CMS vs. HTML for SEO - An Interesting Revelation
- SEO |
I was recently asked by a client whether WordPress or HTML was better for SEO. For the non-technical IM'ers, WordPress produces HTML, so they're effectively the same. Obviously WordPress makes SEO easier with all it's plugins and nice visual options. And, with the right amount of work, Google won't even know that your site is running WordPress. So is there any advantage to a pure HTML site from an SEO perspective?
Obviously there is a massive performance benefit with pure HTML (even if the content is database driven and you've got some PHP in there). Once you start adding plugins you can end up with a dramatically slower site. Google "punishes" slow sites under certain conditions, so that's not ideal. Practically speaking, no amount of clever caching on the WordPress side will solve the fact that it is a clunky CMS, albeit a highly extensible one. And, as we all know, there are shared hosting platforms that will suspend your account if your WP site is using too many processor cycles.
But there's more to this, and this is where it gets interesting. I was inspired by this post on AngelSEO to conduct an experiment. My team and I targeted four niches, two of which are low competition niches, and the other two are quite active from a competition perspective. For each niche we set up three sites: a WordPress site, a PHP/HTML site, and a Joomla site. We've been monitoring them over the past 8 weeks, and what we found was that all the sites initially reached roughly the same ranking. When we applied some offsite SEO (bought a couple of Fiverr social and link wheel gigs for each site) the sites went up in the rankings by roughly the same amount.
However...after 8 weeks we saw an interesting change. The WordPress sites started to fall in the rankings - and I mean drastically fall. We added new posts to the sites as an experiment, and they recovered somewhat, but they didn't get close to their initial ranking. The Joomla sites experienced a similar ranking decay, but not quite as dramatic as the WordPress sites. Their recovery after adding new posts was also not up to their original level. The HTML sites did really well, in that they did not lose their ranking in 8 weeks.
Conclusion: Thus, Google identifies CMS-driven sites, probably by their CSS classes, the "Generator" metadata, and so on. It identifies PHP/HTML sites as static. Google expects a CMS to have regular, updated, unique content. It does not expect the same from a regular site. Thus, if your CMS-driven sites are not topped up with regular updates, you will drop in rankings.
Workarounds: The WordPress geeks will know all about stripping WP down to it's bare essentials. During that process there's a lot of fudging you can do to make WordPress not look like WordPress. By renaming/moving the wp-admin/wp-content etc. directories you further obfuscate the actual HTML generated by WordPress. If you go to this amount of effort, you will save yourself from a rankings decay, but at that level of expertise and work you may as well write it yourself from scratch.
Obviously there is a massive performance benefit with pure HTML (even if the content is database driven and you've got some PHP in there). Once you start adding plugins you can end up with a dramatically slower site. Google "punishes" slow sites under certain conditions, so that's not ideal. Practically speaking, no amount of clever caching on the WordPress side will solve the fact that it is a clunky CMS, albeit a highly extensible one. And, as we all know, there are shared hosting platforms that will suspend your account if your WP site is using too many processor cycles.
But there's more to this, and this is where it gets interesting. I was inspired by this post on AngelSEO to conduct an experiment. My team and I targeted four niches, two of which are low competition niches, and the other two are quite active from a competition perspective. For each niche we set up three sites: a WordPress site, a PHP/HTML site, and a Joomla site. We've been monitoring them over the past 8 weeks, and what we found was that all the sites initially reached roughly the same ranking. When we applied some offsite SEO (bought a couple of Fiverr social and link wheel gigs for each site) the sites went up in the rankings by roughly the same amount.
However...after 8 weeks we saw an interesting change. The WordPress sites started to fall in the rankings - and I mean drastically fall. We added new posts to the sites as an experiment, and they recovered somewhat, but they didn't get close to their initial ranking. The Joomla sites experienced a similar ranking decay, but not quite as dramatic as the WordPress sites. Their recovery after adding new posts was also not up to their original level. The HTML sites did really well, in that they did not lose their ranking in 8 weeks.
Conclusion: Thus, Google identifies CMS-driven sites, probably by their CSS classes, the "Generator" metadata, and so on. It identifies PHP/HTML sites as static. Google expects a CMS to have regular, updated, unique content. It does not expect the same from a regular site. Thus, if your CMS-driven sites are not topped up with regular updates, you will drop in rankings.
Workarounds: The WordPress geeks will know all about stripping WP down to it's bare essentials. During that process there's a lot of fudging you can do to make WordPress not look like WordPress. By renaming/moving the wp-admin/wp-content etc. directories you further obfuscate the actual HTML generated by WordPress. If you go to this amount of effort, you will save yourself from a rankings decay, but at that level of expertise and work you may as well write it yourself from scratch.