SEO Basics — Getting into the Search Engines

Search Engine Optimisation — Getting into the Search Engines

Pay for Inclusion (PFI)

There are a number of engines which offer PFI (Pay for Inclusion). Until recently the biggest of these was Inktomi but they have been bought up by Yahoo! who are now using the crawler and database for their own SERPs (they were, until a few months ago, using Google's results).

INeedHits and PositionTech are two of the boys in the earlier years who took PFI for a number of engines including FAST (Lycos and alltheweb), Ask Jeeves, AltaVista and Overture (and, indirectly, MSN). Inclusion rates are about 20 quid or so a year. What this means is that you are guaranteed your page(s) are crawled regularly (about every 3 days) — but no more. Your ranking depends on how well the copy is presented to the engines.

Bear in mind that signing up to PFI means that usually one page only is included for your annual fee, usually the index page. If you want other pages you have to pay for them. PFI does not mean the entire site is spidered, just the page you pay for. As you can imagine, it can become extremely expensive to do a whole site (and largely unnecessary, since your message is usually conveyed on one or two pages). I will often sign up two pages, the index and an attractor page — like I do for a hotel site, ensuring it's special offers are visible.

I use PFI to ensure my clients' sites are included speedily, especially for pop or topical sites/pages where immediate visibility is the criterion.

Search Engine Spiders

PFI aside, your best bet to getting into the engines is to point a link from an already well spidered site you own. I'll often point links from www.dataperception.com or another site to get the spiders to spin off to the new site. This often gets the ball rolling.

The spiders will naturally follow outbound links from sites — that's how they build a picture of the web. But. Given the volume of pages out there and the ever-increasing huge sums of money now involved with online marketing, many search engine companies are now encouraging PFI and CPC (Cost per Click) marketing models — like Google's AdWords or Yahoo!'s recently purchased Overture — by not spidering sites fully, or indeed at all, in the belief that frustrated site owners will adopt these marketing programs.

It's becoming terribly commercial out there.

The best piece of advice is to push an outbound link from an existing site and also sign up for PFI on one or two of the engines using INeedHits. You can always sign up on the 'free' spidering submission pages of Google, etc. but there is a huge wait for inclusion because of the sheer volume of people submitting and you'll likely wait at least a couple of months, if not longer.

Keywords and Phrases

Beyond the often difficult practice of getting in the engines, your other task, in fact your principle task, is to ensure your page(s) contain sufficient keywords and keyphrases to highlight the significance of its content to the web. This means determining what it is you are promoting and to whom.

There are tools such as Overture's Keyword Assistant (under Tools) which will show you common phrases people are searching on. The trick is to ensure those phrases for your site's topic appear in the title, description and, extremely importantly, the body copy of the page. But don't overdo it. A good rule of thumb is to achieve a 7% keyword density on the page. This means that 7% of the words — the body copy — of the page contain the keywords you hope people are searching on. It's an often tricky exercise because, dependent on the competitiveness of the market (mobile phones is challenging as are home products like sofas and other fittings and furnishings) you may have to up the percentage to 10 percent or, in some cases, reduce it.

To be honest, it can sometimes be a case of best foot forward and hope it ranks well first time round. You then tweak the page once you've benchmarked your ranking from the initial results.

Google's Page Rank

There's a whole lot more to SEOing and marketing a page than getting it in the engines and optimising the copy. Google, for instance, uses its own proprietary ranking scheme called Page Rank (PR). PR (which goes from 0 to 10) basically evaluates a page based on how other pages (sites) on the web consider its relevance by linking to it. My Enigma site is top on Google for 'website development standards' partly because it's well optimised but helped enormously because I've got a couple of high PR links pointing to it.

But getting a strong page rank can be difficult. When a site first gets spidered by Google it is given an arbitrary page rank based upon content quality, relevance to topic and potential popularity. This usually averages out at a PR3 or, if you're lucky, a PR4 — good enough to get you on the map. Thereafter, inbound links (sites with links pointing to pages of your site) from sites with a PR4 or above will help boost your own PR plus help to elevate you in the engines. It's a sort of commendation system.

Site and Page Structure

My own way of viewing or understanding a web page (and site, where I consider it best to go thematic, although other SEMs (Search Engine Marketers) disagree) is to consider each as a book. It has a title, a description (the content summary) and indexing keywords (although as META tags these are generally afforded minimal consideration simply because they were - and remain - spammed to bits) which generally summarise the page content. One prominent search engine apparently now offering some weight to keywords is Yahoo!, probably because it's using the old Inktomi engine which takes keywords into account when ranking a page.

The page itself is the meat. Here is where structuring comes into play.

As with a scientific text, or even a news item, you state the header then a summary followed by further paragraph headers and paragraph text until the article is finished.

Now take 2, 3 or perhaps 4 h2 headings and write a few paragraphs of copy below those. You can then follow on with h3 and h4 headings until your page is finished.

The point to note is get the order right, h1 through to h6 in ascending order. Don't bung an h2 above and h1, etc. When you read a tabloid or broadsheet newspaper the heading is always the biggest text on the page and comes first. It grabs attention (although I don't suggest you go for gimmicky spin!) and says read this first.

You don't have to use all the range of available headings, h1 through h6 but you should always start at the top and work down until you've said all you intend. I rarely ever use an h6, sometimes an h5 but it all depends on the weight of what I'm putting across, it's relevance in the scheme of what I'm offering whether it be a product, service or a political statement.

The more important paragraph(s) should appear in the page summary following the h1 header, where greatest weight is placed. Consider somebody quickly scanning a page. They want to know if the subsequent text is likely to prove relevant. They will read the first few words and decide if it's worth continuing.

Get your audience's (and engine algorithms') attention and supply more of it. But, try not to repeat yourself unnecessarily. What's the point: 'We sell spam, we sell spam, we sell spam' is dumb: it wastes your audience's time and the engines will penalise you and bury you in the SERPs (Search Engine Results Pages). You've said that once so move on and justify how good your product or service is within the following headings and paragraphs.

A word concerning the page title. This is often considered the most important element of a page. Why? Because it's what you see in the browser header bar. That simple. It's the first visual clue to page content. In my view, the title should be considered similar to an h1 element in the sense it is an important top level page descriptor — and it's the element always displayed as the first line of returned search results in nearly every search engine.

Page Hyperlinks

One common problem is finding only one page of your site visible in the engines (if you go PFI you will likely get only the one page — the index — unless you buy into more). The culprit is either JavaScript links, #include files that are not server-side script generated (ASP/PHP) or too much bulk copy in the page pushing the menu links too far down (although this is unlikely in most pages unless yours are very big).

Try to maintain all links as static href jumps as close to the to top the page as possible. If you're using an external include file between JavaScript declaration containing doc.write statements or similar these will not be seen by spiders, in which case you should also supply a complementary menu structure on the page (as you should in any case since some surfers disable JavaScript and they won't get beyond your first page).

Further, maintain context sensitive links within your body copy to relevant pages. This is important if you use dropdown menus where the link will not be immediately visible. Consider making these anchor links with copy relevant to the page either side of the link.