Today's column is going to be a case study of a web site that made great progress simply due to basic code cleanup work. What's interesting about this case study is that the things we did aren't the first things you think of when you think of SEO, yet addressing these types of things can often make a huge difference.
The site is Work Coach Cafe. The site is a blog with regular posts on career advice and coaching.
We got involved in the site at the end of December 2010. Because that was the holiday season, there was a natural dip in traffic, but if we look back to November it was doing about 5,000 visitors per week. When we looked closely at the site we found some structural problems:
- The site performance was slow, with Google Webmaster Tools showing page load times taking up to 15 seconds (slower than 92 percent of sites).
- A crawl of the site by Xenu's Link Sleuth revealed up 49,662 broken links.
- There were 743 301 redirects in place from pages that were long gone, including double redirects in some cases.
We were concerned about site performance since Matt Cutts had identified performance as a ranking factopr in April of last year. It turned out that one of the big problems was that the site was on a low performance shared server.
We moved the site to a different server where it continued to share space with other web sites, but in a much more controlled environment. For the record, the cost of this new server (from Pair.com) was quite a bit more, but offered a much higher performance. As you can see here, performance improved quite a bit, with average page load times in the 5 second neighborhood:
The SEO impact of the broken links is a little harder to assess. However, given the sheer quantity, we were concerned that it could be affecting the perceived quality of the site.
When we investigated the broken links we found that many of these were related to template specific issues, such as references to missing CSS files or images. Through a lengthy process of fixing problems and then recrawling the site we worked until all broken links were eliminated.
We also dug into the 301 redirects. You may ask why we spent the time on this activity.
However, there is some loss of PageRank through a 301 redirect. I asked this question in an interview with Matt Cutts ast March. Here's an excerpt from the interview:
That's a good question, and I am not 100 percent sure about the answer. I can certainly see how there could be some loss of PageRank. I am not 100 percent sure whether the crawling and indexing team has implemented that sort of natural PageRank decay, so I will have to go and check on that specific case. (Note: in a follow on email, Matt confirmed that this is in fact the case. There is some loss of PR through a 301).
These were basically the three steps we took. No link building campaigns were undertaken. No keyword optimization was done. More broadly, no other SEO effort was invested in the site.
Results
So how did we do? The results suprised even us. Here is chart showing what they were:
In a very short period of time, traffic has scaled to about 7,000 visitors per week. This is roughly a 40% growth (and for the record about double where the site was on a year over year basis).
Summary
Much of this effort came down to simple web development hygiene. In developing web sites, errors have a tendency to accumulate. If you don't go back and address them, what begins as an infinitesimal impact begins to grow. Eventually, the cost becomes very, very real. Don't overlook this part of your SEO efforts. It is clearly worth the trouble.
