google | The Internet Marketing Driver

The Magically Moving Meta Robots Tag And The Potential SEO Danger It Brings [Case Study]

September 19, 2018 By Glenn Gabe Leave a Comment

The Magical Meta Robots Tag

{Update: Google’s John Mueller replied to my question on Twitter about this situation. You can read John’s feedback here.}

I’ve been helping a client with a large-scale CMS migration recently and came across a very interesting, and potentially scary, situation. It’s a large and complex site with many moving parts, so there is a lot to process (both figuratively and literally).

The site has over one million pages indexed, and many more that get crawled based on the proper use of noindex in certain areas of the site. That was a big change for the site when I first started helping them a few years ago (after they got pummeled by a major algorithm update). Therefore, I’m always making sure those areas of the site are handled properly (as changes and updates go live on the site).

And now that a CMS migration was under way, I made sure to check on those areas of the site to make sure all was ok. It ends up they were ok… but they sort of weren’t either. Confused? Then read on.

The Magic Show Begins – Typical checks after a CMS migration
As I started crawling and checking the site after the migration went live, I noticed something very strange. The number of noindexed pages during my crawls was sometimes lower than what I expected. And while manually checking those pages, I noticed that my trusty AYIMA plugin sometimes went red (indicating the page was not indexable), but sometimes it didn’t. I found that odd and began to dig in further. Note, other plugins like the DeepCrawl indexability plugin also showed those urls were indexable. Again, strange…

DeepCrawl Chrome Plugin for Indexation

Static HTML Correct, Rendered HTML Not So Much…
When checking the areas of the site I mentioned earlier, and digging into the source code (the static html), I noticed that the meta robots tag was in the proper place and contained the correct directives. For example, it was placed in the head of the document and contained “noindex, follow”.

Meta robots tag from static html.

But we know that Google renders the page and that rendered html can be different than the static html. For example, JavaScript-based content will be rendered and can be indexed by Google. So, I started to check the rendered html via a number of tools like Google’s mobile-friendly test, the rich results testing tool, Chrome Dev Tools (using inspect element), and then various crawling tools like DeepCrawl, Screaming Frog, and Sitebulb. They have the ability to render the page, revealing what the final rendered html would reveal.

Note, the rich results testing tool will render the page via Googlebot desktop, while the mobile-friendly testing tool will render the page as Googlebot for Smartphones. Chrome Dev Tools will show you the rendered html via “inspect element” and then crawling tools like Screaming Frog, Sitebulb, and DeepCrawl have options for crawling with JavaScript rendering on. Therefore, it’s good to test across all of the tools to see what the results are across both mobile and desktop.

Render JavaScript option in Screaming Frog (first image) and then DeepCrawl (second image):

Render JavaScript option in Screaming Frog.

Render JavaScript option in DeepCrawl.

Wait, what just happened?
When I checked the rendered html across those tools, I witnessed the magically moving meta robots tag. The robots tag, and some other head elements, were now in the body of the document and not the head. I had to double and triple check this to make sure I was seeing it properly.

The amazing moving meta robots tag.

The meta robots tag went from line 105 in the source code to line 1,181 in the rendered version of the code. And again, it was NOT in the head of the document, where it needs to reside. It was now in the body! Here is a screenshot from Screaming Frog showing both the source html and the rendered html. Notice the line numbers next to each meta robots tag:

Screaming Frog static versus rendered html.

But Is It Safe?
When spot-checking urls containing the magically moving meta robots tag, they indeed were noindexed. Therefore, I’m assuming Google is using the meta robots tag from the static html and not in the rendered html (more about this in the next section). But will that always be the case? Again, on a site with millions of noindexed pages, and a history of getting hit by major algorithm updates, is it safe to believe that the magically moving meta robots tag won’t eventually cause problems? All good questions.

Checking the meta robots tag via the inspect url tool in GSC.

Google’s 2-phase approach for indexing JavaScript-based content
I passed all of this information along to my client and that was sent to the dev team. I’m still waiting to hear back about why this is happening, but needless to say, it could be very dangerous on a large-scale site with many moving parts. In addition, this is happening in sections of the site that absolutely should be noindexed and not released into the SEO wild.

Also, Google has explained that it uses a two-phase approach for indexing JavaScript-based content. The first phase will crawl and index the static html. And the second phase, which can take a few days or longer, will process and index the rendered html. And remember, the rendered html is where the meta robots tag ends up in the body of the document (for this case).

Here is John Mueller explaining that inserting canonical tags or meta robots tags via JavaScript can cause issues based on the initial phase of indexing versus the second phase (where it renders and indexes the JavaScript version):

So, theoretically for my client, Google would first see and process the proper meta robots tag in the static html, but then it would process and index the rendered html, which has the meta robots tag in the body of the document. If that’s the case, then Google may not decide to noindex the page in the long-term. I have to dig in further across those areas of the site to see how many pages are being noindexed versus indexed over the long-term (even though I’m hoping we just fix this problem sooner than later so we don’t need to worry about this!)

Again, with millions of pages currently being noindexed, I would NOT want to see a chunk of those pages suddenly get indexed. Remember, this site has felt the wrath of major Google algorithm updates in the past. It’s doing very well now, but the battle scars remain.

The CMS or the implementation?
Since my client just moved to a new CMS (a popular enterprise CMS), I naturally starting wondering if the problem was CMS-based or if it was due to the implementation of the CMS. Companies typically tailor the CMS implementation with various customized elements and code, and that can always cause unforeseen issues.

So, I decided to check several others sites running the CMS and they looked fine. The meta robots tag was on the same line in the html in the static and rendered versions (and it was located in the head of the document). Therefore, this looks to be an implementation problem. That’s often the case with CMS migrations. It’s not necessarily the CMS, it’s more about how the implementation goes. That’s where bugs can creep in and potentially cause big problems SEO-wise.

Screaming Frog static versus rendered html for another site running the same CMS.

Update: Feedback From Google’s John Mueller
John Mueller replied to my question about this situation today on Twitter. John explained that if the meta robots tag is in the right place in the static html, then that should be fine. He explained that it’s different when you need to do something specific after indexing and rendering (like adding hreflang via JavaScript). That’s great to know. I am still going to make sure the problem is fixed, since I’m not a fan of having magically moving meta robots tags, but it’s great to know this shouldn’t be a big problem. That makes sense, since I am seeing those pages accurately noindexed. Here’s John’s tweet:

John Mueller's tweet about the magically moving robots tag.

Quick Tips For Stopping Magically Moving Meta Robots Tags
Based on this case, I have some closing tips for SEOs and site owners that are going through a CMS migration (or for anyone that wants to make sure the meta robots tag, rel canonical, or other important SEO code is set up correctly).

