|
|
|
|
 |
It’s all in the planning (6) Achieving good search results According to research in various places, search engines are responsible for perhaps more than 60% of visitors to any given Web site. As a statistic, that’s a tad fatuous, you might say, in that it’s entirely possible to create a site that ranks highly on search engines for certain keywords, yet is never seen in search results. That’s because its keywords are never searched on. So, how do you achieve a good “hit” rate on a search engine? First, we need to examine how they work. Basics The aim of most search engines is to index each and every page on the WorldWide Web – a task that’s all but impossible. The idea is to visit each page (or document) and create an index of the words on it, ignoring such frequent ones as “the” and “and” and “is” and so on – in other words, looking for the meat, the substance of the document. That way, the basic subject matter of each document can be determined. Some search engines count the frequency of the words actually on the page, and assign “key words” according to how often they’re used. Others expect you to provide a list of keywords in your meta tags, and work from there, while others use a combination of the two, referencing the meta tag keyword list against their own generated keyword list for relevancy. Whichever way round the engine works, the end result ought to be a very large database of pages indexed by keywords, with some relevancy information thrown in. So, if you search on any of the major search engines using the string “Strange Brew Band” you’re actually asking them to search on the three key words. What the engine itself will do is to check its indexes and look first for documents where all three words exist – and, depending on how advanced your request is, will apply weightings to how far apart they have to be. If I’ve done my job properly, you ought to see the Strange Brew site appear in the first page of results, especially if you can limit your search to UK sites. On some, it will appear at Number One! How In order to get that kind of placement, you must first decide what manner of key words potential visitors to your site will type into search engines. In order to retain them, as well, you need to create a list of key words that are very relevant to your site. Hence the “key”, I suppose! One of the easiest ways of doing that is to list the words on each page in order of frequency, ignoring the obvious candidates (“The” “and “ “is” “but” and so on). There are tools for the job available for download all over the Web. If your writing is halfway decent, the list of words you come up with will accurately reflect the subject matter of your document, and you’ll be able to see which of your list are the important words – the ones most relevant to your site as a whole – make a careful note of these. What you then need to do is to insert the top ten or twenty into your meta tag keyword list on that page. From the Strange Brew site’s main index, here’s the statement we use: <meta name="keywords" content="Band..., Booking..., Venue, Radio, Hotel, Public, Dates, gigs., book, band!, Dates, gigs., book, band!, gig, band, Fender, Marshall, sing, Guitar,, vocals,, Telecaster,, effects., singing, bass, vocal, guitar,, Strat,, ballad, groove,, country, harmonies, group,, Vocals,, showbands, groups., ballads,, hard, rock,, country,, pop, Sound, sound., band’s, on-stage, band., BBC, Radio, Newcastle"> That statement links with the page title and the page description, both of which have a bearing on search engine placement, since they’re keyword lists as well, as far as the engines are concerned. Again, here are the one the Strange Brew site uses: <meta name="description" content="Strange Brew - The Covers Band for all reasons"> <title>Strange Brew, The covers band for all reasons</title> Weighting Now, here’s the clever bit – most search engines will read the page title and the page description first, then examine the keywords, and, finally, compare the list so generated against the body of the page to determine how relevant the list is to what the document is actually about. If you nip off to the site and compare the information here with what you see on the Index page, you’ll see that the three HTML statements between them cover every combination of key words, and also a few common mis-spellings and mis-punctuated forms. You’ll also see that there are terms on there that don’t appear in the Index page body – but they do appear in other pages of the site. That allows more sophisticated search engines to index against the site as a whole, and it doesn’t count as “spamming” the engine (trying to fool it). The page title and page description carry an awful lot of weight in this exercise, so it’s as well to consider them carefully. They need to be very descriptive, and they need to have at least the most important, and perhaps the three most important key words in them. That’s probably enough for this time – I’ll delve further into the tools you can use and how best to optimise things next time.
|
