Lately I have been working with a client who runs a successful e-commerce brand here in Australia. E-commerce is very much the meat and potatoes of what a lot of modern day web-development toolkits are supposed to support so using ASP.NET 3.5 gave us quite a leg up in terms of development time. As my client nears completion of their project several of the loose ends are starting to get tied up – including the big question of how well is Google going to index the site content.
To do this they have engaged the professional services of an external agency to give advice on how to best structure the site. Like a lot of organisations offering Search Engine Optimization services, it is an adjunct to their existing offerings, in this case design work. One of the suggestions that has come through is that we change the structure of the URL space to include frequent search term keywords.
At first this strikes me as a fine suggestion, but after a little pondering I started to wonder why that would really improve things. First of all, the job of a search engine is to index the content of a site and then determine its relevancy when a user searches for a term that might have been present on that page. Google (as our token search engine) is going to use a number of factors to determine the content relevancy:
- The textual content in the page.
- The title of the page.
- The meta tags in the page.
- The number of authoritive inbound links to the page.
- The URL of the page.
There are probably more – but these are the ones I know about. Now if you look at that list the URL is certainly one of the things that it rates on but it surely isn’t the only one, and my argument is that the URL (domain name aside for a moment) is equal partners with the content of the page. I know a little bit about performance tuning and some of the theory applies here in that you need to know the relative benefits of performing each optimisation before you make it. In this case I don’t think Google has published their ranking algorithm so any optimisation, especially ones that could yield such a minimal result are a little premature.
SEO seems to have become a bit of a cult where the devices of workship are a few key practices and the belief that they hold some mysterious powers. Your SEO consultant will come in and rattle their bag of bones and make a few very expensive implementation suggestions. You’ll dutifully go off an implement them and sure enough, your web-site will turn up in Google and might even move a few points.
The problem is that you made so many other changes to the site in that time that it is impossible to know which one had an impact (URL, structure or content?). Without a baseline and an understanding of the algorithm you are just going on faith.