Test your site via multiple tools that enable you to view the rendered html. That includes the mobile-friendly test (mobile), the rich results testing tool (desktop), Chrome Dev Tools (using inspect element), etc.
You can also test your site via multiple crawling tools that render JavaScript, including Screaming Frog, DeepCrawl, and Sitebulb. Then check the static html versus the rendered html. And make sure to check important reports to see what has changed compared to previous static crawls. For example, you might see noindexed pages drop off a cliff when you compare a static crawl to a JavaScript-rendered crawl. If so, why?
If it’s a third-party CMS (and not custom built by your company), then test other sites running the CMS. You might see consistent behavior across those others sites, which would signal a problem with the CMS itself. But if they look fine, then it’s probably your implementation of the CMS. That’s important to know.

Summary – For SEO, don’t rely on magically moving code
This was an interesting case for sure. Any time you see important directives magically moving around the html code, you definitely want to dig in and find out what’s going on. And then it’s important to quickly address the situation with the dev team. That’s especially the case for large-scale sites with many moving parts. I’ll try and post an update after my client surfaces the problem and applies a fix. Needless to say, I’m hoping the magic show ends soon. :)

Meet Newsguard, A Team Of Quality Raters For News Publishers – And Another Way To Check Site Trust, Credibility, and Transparency

August 30, 2018 By Glenn Gabe Leave a Comment

Newsguard for news publishers.

Based partly on the August 1 Google algorithm update, there’s been an extreme focus from site owners and SEOs on building and demonstrating E-A-T (expertise, authoritativeness, and trust). Now, the concept of E-A-T has been documented in Google’s quality rater guidelines (QRG) for years, but the recent update did seem to have an element that turned up the volume from an E-A-T perspective (and especially for sites focused on health and medical.) That’s why Barry Schwartz decided to name it the Medic Update.

Keep in mind that many different types of sites were impacted by the August 1 update and Google even confirmed they didn’t target sites that could be categorized as Your Money or Your Life (YMYL). Also, it’s important to know that you can’t build E-A-T quickly. I’ve mentioned that in my previous posts about major algorithm updates. You can, however, build strong E-A-T over time if you are doing the right things (building great content written by experts in your niche, using Social to reach a targeted audience, naturally building links from well-known sources, etc.)

Regarding the Quality Rater Guidelines (QRG), Google’s quality raters are asked to evaluate its algorithms, and E-A-T comes up many times in the 160-page guide. Note, raters cannot directly impact the rankings of any specific site, but their feedback is sent to the engineers, who then refine their algorithms. Therefore, they sure can impact rankings indirectly.

Based on what I explained above, site owners tend to have a few important questions:

Is my site and content trusted?
Are my writers perceived as experts in a niche?
Which specific factors on my site would cause users (and Google) to not trust the site?
Is there any way to see ratings for my site?

When helping companies that have been significantly impacted by a major Google algorithm update, it’s important to review the site overall, identify all problems that could be causing issues, and form a remediation plan for fixing those problems as quickly as possible. And helping companies determine perceived E-A-T is important. For example, do you have expert authors, are you deceiving users, does your content exhibit a high level of authoritativeness, etc.

The topic is extremely nuanced, since each niche category is different. For example, a medical site is different from a financial site, which is different from a dating website. And those sites are much different than coupon sites or lyrics websites. Again, there are many different types of sites on the web. And to address the last question from above about actually seeing ratings, there’s no way to see what raters thought of your site while evaluating Google’s algorithms.

That’s unless you’re a news publisher… Did I get your attention? :)

Well, you can’t see what Google’s quality raters think, but you can see view what Newsguard’s team of analysts think. And that can be a good proxy for understanding issues your site has regarding trust, credibility, and transparency.

Newsguard – Publicly evaluating trust and transparency of news publishers
Based on what I explained above, it’s great when you can get objective third parties to review a site (almost like your own version of Google’s quality raters). You can do that on your own, but there’s a good amount of work to do in order to run effective user studies. That’s why it’s great when you can find public sources of information from other services that can yield strong feedback about your site. That’s exactly what Newsguard is doing (for news publishers).

On its site, Newsguard explains that it aims to fight fake news, misinformation, and disinformation. I’ll explain more about Newsguard in the next section, but think of it as a way for the average person to quickly identify organizations that are adhering to important journalistic standards, and not participating in misleading the public.

Newsguard has a Chrome plugin that you can install that provides a quick way to see ratings for news publishes across Search and Social Media. You can also see ratings when visiting the sites in question. As soon as I read an article about Newsguard, I had to check it out. And after using it, I found it fascinating to receive objective feedback about the trustworthiness of various news organizations. It had me wishing there was a Newsguard for every category of website.

How It Works – Criteria and Nutrition Label
Newsguard has analysts that are trained journalists objectively rate news publishers based on nine criteria broken down into two major categories, credibility and transparency. Each criterion is weighted and worth a certain number of points. If a site receives 60 points or higher, it gets a green label (which is displayed in the browser, in the Search results, and across Social Media sites). If it falls below 60, it gets a red label. There are also other labels for user-generated content, satire and humor sites, etc.

For example, here’s what the plugin looks like in the browser window:

Newsguard in Chrome.

And here’s what it looks like in the Search results:

And here’s what it looks like across Social Media sites:

And you don’t just receive basic feedback (like the score). You can see a full “nutrition label” to learn more about why the site received that score, what the analysts listed as reasons for the scoring, etc. Newsguard explains they are completely transparent about why a site received a certain score. They also reach out to the sites being reviewed for more information and will publish that too. And if a score needs to be adjusted based on new information, Newsguard will document that information as well. You can read more about their corrections policy on the site.

For example, here’s an example of a nutrition label for CNN:

Newsguard nutrition label for CNN.

And you can read the full rating with information from analysts:

9 Factors Evaluated By Analysts
You can read more about the criteria on the Newsguard website, but they look at a number of factors that are important for any news publisher (and it’s important to note that there’s overlap with certain parts of Google’s quality rater guidelines).

For example, from a credibility standpoint, does the site:

Publish false content.
Gather and present information responsibly.
Correct or clarify errors.
Handle the difference between news and opinion responsibly.
Avoid deceptive headlines.

And from a transparency standpoint, does the site:

Disclose ownership and financing.
Clearly label advertising.
Reveal who is in charge, including conflicts of interest.
Provide information about content creators.

Again, each factor is scored and weighted, with 60 points being the difference between a green label and red label. You can see the weighting below:

Newsguard criteria and points system.

And for those of your reading this post involved in SEO, your QRG antennae might have gone up several times when reading those bullets points. For example, they mention deceptive advertising, bios for content creators, deceptive headlines, and more. Again, wouldn’t it be great to have this objective feedback as a starting point for every site on the web?

Examples of Site Ratings:
Below, I’ll provide a few examples so you can quickly see two sites on opposite ends of the spectrum. One receives very strong ratings, while the other sits below the 60-point threshold.

The New York Times (view the page on Newsguard.com):

Newsguard ratings for The New York Times.

A news publisher with 6 major categories of content, including YMYL topics, that received a red badge (less than 60 points):
Newsguard nutrition label for news publisher with less than 60 points.

