While we love Local SEO, a good portion of our clients are large non-local sites with a lot of technical SEO challenges. Today I’d like to talk about the dreaded Virtual Canonical Loop, an obscure technical issue that will absolutely kill your organic traffic, regardless of how big/small your site is. Here’s how it works.
You have a URL that canonicalizes to another, say https://www.site.com canonicalizes to http:www.site.com. So in the source of https://www.site.com you see:
<link rel=”canonical” href=”http://www.site.com”>
No big deal. Pretty standard.
Unfortunately, http://www.site.com 301 redirects to https://www.site.com. That in and of itself is no big deal. Canonicalizing http to https or vice-versa is a best practice to ward off duplication problems.
The problem is that when you canonicalize to a URL that 301 redirects back to the URL that is canonicalizing to it, Googlebot and Bingbot act like kid on a tilt-a-whirl after eating a giant corn dog, a slushee and some deep-fried oreos. That’s SEO blogger for “pukes all over you”. So you experience something like:
Besides getting covered with amusement park vomit, other maladies that Virtual Canonical Loops can cause include:
inability to rank for brand queries
massive reduction in organic rankings
a royal reaming out by your boss
loss of job
The fix is to either remove the canonical tag or turn off the redirect. I prefer removing the canonical.
If you experience a Virtual Canonical Loop for more than four hours, please consult your SEO consultant. Individual results may vary. Void where prohibited.
1) Google’s Mobile Testing Tool is Wrong/Incomplete - If this is true, it means that page speed, and other factors not accounted for by the tool, are factored into the mobile-friendly tag or mobile algo update.
2) Google Doesn’t Really Understand Mobile - Honestly, if a 5mb page that takes 12+ seconds to load is mobile friendly, then Flying Spaghetti Monster help us all, and our data plans. I feel like this deck by Jonathan Coleman offers a fantastic explanation about why site speed is critical for both SEO and user experience. Having improperly sized touch targets is frustrating, but not as frustrating as sitting around and waiting for a page to load when I need quick answers.
Have you run into any similar issues with the mobile testing tool? Let us know in the comments!
Update 4/11/15: According to a little birdie who spoke with Gary Iyles at the BrightonSEO conference, Gary “guaranteed” pack rankings won’t be affected. Guess we shall see…
Google’s pending Mobile-Friendly algo update has become the Apple Watch of SEO – I can’t look at my Twitter feed without a blur of “How to get ready for Mobilegeddon” posts streaming by, so I might as well jump on in. Jennifer Slegg over at the SEMPost reports that Gary Iyles, a Google Webmaster Trends Analyst, said at the Brighton SEO conference that Google’s Mobile Friendly Ranking Signal Will Not Affect Local Pack Listings: According to Illyes, if a local business is appearing in the Google Pack results, but does not have a mobile friendly site, their listing in the Pack results will NOT be impacted. Those Google My Business listings will remain. Now I have no clue how this mobile update will affect anything and I am generally wary of Google’s ability to know how an update will affect anything, particularly Local rankings, until it actually pushes the button and releases the Kraken. But Iyles’ statement strikes me as potentially a very Googley version of Truthiness. Here’s my tin-foil hat translation:
The Local Pack algorithm will not be changed. We are not touching it. Any time we touch that thing, all hell breaks loose.
Your GMB “listing” will not be impacted! It’s mobile-friendly after all. Pay no attention to the fact that I didn’t say anything about your GMB listing’s rankings in the Google Pack results. Those Google My Business listings will remain in our index - all your GMB dupes will too.
Don’t worry about the fact that 2014′s Pigeon update supposedly increased the influence of “Web Ranking Signals” on the Local Pack algorithm and that we said “we will be expanding our use of mobile-friendliness as a ranking signal”.
Even though we may be changing our “Web Ranking” algorithm, as I said, we are not touching the Local Pack algorithm. It’s still going to rely on “Web Ranking Signals” in the same way it has since Pigeon came out.
Update 4/11: Dr. Pete had the following interesting observation:
There’s a reason Google is suggesting 4/21 won’t impact local on mobile. Local on mobile is already different.
Today Google published “An update on doorway pages”. The post implies that Google will soon be updating its algorithm to weed out offending doorway pages from the SERPs:
Sites with large and well-established doorway campaigns might see a broad impact from this change.
While sites with spammy set ups such as multiple exact match domains that link to the same landing page are likely the targets (get ready for more Lawyer leads SEOs!), Google’s well-intended algorithm updates often go too far so brace yourselves for collateral damage, even if you think you have a good set up. Some potential issues:
Local directory sites with geo-SERPs that have the same content. For example, when one gas station serves two rural towns a site will have two pages – Gas Stations in City A & Gas Stations in City B – that have substantially the same content and lead to the same gas station profile. If these get hit, these sites will likely have to choose to only display results that are actually in the searched geo, which could have huge implications as to how data is displayed across the entire site.
Service area pages that don’t have unique content. We have long seen issues with sites that use cookie cutter text to create multiple service area pages. While plenty of sites still get by with it, I imagine this algorithm could finally snuff out this tactic. Get your content writers ready.
Franchise businesses – they have all of these issues in spades – particularly those who have microsites for every location. While cookie-cutter implementations make sense to easily create a lot of local sites, it’s possible Google will throw these local babies out with the bathwater.
I don’t want to be too alarmist. This update may be no big deal, but when I see Google use the phrase “broad impact”, I tend to get a bit paranoid.
And then there’s this:
For example, searchers might get a list of results that all go to the same site. So if a user clicks on one result, doesn’t like it, and then tries the next result in the search results page and is taken to that same site that they didn’t like, that’s a really frustrating experience.
At the beginning of every new engagement we ask clients for access to their Google Analytics, Google Webmaster Tools, Adwords data, Bing Webmaster Tools and any other analytics they have. 95% of the time, the client does not have Bing Webmaster Tools set up and 50% of the time they don’t want to go through the (very minor) hassle of setting it up. Most clients are pretty focused on Google of course, but in some cases Bing can still drive significant traffic. And ignoring Bing because it’s not Google can be a big mistake. Case in point:
On January 15th of this year a large news site client’s Bing referrals dropped by about 150,000 – 200,000 sessions/day. So they were losing about 6 million visits per month. That’s not chump change. We did all of the standard technical SEO diagnoses and the site was clean. Google traffic looked fine. So we were pretty sure the problem was not on our end. The only thing I could do was email Bing’s webmaster support which I fully expected to be a black hole. I got a canned “thanks for the submission” response and pretty much forgot about it. But you’ll be amazed at what happened next!:
click to enlarge
I ran into Duane Forrester, Bing’s Webmaster relations guy, at SMX West this week and asked him what the standard procedure was for sites that get whacked by Bing. He claimed that the Bing Webmaster Tools team takes these support inquiries seriously and the key is to be credible and patient. He stressed that you should not have multiple people submitting support tickets for the same site and described this partly as a resource issue – as in they can only spend so much time on any one site/ticket – and partly as a credibility issue – you should be confident that you have ruled out issues on your side and having ten people pile on complaints to Bing support isn’t going to make them move any faster. Of course the flip side is that asking site owners to be patient while Bing keeps you in the dark and figures out if they screwed something up is a pretty big ask. But there aren’t many other options.
The good news for this particular client is that it appears that the issue was on Bing’s side and they responded to our request. So set up your Bing Webmaster Tools. You’ll never know when you’ll need it.