301 Redirects Pass Along Corruption

By September 26, 2012Search Engine Optimization

I thought a bit about what I wanted to post today, and no current SEO-related subjects really interested me, except one. I wanted to briefly describe an experiment that we recently conducted here at Clix. I think the experiment was fairly controlled and the results are relatively applicable for future SEO.

Background Info

We had an old website that was affected by Google’s Penguin update in April. The site was extremely focused and abided by older on-page SEO techniques. As for off-page optimization, the site had about 40 backlinks (according to some of our resources), and few of them were questionable in nature.

The Problem

The site had been ranking very well on Google, and is ranking very well on Bing (even to this date). However, after the Penguin update, our site’s rankings began to slip. It was like watching a car crash. You knew what was happening, but you couldn’t do anything to stop it. Within a couple of weeks, the site had fallen off of Google completely.

The Experiment

The site was gone from Google. As a result, we bought a new domain and threw the files on it. That in and of itself was not the experiment. Rather, we wanted to see what would happen if we 301 redirected from the old domain, that had apparently violated quality guidelines, to the new domain. This would pass value from one domain to another, but would the association corrupt the new domain?

In order to ensure that the experiment did not contain any additional variables, we changed the robots.txt on the old domain to Disallow. We also notified Google Webmasters that the old domain was moving.

The Results

A few weeks passed and nothing happened. Google was stagnate and Bing was status quo (insert thumb twiddling). Then, one day… we saw a hint of life. The rankings for particular keyword phrases crept back into Google’s top 50, and our hypothesis that the site would return to the creme of the crop was coming to fruition during the next few weeks. However, we did have one other prediction or “gut” feeling that the new domain’s association with the previous domain would be the downfall… We were right. The site came back to a similar state to its previous domain, but subsequently fell off. Between August 9, 2012 and August 10, 2012 (exactly two months after we made the domain switch), the site went from ranking for 15 keyword phrases on Google to none.

Conclusion

The 301 redirects passed along the old domain’s value to the new domain with the unoriginal content, and it catalyzed the ranking process. However, the association with the old domain’s profile caused the site to completely fall off once again. Most likely associated with a panda refresh.

Next Steps

We are going to move the site to a brand new domain (again). We will not institute any 301 redirects. However, we will use the same content and on-page optimization.

I’ll let you know how it goes in another two months!

Leave a Reply