Not listed? Submit your site.
As you can guess, the major limitation here is that this is a very manual process for Newsguard. That’s why I came across a number of news publishers that were in the process of being reviewed, or not even in the system yet. I don’t know how long it takes analysts to review each site, and then how long it takes for editors to review those ratings, and then for the co-CEOs to make a final review, but the process doesn’t seem quick. I’m not saying it should be quick, but I can see them getting swamped with requests very quickly. You can read more about the review process on the site.

So if you don’t see a rating for your news site, then feel free to submit it to Newsguard. I submitted a few dozen during my testing. It should be interesting to see how long it takes for those submissions to be reviewed. I’ll share more on Twitter when I see some of those submissions finally receive scores.

Here’s an example of a site in the process of being reviewed:

Site not reviewed yet.

A Note About Google’s Major Core Algorithm Updates – How do sites with failing grades fare during major updates?
While testing Newsguard across the web to check news publishers, I kept wondering how good of a proxy it was for Google’s quality rater guidelines (and how that would manifest itself during major algorithms updates).

For example, would failing sites also see negative movement during broad core updates? Let’s face it, publishing false information, providing deceptive advertising, using deceptive headlines, and more certainly are things that could cause algorithmic problems. But remember, Newsguard is only tackling a piece of the quality rater guidelines (and clearly not looking at everything Google is looking at). There are 160 pages in the guidelines and Newsguard focuses heavily on trust, credibility, and transparency.

But I was still interested in seeing the correlation. I’m not saying a failing Newsguard grade would mean you are going to tank during a major Google update, but it sure can provide some important feedback from Newsguard analysts about how your site is perceived. I plan to dig in more to see the connection between low ratings and algorithmic hits, but for now I just checked some news publishers with failing scores. I provided two examples below.

This news site failed and had previously been hammered by multiple major algorithm updates over the past two years. But, they did work hard on fixing many issues on the site recently and ended up surging during the August 1, 2018 update. So the ratings were before that happened. I’m recommending that the site contact Newsguard to provide additional information about their recent changes (and to see if they can get a fresh review).

A false positive from Newsguard.

And here’s a site that got hammered by the May 2017 update and has never come back. I haven’t analyzed this site heavily, but you can quickly see many serious issues when visiting the site. It’s pretty clear there are quality issues, UX barriers, and aggressive advertising problems throughout the site. A failing rating is justified here in my opinion, and they did get hammered by a major algorithm update.

Newsguard ratings for site hit by major algorithm update.

Summary – Free quality rater feedback from Newsguard
With many marketers now heavily interested in Google’s quality rater guidelines and E-A-T, I found it interesting to learn more about Newsguard, the criteria they use to rate sites, and to see how various news publishers fared with their scoring system.

At a minimum, news publishers should review their scores and look to improve, where possible. Again, it’s like seeing what Google’s quality raters think of a site without having access to the real quality raters. I believe having a public resource like this can be extremely valuable for site owners, journalists, and SEOs. I just wish there was a Newsguard for all sites. Hey, maybe there will be some day.

Analysis and Findings From The August 1, 2018 Google Algorithm Update – A Massive Core Ranking Update

August 9, 2018 By Glenn Gabe Leave a Comment

The August 1, 2018 Google Algorithm Update (The Medic Update)

On August 1, 2018 Google rolled out one of the biggest algorithm updates I’ve ever seen. It was huge and many sites saw significant volatility across categories and countries. The 8/1 update followed similar broad algorithm updates in March and April of 2018, which also caused significant volatility across the web. I wrote a 2-part series covering those spring updates, which you should definitely read. Google’s Danny Sullivan confirmed this was a broad core ranking update just like those updates.

Many sites across the web have been impacted by the August 1 update and Google estimated that it would take over a week to fully roll out. That seems to be correct, as I’ve seen a number of sites see much more impact Monday into Tuesday of this week. Also, Google confirmed yesterday that the update has fully rolled out, so it did take about a week.

I’ve been heavily analyzing the update since it rolled out and will cover a number of important topics in this post. Specifically, I’ll cover confirmed surges, including some of what those companies have completed after being negatively impacted in the past. I’ll cover fresh hits and what I’m seeing across those sites that were negatively impacted. And I’ll also cover important information from Googlers about this type of algorithm update (some of which has not been covered yet since the update rolled out).

So strap your battle helmets on. Let’s take a look at Battle Ground Google, The August 2018 Edition.

Significant AND Long-term + Relevance AND Quality
I help a lot of companies that have been negatively impacted by major algorithm updates. They are typically large-scale sites, but I do help some small to medium sized sites as well. The sites I’m helping have typically lost a significant amount of search visibility, rankings, and traffic during major updates, and sometimes have dropped over several updates (making the situation even worse).

As I’ve said in the past many times, broad core ranking updates take many factors into account. There’s never one smoking gun. Instead, there’s typically a battery of smoking guns. Actually, I tweeted that soon after this update rolled out based on what I was seeing across sites that were impacted. And it was awesome to see John Mueller retweet that! That makes sense, since John has communicated similar things a number of times during webmaster hangout videos.

John Mueller Retweet August 1, 2018 Algorithm Update

Google is evaluating relevance and quality at the site-level. And “quality” can mean a number of things. For example, content quality, user experience (UX), aggressive, disruptive, and deceptive, advertising, technical SEO problems that cause quality problems, and more. Therefore, it’s important to fully understand all of the problems with your site and then form a serious plan of attack for remediation. Remember, Google is on record explaining they want to see significant improvement in quality over the long-term.

You will not recover by putting band-aids on the situation. You will not recover in a few weeks, or a month or two. You must implement significant changes to improve your site and make sure they stick around for the long-term. Don’t roll back changes after a few weeks or a month or two. That may not be long enough for Google… Note, I’ll cover surges soon and what my clients worked on to improve their sites (and how long that took).

Relevance, Which *Includes* Quality Overall – From John Mueller
After the Brackets Update on March 7, John explained that the update had a lot to do with relevance, and the industry went wild with that comment. But after analyzing many sites and helping a number of sites recover during the March and April updates, I explained that it seemed quality was a part of the updates as well. It was easy to see that after analyzing many sites impacted in both March and April. I wrote about this in my 2-part series about the updates in case you want to read more about that.

Well, it was awesome to hear John explain that relevance INCLUDES quality overall! And that includes how you present your website, like sites that provide ads above the fold. Yep, he said ADS. :) He explained this in another webmaster hangout video. I cheered from my office, literally. Here’s the video:

That made complete sense based on what I was seeing across sites, so it was great to hear him clarify his original statement about relevance.

A note about Your Money or Your Life (YMYL) and E-A-T
Many people have explained the amount of YMYL sites impacted during this update, and specifically, health and medical sites. Actually, Barry Schwartz saw so many health sites impacted based on hearing from site owners that were negatively impacted that he called it the Medic Update.

I agree that many health and medical sites were impacted during this update, but they weren’t alone! I have a spreadsheet of over 210 sites impacted by the August 1 update and many are NOT YMYL sites. For example, there are entertainment sites, coupon sites, lyrics sites, e-commerce sites, sites containing online games, and more. So yes, it does look like Google made adjustments that would impact YMYL sites, but the update didn’t isolate those sites.

