Search Engine Concepts: search marketing services

HOME | ABOUT | PORTFOLIO | RESOURCES | CONTACT

Top 5 SEO Design Mistakes

Search engines need to be able to 'read' the information contained in a web page in order to assess its relevance for a search query - however, some techniques aimed at making web pages more attractive or more functional for the user can also end up obscuring the content from search engines, leading to poor rankings in the results. For this reason, it's important to consider search engine optimisation from the outset and ensure the design is search-engine-friendly, to avoid having to re-design it later.

Below I present what I consider to be the top 5 website design mistakes in SEO terms - not necessarily the worst mistakes, as most can be fairly easily remedied, but those which I see most often in my day-to-day consulting.

1. Flash, Java and Javascript

Web site designers LOVE Flash! (It gives us a chance to show off our creative skills…) However, unless there's a strong argument for benefit to the user, it's probably best to steer clear of it as a format for your whole site if you want it to be search-engine friendly, because at the moment automated search engines can't read it - and unless you also create a non-Flash alternative, directories often won't include you either.

That's not to say that a site can't combine Flash and non-Flash content - just make sure that the search engines can see plenty of non-Flash, fully optimised content too (and YES, that absolutely applies to the index page - if your site opens with a gorgeous Flash intro but nothing else on the page, the search engines will ignore it altogether, and directory editors will probably leave before the animation's finished).

Don't be tempted to use cloaking software, redirection scripts or META commands to get around the problem either - these techniques are known to be used by spammers and could result in your site being banned from the index.

You also need to ensure that all your pages are easily accessible to maximise your chances of achieving good listings - this means making sure that your links can be followed. Use of Java and Javascript can sometimes render links unreadable to engines, so if you're not sure, play safe and include text links at the foot of every page, or use the 'NOSCRIPT' tag to make the link available.

Alternatively, create a site map or list of contents page with plain text links to every page on your site, and make sure all your pages also contain a text link to this page.

2. Poor quality, irrelevant or unoriginal content

Content is still king on the web, and no matter how well optimised for keywords your pages are, if they don't have anything interesting to say it's unlikely they will ever do well in the search engines.

This is for a couple of reasons:

i) Site popularity: a combination of link popularity (i.e. the number of, and the quality of, other sites linking to yours), click-through rates (the number of users who clicked on your listing in the results), and post-click behaviour (such as the number of pages viewed before exit or return to the results). If the content is poor, or not what the user expected, a) no-one will recommend or link to the site and b) no-one will click through to the site, or if they do, they won't stick around once they get there.

ii) Competition: there are just SO MANY websites competing for search engine and directory real estate now that quality of content is the only way to differentiate them. It's the search engines' mission to identify the best sites for their users in terms of accuracy, relevance and usefulness - and they're getting pretty good at it. These days a site generally can't fool a search engine into thinking it's the best result for a given query - it actually has to be the best (see There Are No Shortcuts). This may seem like extra work for a website owner but in fact it's probably easier and cheaper than trying to develop a long-term strategy for fooling the engines with substandard content!

3. Poor keyword focus.

Keywords and phrases are an important factor in how search engines (and directories) rank web site results, because they indicate a site's relevance to a particular search term. Search engines look at the position, frequency and context of keywords on the page as well as in the html code (in particular the Title tag, also the meta description tag to some degree), to gauge relevance - and the more competing keywords there are on your page, the less relevant it is likely to be.

Note too that search engines look for the closest match to a search term when ranking, and that people nowadays tend to type in phrases rather than single words to find what they are looking for - therefore the secret to success will be in identifying what search terms your customers are actually using, and then optimising your web pages for these.

However, you need to strike an optimum balance between the incidence of your targeted keyphrase on the page (also known as 'keyword density') and the effectiveness of your copy - overdoing it could put off your customers, or worse, raise a flag to the search engines for keyword stuffing or over-optimisation.

Copywriting for both customers and search engines is something of an art, but the two most important things to remember are:

  • keyword focus: you must know exactly which keyphrase you are targeting and include it, perhaps with one or two variations, at intervals in your copy.
  • natural flow: the copy should not sound forced or unnatural and needs to utilise good sales copywriting techniques to convert visitors into customers.

4. The pages are dynamically generated.

Many larger sites these days use some kind of content management system (CMS) to dynamically create their pages from pieces of content stored in a database. Although many CMSs do produce perfectly search-engine friendly sites, there are still some that generate pages which are inaccessible to search engines. Some common problems include:

  • the URLs include a unique user session ID that search engines can't replicate and therefore can't follow
  • pages are entirely generated from data passed from a previous page and use the same URL every time (eg www.site.com/page.html)
  • pages are accessible under more than one URL on the same site and therefore excluded for duplication
  • pages all have the same Title and/or meta description tag and are therefore assigned low relevance scores or excluded for duplication.

It's usually straightforward to fix these problems, if you have a friendly developer who can modify the scripts for you!

5. The site uses frames.

Less of a problem than it used to be as designers become more search engine savvy and this format declines in popularity, the 'frames' format is renowned for causing problems with the search engines, primarily because they make it difficult for search engines to analyse a site's link structure and so determine a page's relative importance in the hierarchy.

In addition, framed pages are still indexed independently of each other, which can result in, for example, visitors arriving at pages without navigation frames or branding, leaving them unable to work out where they are or how to explore the site. Usability issues like this will affect both click-through and link popularity, thereby directly affecting search engine rankings.

Don't panic though - there are work-arounds which don't involve re-designing the entire site! For example:

i) Create a new entry/home page which doesn't use frames, but which is optimised with keywords, titles, headings and page copy to be search-engine friendly.

ii) Optimise the existing code using the 'NOFRAMES' tag, which all search engines can read and support.

iii) Submit each of your pages separately and include links on every page to return to the main frameset, or add Javascript to the code of each page which will restore the frameset automatically.

iv) Make sure your site has a site map that is linked to from every page.

Other potential problems

This list is just a general introduction to the various issues which need to be considered when optimising a web site for the search engines - ideally at the design stage.

You can gain a much more detailed knowledge of how search engines work by looking through some of the sites listed on the Resources page, but in the meantime here are a few additional reasons why your site may not be achieving good listings:

i) It may have been banned for 'spamming' - search engines and directories are very sensitive to attempts to subvert the index, and penalties can be severe. Keyword 'stuffing', incautious linking, cloaking, duplication, and redirection can all trigger spam alerts.

ii) It may have been rejected by the major directories for minor rule infringement (e.g. no contact details provided on site, title submitted not matching actual site title, inappropriate category selected for submission) - since many search engines use directory information as well as their own to provide results, exclusion from a directory can be disastrous.

iii) Poor directory listings. Even if you do manage to get listed in the major directories (Yahoo!, Looksmart and the Open Directory), you are unlikely to see any real benefit unless your listings include your key search terms. Since your listing is at the editor's discretion, it's advisable to research and draft your suggested description carefully before submitting.

iv) Password protection - search engines cannot read password-protected pages (obviously), so some of your best content may be going unindexed. You can sometimes contact a search engine or directory and offer to provide a password for indexing or reviewing purposes.

v) Image maps: some search engines cannot read links in an image map, so if your site uses one of these for navigation, you will need to supply text links too.

SEM EXPLAINED

Understanding the Basics

No Shortcuts!

Elements of Website Success

Top 5 SEO Design Mistakes