As a WebmasterWorld Moderator for the Google SEO section I am asked questions everyday and honestly most of them really upset me. I don’t mean a little upsetting, I mean like turning into a big green monster that starts saying “me hulk smash tings”. Why am I so worked up? Because the questions are simple and already answered? Nope. Just because someone answered that question last year does not mean the answer is still correct. There are fifteen year old SEO articles exposing the secret of using meta keywords tag to rank #1 and that is definitely not the case. What gets me upset is when people ask simple questions that they could easily test themselves and discover the truth first hand instead of relying on anonymous and unproven internet sources.
Back in 2012 I was asked if keywords in url structure outweighs backlinks in Google’s ranking formula. From my everyday experiences I knew that currently backlinks far outweigh url structures in most situations (there are exceptions to every rule). Instead of giving a simple answer I decided to run this simple test myself. I launched two brand new blogs for [goodroipow] a word that previously did not exist. Here are the URLs:
You’ll notice one site has the keyword in the domain and filename. The other has no keyword mention in the url. I placed identical text on both pages to remove the chance of the on-page content influencing the rankings. Then I linked to just the non-keyword url from WebmasterWorld (nofollow) and from Twitter (follow but redirected) and I waited … two minutes later Graywolf retweeted me ok now I got back to waiting for Googlebot to crawl these sites, then index them and then determine how the rankings will look.
The next morning I woke up and ran over to Google to see how well my test worked and I found nothing – (lesson #1 – Google is taking more than 1 day to index new sites even when linked from Twitter & WebmasterWorld with a few visitors). Google did return several suggestions for search terms that are spelled similarly but no thank you Google I did not make a typo.
The next day I woke and still found no results (lesson #2 – Google is taking more than 2 days to index new stuff).
On day 3 Google finally returned one result when I search for goodroipow – they returned a paper.li url which had scraped the keyword from Twitter. (lesson #3 – Established websites (even ones that are scraping) can easily outrank new sites). To make things even more interesting the paper.li page had already updated itself with new twitter messages and no longer had the keyword on its live page.
Finally on day four I checked the Google serp and found the winner – Non-keyword url with backlinks was ranking #1 (lesson #4 – Backlinks are generally more important than url structure). But I also noticed that Google didn’t even rank the keyword url in the serps. That might be because of it is duplicate content but Google did not display the typical message of “click here for similar webpages”. Maybe Google hasn’t visited the keyword blogspot post, so this test might not be finished just yet.
One Month Later
The exact page on blogspot still doesn’t rank, but my blogspot user profile mentioning the phrase in the post does. Graywolf’s Twitter page also ranks.
Three Months Later
The exact page on blogspot page still only ranks when you click the “repeat the search with the omitted results included” link, only the user profile page ranks in the natural SERP listings.
Now of course this simple test is not perfect. For example what if Google isn’t rewarding backlink but rather they are rewarding usage data and since I posted the non-keyword url in WebmasterWorld it had more visitors. That is a possibility and there are many variables that could be at play. It should be interesting to see how the serp changes after this blog post is published. The fifth lesson that I learned from this is that even when you know an answer to a question, running a simple test to reconfirm the answer can help lead you to learn new things that you didn’t know.
photo credit: Avengers Movie Website