For example, here is a lyrics site that was impacted and a coupon site that was impacted:
A large-scale lyrics site that dropped heavily during the August 1 update:

A lyrics website that dropped during the 8/1/18 Google algorithm update.

A large-scale coupon site that surged during this update:

A coupon site that surged during the 8/1/18 Google algorithm update.

But YMYL sites are extremely important to Google. They are held to a higher standard, since they can “impact the future happiness, health, financial stability, or safety of users.” Therefore, it makes complete sense that Google would always look to enhance its core ranking algorithm with regard to E-A-T (expertise, authoritativeness, and trust), site reputation, author reputation, etc. It does look like Google refined how this works in the latest core ranking update on 8/1, but again, there were many sites outside of YMYL that were impacted. It’s important to know that.

With that out of the way, let’s dig into some surges.

Surges – The Long And Tough Road Back
The first client I’m going to cover is in the entertainment space. They have been impacted by a number of updates over the years and were destroyed by the June 2016 update (which was a huge algorithm update).

When they contacted me, they explained that they had tried to fix a number of things, but they thought they could be missing some important items. They were simply not recovering during subsequent algorithm updates. So, I began a long-term engagement to help them root out all quality problems and form a strong remediation plan.

My audit and crawl analysis uncovered many problems across the site. In total, I sent 84 pages of findings in Word during the entire engagement covering everything from low-quality content to aggressive and deceptive advertising to affiliate problems to technical SEO problems to performance issues and more. There were many, many things wrong with the site from a quality standpoint. It’s always interesting to go back and see how many findings were sent when analyzing a large-scale site with many quality problems. And it’s important for other sites to understand that too. There’s rarely a single issue causing massive drops during core ranking updates.

My client’s team worked hard to implement those changes, including the always-tough conversation with the monetization team. When you’re down, the knee-jerk reaction is to increase advertising to make up for the loss of traffic. But I was telling them to drastically cut down on the advertising situation, which was over-the-top from a disruptive and deceptive standpoint. The conversations were tough, but they did tone it down greatly.

They implemented a boatload of changes over four to five months, but remember, Google needs to process these changes over the long-term. So they kept driving forward while hoping they would see positive movement during subsequent algorithm updates. Remember what I said about “long-term” and “significant” from earlier?

The engagement ended in April and many items had been implemented at that point. I told them to keep driving forward, keep publishing killer content, keep enhancing the user experience, and to NOT fall back to old ways from an aggressive advertising standpoint.

Then August 1 arrived, and BOOM, they began to surge. It was awesome to see. Here is their search visibility trending from SEMrush.

An entertainment website surging during the core ranking update on August 1, 2018.

According to Google Analytics (which I cannot share a screenshot of due to an NDA), they are up over 48% since August 1 (and specific areas of the site are surging more than that). That accounts for more than 822K additional sessions from Google organic alone in just the past week. Clicks and impressions in GSC are off the charts. My client explained that this is the most traffic they have seen in about two years.

Another Surge – YMYL Health Site
The second client surge I’m going to cover is a large-scale health site that has seen a significant drop in rankings and traffic over the past 18 months. They approached me after seeing multiple dips during major algorithm updates and wanted to surface more problems from a quality standpoint.

Similar to what I explained above, I dug in heavily and sent through 93 pages of findings during the first engagement and then 51 more pages through during a second engagement (for a total of 144 pages of findings). Yes, there were many problems with the site from a quality perspective. And the site contains so many different areas of content that I was surfacing more and more problems with every twist and turn. Again, it’s important to know the amount of work the company did to improve the site. Their team was great and moved very quickly to make changes. And they continue to make changes to this day, but many of the significant changes were completed months ago.

Similar to my example from earlier, they worked on a number of areas, including nuking low quality or thin content, fixing a very aggressive and deceptive advertising situation, improving the UX, fixing technical SEO problems across the site, and more.

And then on August 1, they began to surge across many key areas of the site. Here is their search visibility trending from SEMrush:

A YMYL health site surging during the August 1, 2018 core ranking update.

As of today, various sections of the site have increased anywhere from 20% to 60% since August 1. Note, they still have a long way to go to get back to where they once were, but this was a significant move in the right direction. It was great to see.

And it’s also worth noting that some site sections decreased! So, on a large-scale site with many different sections, you can definitely see ups and downs during major algorithm updates by directory. That might be obvious to site owners of large-scale sites, but I’m not sure everyone knows that.

Those are just two examples of large-scale sites seeing improvement during the August 1 update, but the moral of the story was the sites surging have completed a boatload of work to improve their respective sites. They didn’t find one smoking gun. Instead, they found a battery of smoking guns and worked their hardest to root out those issues and improve their sites overall. In addition, they kept those changes in place and kept driving forward with improving the site. They didn’t revert back to old ways, which is hard to do especially from an advertising standpoint.

Fresh Hits And Digging Deep(er) With Sites Negatively Impacted
Once the update started rolling out, it was amazing to see how many sites were being negatively impacted. And some weren’t just being hit, they were obliterated. That includes some of the largest sites and brands across the web.

Here are some examples of serious negative impact. I have many examples of this happening, but I’ll just provide a few:

Massive drop during the 8/1 broad ranking update.

Another huge drop during the 8/1 Google algorithm update.

Yet another massive drop during the August 1, 2018 algorithm update.

When sites drop, it’s sometimes easy to take a quick look at them and try to figure out what’s wrong. But when you do that, it’s easy to miss serious problems that lie below the surface. Often, the only way to really see what’s going on is to dig deep, crawl the site, analyze the various areas across a site, etc. Remember, Google is on record explaining they take every page indexed into account when evaluating site quality. So, if you just look at five pages, you could be missing most of the SEO iceberg.

SEO problems deep in a site.

Therefore, in addition to reviewing top pages from sites that dropped during this update, I also crawled a subset of each site (10K pages per site). It’s not perfect, but that at least enabled me to see a bigger picture than just checking a few pages. And I’m glad I did. After crawling those sites, you could clearly see a number of problems across content quality, technical SEO, UX, aggressive advertising, and more.

And from a YMYL standpoint, you could clearly see sites ranking that probably shouldn’t have when it comes to important topics that can impact the “future happiness, health, financial stability, or safety of users”. Remember, the update DID NOT just target YMYL sites. Many different types of sites were impacted, but Google clearly was adding something new with regard to E-A-T, adjusting the threshold for YMYL sites, etc. Hard to say exactly what they did, but many were impacted.

Let’s go through some examples below. I’ll provide a few key findings per domain based on crawling and analyzing the site. Again, I didn’t crawl the entire site… I crawled a subset of each site (10K urls) to get a sample of what’s going on there. I can only imagine what I would find if I spent more time…

YMYL – Health Site
YMYL site dropping during the 8/1 update.
I surfaced hundreds of thin pages (650+) from just the crawl of 10K urls, so there are probably thousands like these pages on the site now if you take the entire site into account. Many of those thin pages contained a boatload of ads and very little main content (sometimes just a line or two of content, or just an image).

