Everything you need to know about SEO, published jointly with Search Engine Land every Thursday.
Mobilegeddon Hits & Misses
A month into Mobilegeddon, columnist Bryson Meunier examines five sites hit hard by Google's mobile update, and five queries that still need work.
It has been more than a month since Google’s mobile update occurred, and many marketers are moving on to the next thing, having not seen much of an impact from the update formerly known as “Mobilegeddon.”
Not every site is so lucky, however. Looking at the data, it’s clear that many sites were not only hit by the mobile update, but hit hard.
While we don’t have traffic data for every site we don’t own, tools like SEMrush make it easy to see which mobile rankings were lost after April 21, and how that compares to desktop.
For example, looking at Moz.com, which is the example we predicted could lose up to 3% of total traffic back in March, we can see that both desktop and smartphone traffic were affected since the April 21 update.
If we compare April 26, five days after the mobile update, with May 25, we can see that it’s no longer ranking for 293, or 18%, of the top three listings it had before the mobile update, compared with 230, or 14%, of the top three keywords in desktop rankings.
Overall, it lost 2,524 total listings on mobile in the same period, compared with 1,996 for desktop. So rankings are sliding in general, but more so for mobile, according to this data.
Sites Hit Hard By Google’s 4/21 Mobile Update
Honestly, though, Moz didn’t have much to lose to begin with, as we explained in the March column. Many more brands seem to have taken a much bigger hit as a result of this update. SearchMetrics, Econsultancy and iCrossing have all mentioned some of these brands recently.
Looking at SEMrush data we start to see that some sites lost as much as 35% of their rankings for top three keywords in smartphone search results (compared with just 10% lost for desktop rankings).
Do a search on [free music downloads] and it’s very likely you’ll see much different results on smartphone than desktop, with apps results listed first, then Web results; 70% of the top 10 results are mobile-friendly, but the top listing on smartphone and desktop — Last.fm — is not.
In fact, no pages on Last.fm are mobile-friendly, according to SEMrush data. And this is likely why it seems to have been hit so hard by the mobile update. Last.fm lost more than 13,000 of its mobile top three rankings from April 26 to May 25, or 35% of the total. It lost desktop top three rankings as well, but only about 20% of the total.
And looking at three usable mobile-friendly sites right behind it in the search results that are just as relevant and authoritative, it’s a wonder it didn’t lose more than that.
Two of the domains on this list are top cars sites, which tells you that many sites in this industry decided to focus on their apps to the detriment of their mobile sites.
Unlike Last.fm, Cars.com is actually somewhat mobile-friendly, with 52% of its URLs being flagged as mobile-friendly by SEMrush. But the remaining URLs have hurt it, resulting in a loss of 29% of the top three smartphone keywords that it had been ranking for on April 26.
Contrast this with a loss of just 8% of total desktop keywords in the same period, and it looks to me like a -geddon of a mobile variety.
The Duke Boys may be using Autotrader on their smartphones in a recent television commercial, but it’s not the mobile website. Autotrader.com is just 5% mobile-friendly, according to SEMrush.
Autotrader only lost 16% of its top three mobile keyword rankings since April 26, but when you consider that it gained desktop keywords during the same period, this is clearly an instance of Google’s mobile update hitting hard.
When the smoke cleared, Autotrader.com lost more than 12,000 top three smartphone keywords ranking, while gaining more than 13,000 on desktop.
Like Cars.com, RankingsandReviews.com (owned by U.S. News & World Report) isn’t terrible when it comes to mobile-friendliness, with 55% of its pages being mobile-friendly, according to SEMrush.
But since has it lost twice as many top three keyword rankings on mobile as on desktop since April 26, and three times as many mobile keyword rankings than desktop when you look at all keywords, it’s clear that RankingsandReviews.com was rated poorly by the Google mobile update.
Bloomberg.com is 42% mobile-friendly — and another clear loser from the Google mobile update, according to SEMrush data.
While it only lost 674 top three keyword rankings on desktop since April 26, it lost more than 16,000 top three smartphone keyword rankings in the same period. When we look at all keywords, it’s lost 108,000 mobile keyword rankings, compared with 20,000 in desktop.
We’ve reached out to Last.fm, Autotrader and Bloomberg for comment and will update this post when we hear back.
Cars.com said that it saw a “modest decline in traffic to select pages,” but it has implemented some updates and believes it has since rebounded. U.S. News & World Report said RankingsandReviews.com’s mobile traffic was not affected by Google’s update, and it is optimizing all remaining sections of its websites for mobile devices.
Smartphone Search Results That Still Need Work
While I am seeing some sites devastated from this update, many search results haven’t changed at all, and mobile-unfriendly sites are ranking above mobile-friendly sites with similar or equal relevance and authority.
Knowing that Google is continuing to improve its mobile-friendly algorithms, I’m pointing these five out to help the search quality team in that endeavor. I’m sure there are more Mobilegeddon fails out there, and I would encourage you to share them in the comments.
