An intensive analysis of the May 2020 Google Core Update by MHC

On May 4, 2020 Google announced that they were releasing a core quality update. So far, this is shaping up to be an update with widespread effects, possibly even more impactful than the August 1, 2018 Medic update. We had hoped that the main focus of this update would be to help demote sites that were publishing harmful advice relating to COVID-19. While we do feel that Google has reassessed whether they can trust the content of many sites, they appear to have done much much more with this core update.

In this article we will share our observations on what we are seeing with sites that have either improved or declined significantly with this core update. While we cannot say with certainty exactly what Google changed, our hope is that our observations on what type of content Google is rewarding will help you to make your site’s content the best of its kind.

Methodology

When we assess an update, each of the MHC team auditors spends time reviewing the changes in traffic patterns observed for each of our hundreds of current and past clients. Our client base consists of a wide variety of sites that have come to us for advice on improving quality. We then review content that improved or declined significantly with the update. Each of our auditors have extensive knowledge of the Quality Raters’ Guidelines and Google’s advice surrounding core updates. Every time a Google employee or document gives us a clue as to something that could be measured as a sign of high or low quality, we document it and our team discusses its potential value. We have done this for over eight years now, and have three Google docs, totalling over 1500 pages of transcribed Google videos, and more. Our goal in doing all of this is to have as much information as we can as to what it is that Google likes to reward.

With each core update, we take note of commonalities amongst pages that saw declines or gains. We apply our knowledge of what Google has told us on E-A-T and website quality, and then look at a great number of other factors including things like whether our clients have made changes that appear to have been rewarded, whether competitors are thriving, whether link building has happened on a large scale and more. We look at keyword rankings, seasonality and many other factors. 

We then have several team brainstorming sessions in which we come up with theories on what Google has changed, and then challenge our own theories. Our team also assesses the comment section of several forums where the update was discussed. We also review tweets and blog posts from people who have written their thoughts on the update.

We publish our early findings in the premium version of our newsletter.

Our goal is not to figure out Google, but rather, to get more and more clues on what it is that we can advise our clients to do in order for Google (and searchers) to consider their website the best of its kind. With this in mind, the rest of this article will discuss our observations so far on the May core update.

Summary of our observations

In this article, you will read our thoughts on the following changes that we feel were a part of the May 2020 Core update:

  • Relevancy: Google appears to be getting better at understanding what it is that a searcher is trying to find and surfacing sites that have the best answer for that query.
  • Expertise: Many of the articles that improved in rankings with this update contained an element of first hand expertise.
  • Authority: While authority is still important, many smaller websites that would not be seen as giant authorities in their verticals saw improvements and in some cases, were able to outrank very authoritative sites. 
  • Trustworthiness: Signals that Google can use to help determine E-A-T were likely reassessed as happens with most core updates.
  • Link quality: Many sites that saw declines had unnatural links, or links that could be considered “grey hat”, but essentially self made for SEO purposes. We believe that Google may be either putting less emphasis on links now that they can better understand content, or, they may be better able to understand which links are truly votes from other people who are recommending your content.

We’ll explain each of these in greater detail throughout this article.

What happens during a core Google update?

Last year, Google shared with us this blog post they called, “What webmasters should know about Google’s core updates.” It is absolutely packed with information to help us understand what we should be doing when faced with a traffic drop after a Google update. The blog post tells us that several times per year Google will “make significant, broad changes to [their] search algorithms”. They go on to say that the goal of these updates is to “deliver on [Google’s] mission to present relevant and authoritative content to searchers.”

While past core updates have seemed to focus heavily on trust, and protecting the safety of searchers, we feel that the main advancement Google made with the May 2020 core update is in understanding what it is the searcher is looking for, and presenting them with helpful results. In other words, Google got better at determining relevance.

What is relevancy?

In order to understand the changes that we believe have happened with Google algorithms, it is important to discuss what Google means when they tell us that core updates help them to present the most relevant results. Google has some very interesting documents to explain how search works. They break search down to several steps. The first is to understand the “meaning of your query” and the second is to understand the “relevance of webpages.”

Meaning of your query: “To return relevant results for your query, we first need to establish what information you’re looking forーthe intent behind your query. Understanding intent is fundamentally about understanding language, and is a critical aspect of Search.”

