. An engine that takes a submission one week may drop it the next. Site reports will show which search engines 'spider' a site.
5.2.2.6 Paying to be listed
There appears to be a definite even dramatic'(Chaffey 2003) move towards a 'pay for inclusion/ranking model' by the major search engines (Laudon and Traver 2004). The paid search market was worth $1.26bn in 2002 (Harvey 2003) and is a 'booming business' (Elgin 2003). The leaders in this technology are Overture [recently acquired by Yahoo] and Google. Inktomi and Looksmart have also begun some kind of charging for inclusion.
There are 2 types of paid for model. Pay-for-inclusion and Pay-for-placement. With pay-for-inclusion, a search engine includes pages from a web site in its index in exchange for payment. The pay-for-inclusion model is beneficial to search engine marketers/website owners because they know there site will not be dropped from a search engine index. Also, any new information added to their web page will be reflected in the search engine very quickly. Of course, this does increase the life or death stranglehold certain search engines hold over web based businesses. In contrast, pay-for-placement guarantees top positions in exchange for payment. With pay-for-placement, participants bid against each other to obtain top positions for specified key words/phrases. The higher the bid, the higher the web page ranks. The US Federal Trade Commission recently recommended that search engines improve disclosure of paid content within their listing (Green 2002). Many search engine sites, most prominently Google, have consequently begun to feature targeted advertising or smart banners (Wenyu et al 2001) next to search results of pre-determined keywords (Hill 2003).
The 'tricks' to be at the top of the search engine list spread rapidly making it virtually impossible to stay at the top of the list for long. With search engines throwing up an ever increasing number of results from the abundance of sites on offer, some degree of paid listing may look increasingly attractive to companies needing to stand out. Despite this a large number of search engines do not charge for inclusion or placement and the major engines still accept non paying sites. They also still send their crawler programs around the web indexing and including pages in their databases without charge. Indeed, the American Marketing Association (2002) suggests paying for listing can effectively trap a site into a pay only scheme.
6. SUMMARY
Search engine optimisation is a powerful online marketing strategy. When done correctly, it can help a website tap into what many website marketers term natural traffic, from the millions of web searches made daily.
Many web site owners consider search engine optimisation as an afterthought, after a site has already been built. They spend considerable time ensuring that their web sites are compatible with the two major browsers while ignoring the third browser, search services. Building a site thats accessible and friendly to search services is arguably just as important as thinking about how people use Internet Explorer or Netscape Navigator. This is because many of those people will locate web sites by means of these search services. Fortunately, building a website for search services doesnt mean ostracizing human visitors. Rather the opposite. Thinking about optimizing a site for search can often make a sites usability for users better than it was before. A web site that pleases sites visitors, the search engines spiders and the directory editors.
The best site design solution involves planning, as opposed to rectifying problems retrospectively. If a new site is being built or even an existing one being redesigned, understanding how the search engine works, how the target audience searches, and how best to design a site from the onset can save considerable time and expenses. Building a website correctly isnt difficult. Nor is it about using the myriad of tricks that many employ to fool search engines into a temporary high rating. There will always be ways to trick search engines. The reality is, search engines will eventually discover these tricks and add filters to their algorithms to catch out the sites that employ them. This leaves webmasters that employ these tricks in a constant battle to out-race their competitors, employing newer and greater tricks to win the search engine game; that generally strikes against them from the outset. On the other hand, the webmaster that takes the time to properly build a useful, well designed site tends to profit from more stable search engine placings and arguably suffer less grief in the long-run.
Making a site search engine friendly often means implementing small and simple changes that can have a big impact in gaining search engine visibility. A successful search engine campaign consists of several key components. Links, popularity, text, meta tags and keywords. Websites that utilise these key building blocks gain far more than sites that do not use them. Notwithstanding this, the foundation of any search engine optimisation campaign begins with a good, effective website. One that delivers the content for which site visitors are searching. Search engines do help, but unquestionably many web marketers use the black art of search engine optimisation as a lazy mans pancea, masking poor website design and a poor consumer value proposition. Without a strong foundation, a search engine marketing campaign will ultimately fail, which costs web site owners time and money.