There were many thin user profiles indexed across the site. Some thin pages were filled with low-quality supplemental content, with very little main content.

The site also heavily uses popups, which can be extremely disruptive and annoying. I’ve covered disruptive and aggressive advertising many times while writing about major algorithm updates.

There were also a number of technical SEO problems including canonical issues.

And remember, this was only based on crawling 10K urls…

Reference Site

I surfaced over 6K near-empty or empty pages. It looks like a technical problem that is causing content quality problems. These pages are indexed now. And the pages are filled with ads. So you have empty pages (main content) but they are still displaying many ads. That’s not a great combination at all.

I then surfaced more content quality problems including pages containing just a few words of content with a number of large ads on the page. Again, not a great combo.

There were also annoying UX barriers that broke up the content. This was not done in an elegant way at all, and I found it extremely disruptive. I found myself scrolling to try and get to the rest of the main content.

From a SERP perspective, Google is returning a wider mix of results, and not just reference information that can be found on this site. Therefore, the site is battling quality problems, but it’s also dealing with what Google deems to be relevant based on query. So it seems Google was determining that users weren’t happy with the previous results (or that they were simply looking for something else).

Lyrics Site

I surfaced thousands of ultra-thin pages across the site, including blank pages containing ads. I also surfaced a number of pages returning a 503 header response code (by accident). This seemed like a technical glitch as the 503 is still in place now on those pages.

There’s a forum with many user profiles that are very low quality (and often contain very little main content).

There were also many empty artist pages that contained a number of ads. So there wasn’t any main content, yet many ads on the page. I saw this a number of times while analyzing the update. Remember, Google takes every page indexed into account when evaluating quality. Make sure your best content is indexed and nuke low quality or thin content. And if you can boost low quality content, that’s a great path as well.

Low quality or thin content and aggressive and disruptive advertising seemed to be the killer combination for this site. And to add insult to injury, the site still hasn’t moved to https.

YMYL – Health

I surfaced close to a thousand pages of low quality or thin content out of just a 10K url crawl. There was a serious lack of E-A-T for YMYL pages. The site’s authors do not have the expertise to write about sensitive medical topics.

In addition, there were many pages that contained just an autoplay video surrounded with ads and no other main content. Popups were used extensively across the site, also in combination with autoplay video ads. This was very annoying and disruptive. Again, not a great combination.

Then there were pages that should contain video, but the page was basically blank with ads surrounding the empty main content area. Sounds familiar, right? Sometimes there was just a placeholder for video and ads were all over the page.

I also saw a major relevance component at play here. Some rankings that dropped didn’t meet user expectations based on query. So, it seemed like the right move from Google’s standpoint to downrank those urls. For example, the site was tangentially covering a topic, but ranking well for the topic head term. I can’t imagine many users were thrilled with the results.

Recipes

I surfaced over 3K thin or low-quality pages out of a crawl of just 10K. And many had autoplay video ads (although without sound). Even though they have no sound, they are still incredibly annoying.

In addition, popups are used as well! So, the combination of autoplay video ads and popups on the same page had me banging my head against my monitor. And on mobile, there are giant ads pushing the content way down the page, so the entire viewport was filled with an ad. Note, I’ve often called Google’s Above The Fold (ATF) algorithm the beast that nobody talks about. Well, having ads that push main content out of the viewport is not a great idea, and it’s horrible for users.

In addition to many thin pages, I also surfaced empty pages that were indexed. And those empty pages contained ads. Again, sound familiar? There were also many pages with affiliate links weaved into the content. And there’s no disclosure on the page about which links were affiliate and which ones weren’t. I found that pretty deceptive from a user standpoint.

Side note about E-A-T, the site is well-known in the niche, has top writers with strong E-A-T. Therefore, it doesn’t seem like E-A-T was a core problem here. I believe it had more to do with the thin or low quality content mixed with poor user experience problems.

YMYL – Health

There were many UX barriers and technical problems across the site. For example, there was a module in the middle of many pages that just hung while hourglassing. Hard to tell if that was main content, supplemental content, or videos, but clearly not good to have on many pages.

There were MANY weaved ads throughout the content. Actually, it was every paragraph or two… I found that extremely disruptive. It’s ok to do that every now and then, but many weaved into a single article is horrible for users.

There were many amazon affiliate links weaved into the content as well. It seems like the site was focusing more on monetization than the user. And that’s always a dangerous situation. I’ve had many affiliate marketers reach out to me after getting smoked and they had that in place.

There were other user interface problems, especially on mobile. For example, some paragraphs did not align well with images. So there were only a few characters for each line on mobile. It was literally impossible to read. Imagine trying to read a paragraph 3 characters at a time…

Also, and this is clearly important on a YMYL site, there was NO author information. That’s possibly a worst-case scenario for a YMYL site writing about medical topics. So the site doesn’t exhibit much E-A-T, but the authors really don’t demonstrate strong E-A-T.

Many smoking guns, not just one
As you can see, even a relatively quick analysis enabled me to see a number of problems across sites that were negatively impacted. That’s why it’s important to thoroughly analyze a site and surface all potential problems. I’m sure if I really dug into those sites, I would surface many issues that could be contributing to their drop. These broad updates are called “broad” for a reason. :)

Follow The “Leader”, Flatliners, And The Danger In Doing Nothing.
When helping companies that have been negatively impacted by major algorithm updates, I often hear similar things. For example, some explain that they are just following what stronger players in the niche are doing and hoping they will surge like they have in the past. But unfortunately, they are following them blindly. For example, they have no idea if those are the right things to do, if those companies are surging in spite of having those things in place, etc.

The danger, of course, is that you can follow them right off an algorithmic cliff. Here’s a quick example. One of my newer clients always references a certain site with how they follow their lead, implement the same type of functionality, layout, content, etc. Once I started helping them, I explained the danger in doing that. And guess what? That site they follow just got hit pretty hard by the August 1 update. That site has done very well over the past year or so, but just got whacked by the August 1 update. Note, it’s NOT a YMYL site.

Following a site off an algorithmic cliff.

And here’s a site that refused to change after getting smoked by the February 7 algorithm update. Not only have they NOT increased at all. They have actually dropped more. Not good.

A site that never fixed SEO problems continues to drop.

The QRG Is Important, But Combine It With Traditional SEO Audits
Since 2014, I’ve been explaining the connection between Google’s quality rater guidelines and major algorithm updates. Yes, that’s when medieval Panda roamed the web. I think I’ve mentioned the power of the QRG in almost every blog post about major algorithm updates since then (including Panda, quality updates, the March and April 2018 updates, and more).

The overlap between the QRG and what I see in the field is uncanny. That’s why I highly recommend reading the QRG to see what Google deems high versus low quality, to understand how Google treats YMYL sites, to understand the importance of E-A-T, to understand the impact of aggressive, disruptive, and deceptive ads, and much more.