The document goes on to explain that Google attempts to determine “whether your query is seeking out fresh content”, whether you’re looking for reviews, pictures, or some other type of specific information. 

Understanding a user’s query is challenging. If a user searches for “best diamonds”, it is likely challenging for Google to understand whether you are looking for a website that sells diamond rings, an article that describes what to look for in a high end diamond stone cut, or content that discusses the best baseball diamonds in the major leagues. 

It is also interesting to note that the way people search has changed as well. John Mueller commented recently that the new generation of users search quite differently. “They’ll go to Google and ask a full question. We do keyword one, and two and add another if it doesn’t work. The younger generation will get older and this shift will happen in search.”

We believe that Google is using BERT to get much much better at understanding what the intent behind a user’s query is. When Google announced that they were using BERT, it was in a document called, “Understanding searches better than ever before.” Here are some interesting quotes from that article that help us to see what Google’s goals are when it comes to relevancy:

  • “At its core, Search is about understanding language. It’s our job to figure out what you’re searching for and surface helpful information from the web, no matter how you spell or combine the words in your query.”
  • “With the latest advancements from our research team in the science of language understanding–made possible by machine learning–we’re making a significant improvement to how we understand queries, representing the biggest leap forward in the past five years, and one of the biggest leaps forward in the history of Search.”

Relevance of Webpages – Google’s document on how search works says, “Next, algorithms analyze the content of webpages to assess whether the page contains information that might be relevant to what you are looking for.” It describes how, in the early days of search, Google would look for things like whether the keyword you are searching for actually exists on the page, saying that “if they appear in the headings or body of the text, the information is more likely to be relevant.” But they go on to say that “when you search for ‘dogs’, you likely don’t want a page with the word ‘dogs’ on it hundreds of times. With that in mind, algorithms assess if a page contains other relevant content beyond the keyword ‘dogs’ — such as pictures of dogs, videos, or even a list of breeds.”

Does this mean that in order to rank well, we simply need to add pictures, videos, lists, etc. to our content? This is great advice, and could be helpful for many pages, but we feel that with the language understanding capabilities of BERT, Google can go much deeper in determining whether a page is the most relevant result to display for a searcher’s query. 

We saw several examples of articles that started to rank well with this update that were actually quite short, didn’t contain images, and in some cases, didn’t have the scientific references we would like to see. But, in each case, these were actually really helpful pages that did well to answer specific questions that potential customers would have.

Google tells us in that document that they go beyond simply keyword matching, but also look at “aggregated and anonymized interaction data” combined with other signals to “help [their] machine-learned systems better estimate relevance.”

While we have several theories on how Google is working towards providing users with more relevant results, the methods that Google uses to accomplish this goal are likely far too complicated for us to speculate upon. Rather, what we need to know is the following:

 

Google has told us that their main goal is to better understand queries, and also to better understand which content is high quality. The key to improved rankings in 2020 and beyond is tied to truly producing content that is more helpful than what already exists.

 

Examples of websites that had high quality content improve in rankings with this update

When Google announces an update is about to happen, the MHC team waits anxiously to see what will happen with our client base. The majority of what we do is help businesses whose websites are not performing as well as expected on Google. While not all of our clients improved with the May core update, quite a few did. 

We will not share specific websites with you, for confidentiality reasons. Hopefully though, we will be able to share enough information about their success that you can glean some tips to help you improve as well.

Medical information site with nice improvements

This website is run by a medical doctor. After the June 3 core update, the site saw significant losses in traffic coming from Google.

medical info site with nice gains

You will notice in the chart above that the site saw some improvements with the January core update. We had hoped for a bigger jump up as the site owner had worked extensively on several elements of E-A-T including getting more truly authoritative mentions, and greatly improving on the quality of their content.

And then, on March 10, 2020, traffic dropped again following the WHO’s declaration of a worldwide pandemic. If your traffic declined, starting on March 10, you are not alone. Unless you are one of the businesses that grew due to the sales of supplies that are in high demand while we deal with COVID-19, you probably saw declines starting March 10 as well.

