Tuesday, October 21, 2014

Current State of SEO & PizzaHut.com

How Pizza Hut might lead to a discovery about technical SEO

A Funny Thing Happened on the Way to the SERP


I know Google has come a long way in its 16 years. With the introduction of the Hummingbird algorithm and improvements and research into semantic search, Google's goal seems to be to answer our questions before we've asked them. Still, to me, there is something so delightfully unpredictable about human thinking that sometimes I get a little thrill when Google demonstrates that it is still, after all, just a machine.

I've been working in Internet Marketing for over ten years now. Most of that time I've spent heavily involved in search engine optimization, and much of that involved a special area known as Technical SEO. The work was inevitable born out of necessity in that all of the sites I seemed to end up working on suffered badly from a vast array of technical problems that either made it difficult or sometimes impossible for search engines to efficiently crawl and index the content of the sites' pages.

My first foray into technical SEO had to do with an odd thing I was seeing in the Google Analytics reports for one particular client.  I was seeing URLs listed in the landing pages reports that looked like a url, with a portion of the URL copied and re-pasted at the end. For example, something like this:


(this is just an example, it didn't happen on my blog, but I don't want to use the real company's URL out of respect) 
What was especially curious about this was that this URL not only reported, but it would render the page just fine. When I asked the eCommerce manager about it he said "Oh yes, I've seen those for years. I don't know where they are coming from. Googlebot is just stupid." Until I came along, this was the accepted explanation. Honestly, I had no idea why it was happening, but something told me that if it was Googlebot, it was happening because of the way the site was presenting data.

Thanks to George Andrews, Dr. Pete Meyers and the Moz Q & A, we discovered that the root of the problem was that we had chose to implement quasi-relative (they had incorrect syntax) rather than absolute URLs for our canonical tags. According to Dr. Pete: 

"Your home-page version isn't really either an absolute or relative URL - you really should have the 'http://' (protocol) in that URL. If you're being nitpicky, that's an improper URL, and Google could end up interpreting it as something like:  http://www.companyname.com/www.companyname.com "

That's exactly what happened. Once we fixed it, the odd links immediately stopped showing up in Google Analytics. So yes, it was brought on by how we had implemented our canonical tags, but I found it interesting that Googlebot, which on the one hand boasts that it is nearly a mind reader, doesn't have the sense to figure out something so seemingly simple.

This brings me to the purpose of this post. I talk to a lot of clients every day. Big businesses, small businesses, sites with tens of thousands of URLs and sites with less than 100 URLs. Because I'm not an algorithm-chaser, much of my work with businesses revolves around helping them achieve their business goals and how SEO can be one thing to help them get where they want to go. But still, every day, at least one client comes to me with a competitor who's participating in some "shady" tactics like participating in blog networks or paying for links who is out-ranking them by using those tactics.

The bottom line is that a lot of that old black hat SEO still works and it frustrates the hell out of real businesses. It also makes it really hard to sell them on #RCS.

This is what I say to clients who present me with that scenario:

"Participating in some of the old tactics like buying 10,000 'do-follow' links, setting up blog networks and involving yourself in search engine spam can still work. But it is a huge risk. I have seen businesses with revenue of $40 million per month get a manual penalty from Google and be completely removed from the index. Is that an email you want to wake up to tomorrow morning? If you are willing to take that kind of risk, what kind of resources, both financial and human, have you set aside to tackle the cleanup from such an event?

If it's just me, and it's just my blog, and that's my whole business, and I'm willing to take that risk it's one thing. But, if I'm PizzaHut.com, and I'm not only a major brand but a major employer with 10's of thousands of people and their families relying on that business for their livelihoods, taking that kind of risk is completely irresponsible."

Google has come a long way, certainly. Search experience is improving, machine learning is getting smarter, but it's still machine learning. There will always be loopholes and issues with how well Google interprets content both on and off the page. These will, as anyone in the SEO industry know, always be fast-moving targets. What works today, very well may not work tomorrow.

The fundamentals of SEO are based on making sure that search engines can access and correctly interpret the information on a website. Googlebot might be smarter than it was 16 years ago, but my 16-year-old born three weeks before Google, is still light years ahead of his coeval.