But it’s not the only thing you should do. I recommend combining the QRG with what you would typically do in a standard audit. Don’t ignore technical SEO, thin content, performance problems, and other things like that. Think about the site holistically and root out all potential problems. That’s the most powerful way to proceed in my opinion.

Google’s Danny Sullivan brought up the QRG when asked about the 8/1 update last week. So yes, definitely read the rater guidelines, internalize it, share it with your team, etc., but don’t forget about all of the other things involved with a site from an SEO standpoint (and how your users interact with your site).

Here’s Danny’s tweet:
Google's Danny Sullivan mentioning the Quality Rater Guidelines after the 8/1 update.

Next Steps For Site Owners:
If you’ve been negatively impacted by the August 1 algorithm update, then don’t sit and wait it out. John Mueller has explained in the past that if you’ve been impacted by one of these core updates, then take a step back and review your site overall from a quality perspective. And then try to significantly improve quality over the long-term. Don’t sit and do nothing. That probably won’t end well for you.

Therefore, here are some closing bullets (some might sound familiar if you’ve read my previous post about major algorithm updates):

Perform a thorough crawl analysis and audit of your site. Surface all potential quality problems and form a plan of attack for fixing them (and improving the site overall).
Read the Quality Rater Guidelines (QRG) and have everyone on your team that’s involved with your site read them too. Then review your site through the lens of the QRG (objectively). Then identify all potential problems and fix them.
Conduct user studies to understand how real people view your site. I’m not talking about having your mother, spouse, or close friends go through your site. I’m talking about objective third parties that can evaluate your site based on specific actions you want them to take. Identify all problems and barriers and remove them.
Analyze your Google Search Console (GSC) reporting, including the new index coverage reporting (which is killer). Find all technical SEO problems (combined with your crawl analysis and audit), and then fix them as quickly as possible.
Continue to publish killer content. Don’t sit and wait while you are fixing problems… Keep pumping out high quality content that users want to read (based on understanding your niche). The more high-quality content you can produce, the more you will exceed user expectations, which will often yield more sharing, which can result in more branded searches for your site and content, while also enabling you to build more (natural) links.
Keep building E-A-T. Note, you cannot quickly build E-A-T… you will need to do this over time. Remember, Google’s Gary Illyes explained when evaluating E-A-T, Google primarily looks at links and mentions from well-known sources. You cannot quickly build those links and mentions, but you can over the long-term by working hard and doing the right things. And make sure to showcase your expertise on your site and in your linked profiles (like Social). Jennifer Slegg just covered Google’s new creator reputation from the latest quality rater guidelines released in July. I recommend reading her post to get a solid understanding of what Google is doing on that front.

Summary – Google’s Mid-summer Classic Was Huge
The core algorithm update that rolled out on August 1, 2018 was massive and many sites across the web were impacted. Although there were a lot of health sites impacted, many others in non-YMYL categories were affected as well. If you have been negatively impacted by the 8/1 update, then it’s important to objectively analyze your site to find ways to improve. And remember, there’s never one smoking gun. There’s usually a battery of them. So go find them now. :)

Trapped In Google’s New Video Carousels – A Dangerous SERP Feature For Some Ecommerce Retailers [Case Study]

July 17, 2018 By Glenn Gabe Leave a Comment

Trapped in Google's new video carousels.

{Update 8/13/18: I started noticing a number of ecommerce retailers being released from video carousels when their category or product pages didn’t contain any video (or if the video just supported the main content on the page). The first signs of the change were on Sunday, 8/12/18 and I’m seeing the changes roll out more widely on Monday, 8/13/18. It’s still early, but I hope the changes stick. I included more information and screenshots below. And you can see clicks return via GSC screenshots, which I added on 8/27/18.}

————————-

In June of 2018, Google rolled out video carousels in the desktop SERPs (after having them live in the mobile SERPs for some time). Google had been testing the desktop feature on and off for a while, but finally pulled the trigger and fully rolled it out. And when I say “fully”, I mean the whole enchilada. More about that soon.

It wasn’t long before many people started noticing the change in the search results, which makes complete sense. The feature is hard to ignore based on the massive real estate the carousels take up and their visual nature.

Here’s an example of Google’s new desktop video carousel:

Google's New Video Carousels in the Desktop SERPs

Also, it didn’t take long to notice that Google had seriously turned up the dial on the amount of video carousels surfacing in the search results. For example, the carousels were triggering for many different types of queries, even if it didn’t make much sense to include video carousels for some of those queries.

Heck, they were even triggering for people’s names! I understand that video would make sense for some people, but definitely not for all. It seemed like Google was pushing the new video carousels too hard. Brodie Clark wrote a quick article covering this situation when the carousels first rolled out.

The video carousels are harmless… but are they? Just ask ecommerce retailers:
Some marketers checked out the new video carousels in the desktop SERPs and automatically thought the amazing real estate and visual nature of the carousel would be incredible for sites gaining those listings. But that’s simply not the case for ALL companies.

Ever since the carousels rolled out, Google’s John Mueller has been asked a number of questions about how to get out of the video carousels. Yes, that’s “get out” and not “get in”. Some questions have been asked via Twitter, while others have been asked in John’s webmaster hangout videos. I’ll cover more about Google’s recommendations soon.

Some ecommerce retailers have had a rough time with dealing with the video carousels. For example, some sites have found their way into those video carousels without even having video on their pages! In addition, it seems that many users are simply scrolling right past those carousels when searching for products. And that can lead to a serious drop in clicks and conversions.

My guess is that they think the carousels just contain YouTube videos or other video clips versus links to ecommerce pages. And that’s a terrible thing for ecommerce retailers that used to rank in the normal organic listings (the ten blue links). When you rank in the video carousels, you are removed from the organic listings. Not good.

Here are some examples of sites now ranking in video carousels that used to rank in the standard organic listings. The pages ranking in the carousel either don’t have video, or only provide video as a supporting feature. And these are ecommerce category pages, not video pages. I’ll provide some stats next.

SERP for “fidget spinner 1 dollar”. Alibaba ranks multiple times in the carousel:

Ecommerce category page on Alibaba.com containing one small video (bottom-right):

SERP for “jewelry sets for less”. Overstock category page ranking in the carousel:
Overstock ranking in the video carousels for an ecommerce category page.

Ecommerce page on Overstock.com without video:
Overstock page ranking in video carousel without video.

SERP for “roller shades”. Blinds.com ranking in video carousel multiple times:
Blinds.com ranking in the video carousels.

Ecommerce category page on Blinds.com without video:
Blinds.com ecommerce category pages ranking in the video carousel.

Show me the data! Clicks and CTR fall off a cliff while impressions remain high:
Again, it seems users are scrolling right by the video carousels when searching for products. How bad is the drop in clicks when comparing ranking in the standard organic listings versus the new video carousels? Pretty darn bad.

For example, I’ve included some screenshots from Google Search Console (GSC) of clicks and impressions from an ecommerce retailer that is experiencing this problem. Note, this a B2B ecommerce retailer, so those clicks can turn into big dollars. The average sale for this company is in the thousands of dollars and not tens of dollars.