The site owner has told us that since the January core update happened, she really has not done much to improve her website, or her E-A-T. But still, she saw improvements in rankings across many pages. This leads us to believe that her May core update improvements were more likely to be due to Google making changes in how they assess her content quality, rather than because of increased efforts on her part towards improving E-A-T.

We looked at several pages that were seeing improvements in search traffic and it was quite obvious that these were incredibly helpful pages. One post that is performing well is an article on eyelid bumps. When we reviewed this page from the perspective of a potential searcher, we found it to be valuable for the following reasons:

  • The article starts off with a story about a family member who found a lump on his eyelid that turned out to be cancerous. 
  • Next, the author herself had an experience where she was uncertain whether a lump on her eyelid was serious or not. It turned out to be cancerous as well.
  • The article gives many statistics that are well referenced.
  • She describes many different types of eyelid cancers, in a way that is easy for a layperson to understand.
  • There is good use of headings to break up content and make it easy to digest.
  • The article has been kept updated as it was originally published almost nine years ago.
  • There is a huge comments section filled with people describing their own experiences with bumps on their eyelids.

It is interesting to note that the majority of the sites that are shown in the SERPs for queries that also surface this article, are giant authorities: WebMD, Cancer.net, and Healthline are still thriving and ranking well. 

In Google’s document on how they fight disinformation, they tell us that they want to, “where possible and relevant, elevate authoritative content from trusted sources.”

We believe that prior to the May core update, in order to rank well for queries that are YMYL (Your Money or Your Life), you had to be seen as a large authoritative site. We are not completely certain on this, but it appears that in many cases Google is now inserting a site or more into search results that is not a massive authority, but appears to be extremely helpful. 

It is interesting to note that in quite a few of these cases, the articles that have been able to push through to compete against authoritative sites are ones that clearly demonstrate real life expertise on a topic.

If I do searches to determine whether the bump on my eyelid is cancerous or not, I still can get great value from reading a site like Healthline, the Mayo Clinic, WebMD, etc. We believe that the authority that those sites have is enough for Google to consider them a trustworthy source for most, if not all, medical queries. But we also find it interesting that Google seems to be allowing a few smaller sites into the top results. It is one thing to learn the straight facts about cancer. It’s another though, to hear personal anecdotes that relate to mine.

While we cannot say exactly how Google determined that our client’s article was one that searchers would find helpful (although it’s probably related to BERT), we can hopefully agree that they have done a good job here.

Lead gen site with incredibly valuable content

Here is another client of ours that is doing well after this update.

lead gen site seeing improvements

This site is a lead gen site. They have worked quite hard to trim out thin content as we had identified a large number of pages that Google’s algorithms could consider doorway pages. Their articles are incredibly helpful. They have a calculator on their site that is by far the most helpful of its kind. While we can’t share with you what industry they are in, a similar comparison would be a mortgage calculator. While there are hundreds of calculators that allow you to punch in some numbers, imagine a mortgage calculator that walked you through several different steps and thoroughly explained each step so that you knew the best way to answer. We’re not talking about a page with a calculator that is followed by a 4000 word article that no one will ever read. Rather, each step on the page is explained extremely well for the reader.

There is not simply a calculator on the page, but also, loads of information that is not just there to pad the word count, but is likely actually read thoroughly by most readers. The content is well divided with good use of headings. The content has also been recently updated to contain pertinent new information about how the worldwide pandemic could affect the choices being made by people who are using this calculator.

The main point we are making here is that even though this site is not necessarily known as the most authoritative in their vertical, they truly do have the most helpful content. 

Supplement review site

This is a really challenging niche! If you write about supplements, there is so much potential for Google to consider your site as untrustworthy. We believe that with the June 3 core update, Google became very good at determining whether medical advice could potentially be harmful to searchers. 

If you run a medical review site and you’ve got pages and pages of reviews touting miracle diets or cures for cancer that are not recognized as valid by traditional physicians, your site probably has not performed well on Google for quite some time.

This site has been on a roller coaster of emotions with each Google update.

supplement review site gains

When reviewing pages that saw improvements with this update, it was instantly clear that they were incredibly helpful. Each medical article was reviewed by a doctor with appropriate medical expertise. What we found most interesting was how well this site did at telling both sides of a story. Rather than simply touting the potential benefits of each supplement they reviewed, they went above and beyond to discuss potential side effects and to discuss whether or not traditional physicians would take issue with this treatment. Their pages also had an excellent and long comments section with people giving their personal experiences.

