{"id":10963,"date":"2021-03-11T02:09:17","date_gmt":"2021-03-11T07:09:17","guid":{"rendered":"https:\/\/ipullrank.com\/?p=10963"},"modified":"2025-07-31T16:10:48","modified_gmt":"2025-07-31T20:10:48","slug":"why-mobile-only-indexing-will-cause-the-biggest-seo-shake-up-of-2021","status":"publish","type":"post","link":"https:\/\/ipullrank.com\/why-mobile-only-indexing-will-cause-the-biggest-seo-shake-up-of-2021","title":{"rendered":"Why Mobile-Only Indexing Will Cause the Biggest SEO Shake Up of 2021"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\n

As 2020 wound down, and I awaited the delivery of my Search Marketer of the Year award, I was honored to join the amazing Aleyda Solis and fantastic Jamie Indigo in contributing to the SEO chapter<\/a> of the Web Almanac<\/a>. I encourage you to check it out because we uncovered a lot of interesting things about the web by quantitatively reviewing millions of homepages.\u00a0<\/p>\n\n

The most compelling insight that I got from our analysis was how often there is a material difference in content served between the desktop and mobile user context. Directly from the analysis, we found a disparity between the median number of internal links being shown across the Desktop and Mobile pages.<\/p>\n\n

\"\"<\/figure>\n\n

From the above snapshot of the analysis, every percentile of homepages that we analyzed had fewer internal links shown for mobile than they did on the desktop experience. In preparation for this post, I asked Tony McCreath<\/a>, the Australian technical SEO (and hairdresser to the stars) that acted as our data analyst to pull a specific subset of the data where we had both the mobile and desktop page for direct comparison. What we found was somewhat alarming.<\/p>\n\n

In our dataset of 5.3 million URLs, we discovered that 30.31% of URLs served a different number of internal links for desktop devices than they did for mobile devices.<\/p>\n\n

\"\"<\/figure>\n\n

Scarier still, we found that only 16.29% of pages served the same word count between Mobile and Desktop. For the technical minded, the Word Count we considered is based on what is extractable from a rendered page using document.body.innerText.<\/p>\n\n

\"\"<\/figure>\n\n

We also took a look at indexability and found that 2.98% of URLs that self-canonical on Desktop do not do so on Mobile. There\u2019s also a subset of 0.51% of URLs where the opposite is true.\u00a0<\/p>\n\n

Keep in mind that our analysis only looks at homepages. So if we were to take the time to extrapolate this out, I believe that we are looking at a major issue across the web.<\/p>\n\n

Why Is this a Problem?<\/h2>\n\n

Google made an official shift to Mobile-First Indexing as of September of 2020<\/a>. Despite that, many of the sites I work on continue to show inconsistent distributions of verified <\/em>mobile and desktop bot crawl activity. Here are a couple of examples:<\/p>\n\n

For a large and popular e-commerce site, they are seeing 58.27% of their crawl coming from Googlebot Mobile.\u00a0<\/p>\n\n

\"\"<\/figure>\n\n

We also have a large site that is one part publisher and one part marketplace that was switched to MFI as of November 2018.<\/p>\n\n

\"\"<\/figure>\n\n

However, despite that indication in Google Search Console, their crawl activity still stands at 75% verified bots remaining Desktop-first.<\/p>\n\n

\"\"<\/figure>\n\n

So, what\u2019s the deal Google? Are we doing mobile-first indexing or naw? If we are, why is so much desktop crawling still happening? Or, is it that you\u2019re still crawling with desktop for URL identification because you know there are more links on desktop, but you\u2019re still using mobile rendering for the indexing process? Somebody get Paul Hahr on the line!<\/em><\/p>\n\n

An Issue is Brewing<\/h2>\n\n

Here\u2019s the thing, mobile-first indexing is not without its own set of problems. For a site that we recently worked with, their bot activity is nearly split down the middle. As a result, Google is continually going back and forth between which version of any given page they want to index.<\/p>\n\n

\"\"<\/figure>\n\n

Their desktop versions are incredibly robust and, as you may imagine, their mobile versions are quite pared down. The difference in word count ranges from a hundred to over eleven thousand words. Similar things can be said about the internal link counts.\u00a0<\/p>\n\n

In cases like this, Google has a Frankenstein representation of a website in their index with an incongruent link graph and subsets of the content being featured. In simpler terms, you may be presenting a link graph that looks like this on desktop:<\/p>\n\n

\"\"<\/figure>\n\n

But due to the inconsistencies in crawling and indexing, Google may index a link graph that looks like this for your site:<\/p>\n\n

\"\"<\/figure>\n\n

Now apply the same idea to body copy on those pages. While it may be obvious, it\u2019s worth stating that any given page\u2019s relevance will inherently change by having a different set of words on it.\u00a0<\/p>\n\n

Google representatives have indicated<\/a> in recent discussions that indexing will become mobile-only<\/em> as of March 2021. In other words, all those sites that are rendering less content on mobile are about to have a bad time.<\/p>\n\n

While much of the SEO ecosystem is preparing for the Core Web Vitals<\/a> component of the upcoming Page Experience update<\/a>, I find myself needing to sound the alarm about the content parity problem.\u00a0<\/p>\n\n

This is the real Mobilegeddon.<\/em><\/p>\n\n

SEO’s Context Problem<\/h2>\n\n

Here\u2019s what we know. Crawling and rendering are two separate processes. For Organic Search, this means there are four distinct contexts that Google may currently consider for any given web page.<\/p>\n\n