1. [used cars]
– 37.6% of total search volume, or 169,132 smartphone searches monthly
– Suggested bid: $2.45
Is Autotrader.com really that much more relevant or authoritative for the query [used cars] than Edmunds.com, Kelly Blue Book, Craigslist, Cars.com or the other mobile-friendly sites directly below it? Chances are the only ones who would answer yes to that question do SEO for Autotrader.com.
And yet, what the searcher sees is one barely-usable-on-a-smartphone site followed by many authoritative and relevant listings that also work well on a phone.
Isn’t this what the Google mobile update was supposed to fix?
2. [online banking]
– 38.7% of total search volume, or 8,585 smartphone searches monthly
– Suggested bid: $9.80
If you were wondering, many searchers who use the phrase “online” don’t appear to think of it just as a desktop word. In this case almost 40% of searchers are using smartphones. Yet the first two listings for this query are not mobile-friendly.
Seven of the 10 listings on the first page are not mobile-friendly, so I can see how for this query it might be necessary to put the first listing, Bank of America, ahead of the third site that appears, PNC Bank. But do you need to give the first site two unfriendly listings on mobile?
Especially for a query like “online banking,” which is most likely being entered because the searcher intends to do banking online on his or her mobile device. If you give users a result that doesn’t work on their mobile device, they can’t complete their goal, and search fails them.
If only Google had a mobile-friendly algorithm that would fix this problem. Hmm…
3. [home value estimator]
– 27.6% of total search volume, or 6,124 smartphone searches monthly
– Suggested bid: $1.59
When you need a home value estimator, you need a home value estimator. Nowadays searchers expect tools like this to work regardless of the device they’re using. Yet the very first result looks like this on a smartphone:
You see two mobile-friendly listings that are not only relevant to and authoritative for the query, but actually work on a phone directly below this listing, so your guess is as good as mine as to why Mobilegeddon didn’t wipe this first result away.
4. [online poker]
– 18.8% of total search volume, or 5,088 smartphone searches monthly
– Suggested bid: $1.78
I mentioned this one in my SMX Advanced presentation on Tuesday. Google does put organic app results direct below the first Web listing for this query, but it wouldn’t necessarily have to if the Web results worked.
Looking to play online poker on your smartphone? Don’t click on the first listing, as you’ll go to a page on FullTilt.com that implores you to download software with an icon of a PC monitor next to it.
And when you click on that, nothing actually downloads to your smartphone, as the software is for a desktop computer and doesn’t work on mobile.
But you’ll get this frustrating message anyway:
This is the best Web result for “online poker” on a smartphone? I understand that many of the listings below it are either also desktop sites or articles about online poker, and that the only usable site on page 1 is blocking its mobile site with robots.txt, but there are usable sites like PartyPoker on page 2.
Why is it on page 2 for this query, when unusable sites hold the first listing? Someone from the search quality team may know, but as a searcher I have no idea.
5. [car batteries]
– 38.9% of total search volume, or 5,757 smartphone searches monthly
– Suggested bid: $4.29
Here’s a possible scenario for smartphone search that you’re unlikely to see on the desktop: I’m stuck on the side of the road because my battery is completely dead. I just called a tow truck, but now I need to know where I can get a new battery. So I enter the query [car batteries] with this in mind.
Google’s ads for the query seem to answer this search intent very well, with copy that’s tailored to a mobile device, and site links that send the searcher to a parts catalog and a click-to-call button for a nearby service center based on GPS.
And yet the first organic result? This:
Conspiracy theorists might point to brand bias or some other nefarious activity on Google’s part. I’m willing to give it the benefit of the doubt. But I don’t think many would argue that this is yet another instance where Google’s mobile-friendly update would have been very valuable, had it been working as advertised.
At SMX Advanced after the Mobilegeddon panel I asked Google’s Gary Illyes what to make of queries like these and whether they’re something that Google would like to fix eventually. Illyes didn’t see the queries, but I used FullTilt.com for [online poker] and Autotrader.com for [used cars] as examples.
Interestingly, he said that some queries that are considered navigational were not affected by the mobile update, and that it’s possible that these queries I’m seeing are considered navigational by Google and that’s why they appear still as the top-ranked site in spite of their lack of compliance with Mobilegeddon. I clarified, asking if he was saying that the query [used cars] could be considered navigational for Autotrader.com, and he affirmed it. (Navigational searches are those in which people use the search engine as a means of navigating to a particular site that they already have in mind.)
This is interesting and has implications beyond mobile SEO, but it still doesn’t make these sites possible to use on a smartphone, or in my mind a positive user experience for Google searchers.
Would they still be considered navigational, I wonder, if Google saw that searchers were pogosticking on smartphone but not on desktop? I would imagine that mobile usability might therefore change whether a site is truly navigational for a query that’s not obviously branded.
Still, great insight from Illyes on why this might still be happening, and I will be watching to see if the user experience improves on these queries and writing about it in a future column.
Frustrated by Mobilegeddon fails? Sound off in the comments below.
Some opinions expressed in this article may be those of a guest author and not necessarily Marketing Land. Staff authors are listed here.