While we can’t share the actual page with you, we felt that it did a good job in answering many of the questions that Google lays out in their blog post on core updates. We would urge you to ask the same regarding your content.

  • Does the content provide substantial value when compared to other pages in search results?
  • Is the content free from easily-verified factual errors?
  • Is this content written by an expert or enthusiast who demonstrably knows the topic well?
  • If you researched the site producing the content, would you come away with an impression that it is well-trusted or widely-recognized as an authority on its topic?
  • If the content draws on other sources, does it avoid simply copying or rewriting those sources and instead provide substantial additional value and originality?
  • Does the content provide insightful analysis or interesting information that is beyond obvious?
  • Does the content provide a substantial, complete or comprehensive description of the topic?
  • Does the content provide original information, reporting, research or analysis?

Glenn Gabe has a great article on the power of user studies in which he encourages us to get real user feedback to help surface quality issues on your website. If you have seen drops in traffic with this core update, getting unbiased feedback on the quality of your pages, and their usefulness as compared to competitors can be very helpful. We have found that often clients will tell us that they feel their pages are significantly more valuable than than their competitors’ pages. Yet, when we look at the pages, we can clearly see that this is not true. It can be very challenging to assess the quality of your own content.

Non-medical affiliate site

Many affiliate sites have seen declines with this update, but not all. Below, we will show a client of ours who runs an affiliate website that focuses on one specific niche. They do a good job of focusing on the parts of the niche that they know well. While this is not their subject matter, imagine a site that writes on many technical topics. Now imagine niching down to only cover one small part such as iPhones, or laptops. 

If you have an affiliate site that reviews products across many different categories, it can be challenging to be a subject matter expert for every single one of those topics. 

There are some sites that can still get away with this. For example, sites like PCMag or TechCrunch are massively authoritative affiliate sites that cover a variety of topics. According to Ahrefs data, they have both fared quite well with this update.

PCMag

PCmag gains

TechCrunch

techcrunch gains

But other smaller, less authoritative sites that write reviews (with affiliate links) on many different topics have not fared as well.

Here is the Google organic traffic for our client who runs an affiliate website based on a small niche that they know extremely well.

affiliate based site gains

This client has been working on improving many elements of E-A-T, so we cannot say with certainty that their success is due to the quality of their content. However, we think it is. When we looked at posts of theirs that improved as much as 70% in Google organic traffic, every one of them contained incredibly helpful reviews of the products they were selling. The most interesting thing was that they had actually used the products they reviewed. Their reviews are filled with first hand experience. They definitely contain more helpful information than any other site out there.

Now, we do not think that you can simply add the words, “We personally reviewed this product”, or something similar in order to trick Google that you are truly a subject matter expert and know what you are talking about. Rather, we feel that Google’s improved capabilities to determine what a person is searching for, combined with BERT enhanced understanding of content, and also their vast knowledge of entities across the web, has caused Google to consider showing searchers our client’s site, despite the fact that they are not known as a massive authority.

More on Expertise

We saw many examples of pages that improved with this update that demonstrated first hand real world expertise on a topic. 

For one of our medical clients, a post of theirs that describes symptoms of a heart attack jumped up into first page rankings with the May core update. Now, there are many authoritative sites that talk about heart attack symptoms. Our client is not one of these authorities.

post on heart attacks gains

The post is written by a medical doctor who shares the stories of several women who have experienced a heart attack. Each of the women explains their symptoms and their thoughts and fears surrounding the experience in incredible detail – much more detail than any other article we have found on the subject.

Imagine you are a woman who is experiencing chest pains. You are concerned that you may be having a heart attack, but you know that the symptoms of heart attacks in women are not always the same as they are with men. The information on sites like the Mayo Clinic, Healthline or WebMD is still extremely valid and those articles rank well. But now our client is ranking amongst many of those pages even though his site is not known as a massive medical authority site. We believe that Google can now recognize when a person may be interested in reading first hand experiences of women who have had heart attacks, and as such, is offering our client’s post as one of the first page results. 