Notice how impressions remain the same (or surge) while clicks drop off a cliff. This B2B ecommerce retailer began testing adding video schema markup (and video sitemaps) in late April and got caught up in the new video carousels by accident. They first started showing up in the mobile video carousels (which were already live in the SERPs), and then began showing up in the desktop video carousels when they launched in early June. I’ll explain more about what they are doing now soon.

A big drop in clicks while impressions remain strong for ecommerce category pages caught in the video carousel:
Click and impressions for page in a video carousel.

Clicks drop while impressions remain strong for page in video carousel.

Clicks drop and impressions surge for page in Google's new video carousel.

Clicks drop heavily while impressions remain strong for page mistakenly caught in a video carousel.

John Mueller’s Recommendations – The Current Fix
I mentioned earlier that a number of people have pinged John Mueller via Twitter and via his webmaster hangout videos about how to get out of the video carousels. John was quick to respond and had some recommendations for getting out of the carousel and back into the main SERPs.

John explained that you can disallow crawling of the video file and thumbnail image, or submit a video sitemap file with an expiration date set in the past.

Here is John’s tweet explaining how to get out of a video carousel:
Google's John Mueller explains how to get out of the video carousel.

And here is a video from John’s webmaster hangout explaining what you can do:

Again, the B2B ecommerce retailer I mentioned earlier was testing adding video schema markup (and video sitemaps) in late April and got caught up in the new video carousels by accident. After seeing the video carousel problem, they removed all video and video schema markup from the category pages, and then removed the listings from their video xml sitemaps to get removed from the carousels. Unfortunately, none of that worked.

They also tested blocking the video thumbnail image, which didn’t work either. Google ended up creating its own video thumbnails. Then out of desperation, they added video schema and video sitemaps back to add an expiration date set in the past. That didn’t work either. As of today, the site ranks in the video carousels for thousands of keywords. They used to rank in the standard organic listings until the end of April and early May when they started testing adding video. And then the desktop video carousels hit the scene in early June and the situation has gotten even worse.

From a business perspective, the site owner estimates they have lost $300K to $500K in sales since this started happening. Their category pages drive a large percentage of sales for the company, so this is really hurting them financially.

Danger Levels Rise: When all else fails, implement some hacks…
Remember, there is no video on the pages in question, so they had to add a video xml sitemap just to add the expiration date set in the past. Since nothing has worked, they decided to implement some hacks, which is always a dangerous thing to do. I never recommend implementing hacks SEO-wise since they can come back to bite you. That’s potentially happening here as well.

For example, the company started implementing more extreme changes to try and get their listings out of the video carousels. They started 301 redirecting to new urls to see if that would break the connection. That hasn’t seemed to work either (although they did see a few of those pages pop out of the carousels in the short-term).

Yes, they are literally redirecting important category pages that rank well for extremely competitive queries as a last resort. That’s crazy… They also started removing the videos from YouTube (those are the videos that Google was connecting with the site) or making them private. Remember, the videos aren’t on the site, just on YouTube. Those hacks didn’t work either.

Again, with an average sale in the thousands of dollars, and the category pages driving a high percentage of those sales, the company estimates they have lost approximately $300K to $500K since late April or early May when they started ranking in the video carousels.

Burning revenue as pages remain caught in Google's video carousel.

It just seems crazy that site owners need to jump through hoops while their revenue tanks due to something completely out of their control (and something they didn’t opt into). This ecommerce retailer, in particular, was blindsided in late April and early May when both traffic and revenue dropped off a cliff for important category pages.

A (Better) Fix Is Needed And Brainstorming Possible Solutions
Since video carousels are being populated algorithmically by Google, any change to that algorithm can take a while to hit the SERPs. Google’s John Mueller has explained that recently with regard to Top Stories (another SERP feature that’s gained a lot of attention). John explained that the Top Stories team needs to implement changes to the algo, then heavily test the changes, and then it could roll out down the line. That could take weeks, or even months before hitting the SERPs.

Here is Google’s John Mueller explaining algorithmic changes for Top Stories:

My guess is it’s the same way with video carousels. My recommendation to Google would be to first tone down the algorithm triggering the carousels. They should only trigger when Google is extremely confident that users want to see videos for a specific query. And if video carousels are triggered for a query, then Google should make sure that the pages being ranked in the carousel ACTUALLY contain video. As I explained earlier, some ecommerce retailers that don’t have video at all are being included in video carousels. And that’s killing their revenue.

Beyond the algorithmic part, there should be some way for sites to easily opt out of video carousels (either at the site-level or page-level). Whether that’s an option in Google Search Console (GSC), a tag that can be implemented via code, a directive delivered via the header response, or an addition to xml sitemaps, something needs to be done so sites can make sure they don’t become collateral damage with video carousels.

A mockup of GSC reporting showing a page being blocked from showing in video carousels:

GSC reporting a url is blocked from showing in a video carousel.

**Update 8/13/18: Many ecommerce retails are being released from the video carousels.**
Actually, a number of video carousels are simply being removed from the SERPs for some ecommerce queries. This is outstanding news for ecommerce retailers that were trapped in video carousels when they didn’t even have video on those pages (or if the video just supported the main content). I began noticing these changes on Sunday, 8/12/18, and then saw the changes roll out more widely on Monday, 8/13/18.

For example, I’ll use the queries from earlier in the post to demonstrate what I’m seeing now. I hope this change sticks. I know the B2B ecommerce retailer I documented in this post is ecstatic right now. But the company knows that anything can change in a heartbeat… I’ll post more updates soon. Stay tuned.

1. Blinds.com had a category page for roller shades (without video on it) trapped in a video carousel. It’s free now:
Roller shades - video carousel missing.

2. Overstock also had a category page without video stuck in a video carousel. Well, not anymore:

Jewelry sets for less - video carousels missing.

3. And Alibaba was released from the video carousel for this query, even though the carousel remained in the search results. But now it’s just populated with YouTube videos (which is accurate):

Clicks Return!
Regarding the e-commerce retailer I documented in this post, it’s been amazing to see their clicks return. When they were trapped in the video carousel, their impressions remained strong, but their clicks dropped off a cliff. Well, as soon as they were released from the carousels, those clicks came back.

Here are three different screenshots from Google Search Console (GSC) demonstrating the changes to key category pages that were once trapped in the carousel. Considering this retailer lost between $600K-800K in revenue due to the problem, they are pretty happy to be back in the standard organic results again.

Released from Google's video carousels. Clicks surge.

e-commerce retailer released from video carousels.

Clicks surge as e-commerce retailer is released from Google's video carousels.

It’s great to see Google make the right changes here. It took a bit of time, but as I explained earlier, that can happen when Google needs to implement changes, test those changes, etc.

**End Update**
——————————

Summary – Escaping video carousels is harder than you think
For ecommerce retailers mistakenly caught up in video carousels, this situation has been a horrorfest. Some have been blindsided by being included in the new video carousels. And then combine that with users that are scrolling right past those carousels when searching for products. And that’s after ranking well in the standard organic search listings where users were clicking their listings. As I documented above, revenue has dropped significantly for those important category pages.

