Every year, Searchmetrics publishes a correlation study of SEO ranking factors. I am a big fan. I love correlative studies. At the time of writing this post, their 2015 study is not out yet, but here is a link to their 2014 study if interested. I had the chance to hear their findings at SMX West this year. In short, the takeaway from the study was not to worry about title tags anymore. I do not agree.
Once again, I LOVE correlative studies, especially when they are well done, like Searchmetrics'. However, correlation does not always mean causation. Looking at over 15,000 keywords using the top 30 results, they found no correlation between keywords in the title tag and rankings. Your mind immediately jumps to, "title tags no longer matter." This struck me as odd. So, I decided to do some testing. Now, keep in mind that I did this on a small scale and my only intention was to determine whether or not Google used our title tags as a ranking factor. What were the results? Also, very compelling.
So, I made an assumption. The study says that title tags do not correlate. Searchmetrics put a lot of weight into this and actually said not to focus on title tags. They did NOT say they were no longer a ranking factor or that they were not used (I just want to be clear on that). However, for the purpose of my little test I made the assumption that this is what was presumed. Simple enough. So, how does one do this? Well, if the only assumption here is that title tags no longer impact rankings, then I just either need to prove that correct or incorrect. So, I simply changed our title tag on our home page. It used to read, "Website Design, Search Engine Optimization | Southern WI Area". I changed it to read, "Award Winning Website Design and Search Engine Optimization". Clearly, if titles still affected rankings then this would have an impact since I removed mentions of our geography from the title tag and added "award winning".
Using comparative data in Webmaster tools and keywords we track, I was able to measure the shifts in our average ranking position (and actual rankings using daily metrics) for several days before the change, compared to several days after. Now, WMT is notorious for given us inaccurate data. So, to be sure, I had a look at large swaths of keywords that I categorized, as well as my other tracking data. I threw out anything with a shift of less than three places, just to make sure I was looking at the more significant ranking swings. In short, I was looking for massive changes in rankings for terms having to do with "award winning" or "Wisconsin". Well, I found some.
We began to rank for 5 key phrases related to award winning web design. The comparison confirmed that we did not rank for these prior at all. We also dropped heavily for every single phrase with "Wisconsin" in it and any geographic terms including Wisconsin-based cities. Interestingly enough, we also increased in rankings for a great deal of our non-geo specific terms (almost as if being Geo-specific was a detriment to national rankings). None of this information is definitive about the algorithm itself, but it does tell us, with reasonable likelihood, that the title tags had a significant impact on our rankings, and specifically to our relevance for terms in the title.
The one thing studies don't account for is experience. There could be a number of reasons why there was virtually no perceived correlation between the title tag and the key phrase that's in the title. What I can tell you, is that in this case, the correlation is deceptive. Titles tags still matter in SEO.
What are my next Steps?
SEO For Businesses Website Development Conversion Rate Optimization Marketing Automation Digital Success Stories