In another case, we had a plastic surgeon client who had several articles about their particular specialty see very nice improvements.

surgeon client gains

Again, while this client is a plastic surgeon, he is not known as the absolute authority. Most people would not recognize his name. He has been working on many improvements on his site including filing a disavow to clean up some work from when he hired someone to build links in the past and other content improvements.

His content that is ranking well does a very good job of answering questions that people may have after having this kind of surgery. Again, you could go to a recognized authoritative medical site to get a list of side effects and that could be helpful. But, somehow Google has now recognized that his content is really helping people. His content contains first hand experience of the types of issues that he has seen his patients experience in real life. It is very helpful.

We do feel that with this update, Google got better at determining when an entity is truly demonstrating first hand expertise on a subject. It’s not the only raking factor though. We can see people saying, “But this site is outranking me when I clearly have more expertise!” But it’s something to consider. If you are now being outranked by sites where it is clear that they do have more expertise than you, regaining rankings could be challenging.

Aggregator sites seeing improvements

But wait! If expertise is being valued by Google, then why are aggregator sites winning some organic SERPs with local queries?

If I Google, “NYC Attorneys”, I see sites like Findlaw.com, Justia.com, Lawyers.law.cornell.edu and Bestlawyers.com. These are all aggregator sites and do not represent real life expertise.

If I Google, “Orlando Realtors” I see Realtor.com, OrlandoRealtors.com (a directory of realtors), and several other directory type sites before I see an actual realtor.

Google has definitely made some changes to the local search algorithm that gives us the “map results”. As we covered in our newsletter, the local results have been going through some sort of ranking changes since April 23, 2020. However, Google has told us in the past that local updates are separate from organic. What we are describing above, is a change to the organic results. 

It appears to be much harder for local businesses to surface as the top result organically when someone is searching for a lawyer, accountant, doctor, etc. 

We believe that this is related to Google getting better at determining the intent of a searcher’s query. 

While someone who is searching for “realtors in [my city]” eventually wants to end up connected with one particular real estate agent, we would guess that most people who do this search actually are trying to figure out which realtor to choose. There are probably some searchers who would say, “Ah, Google shows me this realtor’s website first, so they must be my best choice.” But most probably want to do research and figure out who it is that they want to hire. 

We believe that when Google is showing aggregator sites rather than businesses who actually have first hand expertise, it means that they have determined that the intent of the searcher was to investigate several different businesses. Conversely, in the examples given above where Google appears to be preferring to show pages from smaller, less authoritative websites, it is likely because Google was able to understand from the query that the searcher was looking for pages that discuss real life experience on a topic. 

There is likely more investigation that needs to happen here to figure out if there are other reasons why aggregator sites are dominating right now. It seems that these SERPS are still quite volatile. We will be looking at this further in the weeks to come. 

General E-A-T related signals were likely reassessed

If you have been negatively affected by a core update, even if you have made significant changes to your E-A-T or site quality, you generally need to go through another update before your site sees dramatic improvements. Google’s document on core updates says, “Broad core updates tend to happen every few months. Content that was impacted by one might not recover – assuming improvements have been made – until the next broad core update is released.”

With each core update we believe that Google reassesses several things. We believe that several of our past clients who saw declines in traffic with this update were negatively impacted because they published large amounts of content in the last few months that could be considered untrustworthy.

For example, one of our clients is a news publisher that saw approximately a 30% drop across many pages. When we looked at pages that declined we saw new articles that had several glaring issues that are outlined in Google’s Quality Raters’ Guidelines as things that could make an article low quality content. They appear to have published a large amount of content that touts “at home” cures and remedies. The content is written by someone with no medical expertise. The articles do not contain scientific references, and often contradict scientific consensus.

In another case, we have a client who saw big improvements with the September 2019 core update. They were negatively affected by the January core update, and again the most recent, May core update. While this client has been working hard on improving transparency surrounding their monetization methods, and greatly improving the trustworthiness of their medical articles, they also have a large component of their site that is driven by user generated content.