My hope is that Google implements changes both algorithmically (refining the video carousel algorithm) while also providing ways for site owners to opt-out of video carousels (via GSC, meta tags, the header response, or xml sitemaps). If not, the bleeding will continue, sites will continue to implement hacks to try and get out of video carousels, and users will keep scrolling right past those carousels when searching for products. I’ll keep monitoring the situation and provide an update if I see any changes roll out. I hope that’s the case. Stay tuned.

How Sinister Mobile Popups Only Triggering For Uncookied Users (and Googlebot) Could Impact SEO, Usability, and Monetization

July 5, 2018 By Glenn Gabe Leave a Comment

Sinister mobile popups for uncookied users.

Users hate popups. But they really hate mobile popups. They are annoying, inhibit the user experience, and can cause serious user frustration. But mobile popups can also cause problems from an SEO standpoint.

First, Google has a mobile popup algorithm that can demote pages using an interstitial or popup when users visit from the search results. In my testing, it’s been a pretty weak algorithm, but it’s still there (and the strength could always be increased at any point).

Examples of annoying mobile popups. Just try to find the content…
Examples of mobile popups.

Second, John Mueller has explained that Google could see a popup as the main content on the page. That could be a serious problem, as Googlebot could simply index the content in the popup as your primary content for the url at hand. That can be problematic quality-wise, as the content will clearly be suboptimal (it’s the popup being indexed!)

Google's John Mueller about mobile popups.

Third, and more related to monetization, mobile popups violate the Better Ads Standards. That means Chrome could end up blocking ALL advertising on your site (if it’s a widespread problem). I wrote a post covering examples of Chrome blocking ads and you should read that article if you are wondering how it works and what it can look like.

List of mobile violations based on the Better Ads Standards.

Better Ads Standards for mobile popups.

So, we have a quadruple whammy when mobile popups are being employed (which includes user experience, SEO on several levels, and monetization). But the situation could be worse, and more sinister… One where a site owner has no idea that mobile popups are even being triggered for Googlebot. And maybe it’s across thousands, tens of thousands, or even more pages. Well, that’s exactly the situation I came across recently, and it’s what I’m covering today.

I think it’s important to cover this case since other sites may be experiencing the same situation. And if they are, they might be able to quickly implement changes to rectify the problem. This case study is a great example of how technical problems could cause big SEO problems. Read on.

The situation: Sinister mobile popups invisible to the ‘cookied’ eye
I’m helping a site now that has been hit a number of times over the years by major algorithm updates. It’s a large and complex site with many moving parts. There are millions of pages indexed, and the site has a number of issues that run the gamut of SEO problems, including quality, technical, advertising, UX, and more.

During my audit (which is still ongoing), I started digging into a specific area of the site and loaded a page on desktop. When visiting one of the pages, I noticed a strange lag when the content was loading. Some of the page content was being loaded via JavaScript and the lag I saw just seemed odd, so I decided to take a closer look.

Once the page fully loaded, it looked fine. The content was loaded properly, canonical tags were correct, images were fine, etc. But that lag really bothered me, almost like there was some glitch in the matrix.

Glitch in the Matrix.

So I decided to run a battery of tests on the page type, including checking page renders, measuring page speed, checking mobile-friendliness, as well as crawling a number of urls from that section. And what I found shocked me… and it was sinister.

Popups were only triggered on mobile, and only when cookies weren’t used
After checking the desktop pages quickly in Chrome, I started loading the pages via mobile. I first did this by opening a new tab and checking via Chrome dev tools. I was cookied by the site already (based on working on the audit), so nothing really stood out. The page looked fine.

But when I checked pages in this section incognito, the situation radically changed. I was hit with a mobile popup asking me to set my location. And it covered the entire screen. I tested several more urls and saw the exact same thing. It’s also important to note that I had never seen this popup before during the audit (so I had never set my location). I was cookied by the site, but not via information I set via this popup.

Mobile popup only for uncookied users and Googlebot.

If you just checked the pages manually (and while you were cookied), the pages would load fine. But for first-time users, or incognito without a previous cookie set, the site smacks you in the face with a giant mobile popup. And this was happening on thousands of pages across the site (and possibly more). I’ll cover more about that soon.

For my client (including the team of people I’m working with), they were already cookied, so they were seeing the pages load fine in the sections that were impacted (and not the popup). But remember, Googlebot doesn’t accept cookies, so it has been seeing the mobile popup for some time. Not good.

Here is John explaining more about Googlebot and cookies:

Fetch as Google + The *New* Mobile-friendly Test
Knowing that the mobile-friendly test renders content via Googlebot for Smartphones (and now will render JavaScript), I quickly tested some urls. And boom, the mobile popup was present.

Mobile-friendly test revealing sinister popup.

And then I hopped into Google Search Console (GSC) to test the urls using fetch as Google (to render the content on both desktop and mobile). As you can guess, the mobile popup was there too.

Fetch and Render in GSC revealing mobile popup.

This was clearly a big problem impacting many pages across the site. I immediately created a document detailing the situation and sent it to my client. Their team is working on the situation now and some fixes have already been put into place. But I’m hearing from my client that they are facing a serious game of SEO whack-a-mole. The problem is creeping up in more places than they expected. Remember, it’s a large and complex site.

Playing whack-a-mole with mobile popups.

How to avoid sinister surprises like this:
If you begin auditing a large-scale site, make sure you begin checking each page type across the various sections of the site. Manually view them, check them across desktop and mobile, and then use a number of tools to analyze the situation. And make sure you check the pages via incognito mode in Chrome (to avoid cookies delivering a user experience that’s different than what Googlebot will encounter – and for first-time users).

Check the mobile-friendly test for a quick look at how Googlebot for Smartphones renders the content (which will include rendering JavaScript-based content). Then check fetch as Google in Google Search Console (GSC) using both desktop and mobile user-agents. Make sure you don’t see anything crazy like what I saw in this situation.

From a crawler perspective, crawl the site via multiple tools to get a rounded view of how the pages are being crawled and rendered. For example, Screaming Frog, DeepCrawl, and Sitebulb can all render JavaScript content now (while both Screaming Frog and Sitebulb having the ability to provide rendered screenshots). Crawl sections of the site via those tools to check the content of the rendered pages.

For example, here is Sitebulb’s “live view” feature that enables you to see a snapshot of the rendered page across desktop, tablet, and mobile. Notice the mobile popup:

Sitebulb

Summary – Don’t let technical problems create bigger SEO problems
If you follow the approach listed above, you can avoid sinister problems that are invisible to the naked eye (or uncookied eye). And that can help on multiple levels, including improving the user experience and SEO. Don’t let Googlebot churn through mobile popups on your site versus your top-notch content. As I explained earlier in the post, there are multiple ways those popups can hurt you. Avoid the quadruple whammy if you can. Good luck.