Sunday 8 June 2008

Response to Mistakes in SEO

The task of optimising a website to get good rankings in search results is not easy or quick. There is no single magic trick that will guarantee frequent visitors to a site and convert large percentages of visitors into customers. You shouldn't worry too much about making mistakes but instead think of refinements - probably on a continuous basis.

There are at least 4 factors in selling a product online:-

  1. Having a product that has a market in the first place.

  2. Determining what people search for (on Google, Yahoo etc.) when looking for such a product.

  3. Making sure your website is a good match for the most frequent search terms - plus long tail terms.

  4. Making your website convince customers to buy your product


None of the points are particularly easy to solve and you won't be able to "fix" any of them in a single hit. You shouldn't rely on one, main search term reinforced by a domain name.

Long-tail search terms are very important. For example, on my main website the search terms are invoicing software, invoice software and billing software. I've done research and found that these are the main search terms in the US and UK for my type of product. However, these search terms account for only 15% of my visitors. The rest of my visitors come from long tail terms that may only be mentioned once in the whole website. I don't always deliberately put long tails into the text, I just write text, e.g. in the Support page or the Release History page that simply gives a good variety of words and phrases.

Due to the way search engines work, the people who find your website will already be interested in your product, or at least in the problem your product helps with. My feeling is that your website should then describe the benefits your specific product offers, i.e. why using your product will make handling security easier, quicker and cheaper for users. Be specific, e.g. if there is a lengthy manual process for enabling security then describe how your product is a better alternative.

The other thing to remember is that Google Pagerank isn't everything. Pagerank doesn't guarantee a page gets visitors. Monitor your web stats and see if any changes produce a rise in the number of visitors. Make refinements and see what effect they have.

The other thing to consider is whether you should rely entirely on search engines to get customers. Are there alternative methods of advertising your software? Some sites directly sell software to programmers or have programmers as target customers, e.g. www.codeproject.com or www.componentsource.com. Could you use sites like these in some way to advertise your software?

by ML

No comments:

Post a Comment