When we reviewed pages that had dropped with the May core update, many of them consisted of almost completely user generated content. As such, they had posts giving medical advice that were authored by anonymous users with no demonstrable medical expertise at all, and no external references. 

We suspect that once this client does a thorough review of their content and noindexes or removes this type of content, they will see improvements with the next core update.

Was link quality a component of this update?

At a recent Pubcon conference, Gary Illyes from Google told us that core updates are usually not about link quality. While we have seen many Google updates in the past that did appear to assess link quality, we do not recall seeing a core update that we felt was tied with link quality.

With that said, many of the sites that saw significant declines with the May core update had a link profile that concerned us. 

We believe that either of the following has possibly happened:

1) Google could be putting less emphasis on links now that they can better understand the relevancy of content. In the case of our plastic surgeon client who had his posts outlining his clients’ experiences after surgery improve with this update, the pages that improved did not have many links pointing to them. 

In 2014, Yandex, a popular search engine in Russia experimented with removing links from their algorithms. The experiment did not last long, and they eventually realized that using links to help determine authority was important. Matt Cutts, who was still with Google at the time, said that Google actually had tested dropping links from their algorithms and it made the results “much worse”. Links have consistently been one of the strongest signals that Google can use to determine which content should rank. If a particular article has people linking to it from all across the web, it’s usually a good article!

But what has happened over the years is that many SEOs have become fixated on link building. If you are an SEO who can consistently find ways to get people to want to link to your clients, you are skilled! However, we need to keep in mind why Google values links. Links are an important part of the algorithm because an article with links pointing to it is generally one that several people have recommended. Is this still true though, if those links are ones that you made yourself? 

If pages on your site have benefitted in the past from the power of self made links, you may be finding that they are not doing so well now. Let’s say that you have written an article about a particular topic in your vertical. Consider two scenarios and decide which of these link sets Google would want to value the most:

Case #1

  • Your content goes viral on social media and in the press. It’s so good that people start linking to it and mentioning it across the web.
  • You put out a press release and that gets even more attention. News outlets want to cover your story and are talking about your brand.
  • Other people who are writing stories on this topic, do research, read your article and link to it from within theirs.

Case #2

  • You hire a content writer to write several articles that cover your topics. Because you have relationships with several content publishers, they publish your content and link back to your site.
  • You create an infographic and get it published in several places that are known for publishing infographics which results in several links.
  • You write guest posts for several authoritative sites in your niche, each containing a link pointing back to your article.

Can you see the difference? In both cases, the content got links. In some situations, the links in Case #2 might even be from sites with higher PageRank or Moz Domain Authority. 

Even though we know that we would love to get links like described in Case #1, it is much easier to get them as we have described in Case #2. In the past, Google would often seem to count and reward both types of links. But we can clearly see that in Case #1 people are linking because your content is spectacular and in Case #2 sites are linking for SEO reasons.

If you were Google, which type of link would you want to count?

2) Google could be using BERT to better understand which links are true votes or recommendations of your content

It is possible that Google simply stopped counting links that were obviously self made mentions, links made for SEO, or links that just weren’t true recommendations of your content.

While we don’t know exactly what happened with links in this update, we can clearly see many incidences where content that used to rank well on the part of self made links no longer does. 

Should you disavow?

If you have content that is struggling to rank after this update, and you have built links to that content, you are probably trying to decide whether disavowing those links will help to restore your rankings. We have written extensively on this topic:

Disavowing in 2019 and beyond – the latest info on link auditing

Does Disavowing links work in 2019? Here are some case studies?

In those articles, you will read that John Mueller from Google has told us that if Google’s algorithms see enough unnatural links pointing to a website, they may actually choose to put less trust in your overall link profile. As such, if you have been link building on a large scale, it is possible that a thorough link audit and disavow can help to remove distrust that Google may have, and could help to restore the value you once had from truly naturally earned links pointing to your site.

However, this is not likely to completely restore rankings that have been lost with this update. If Google has gotten better at determining which content is valuable, they may be putting less emphasis on links. Similarly, if they got better at determining which links are legitimate votes, disavowing the ones that you made yourself is not going to make your content more valuable.

It is our belief that the May 2020 core update did not address link quality, but rather, made it so that links were a less important ranking factor than they used to be.

Conclusions

Update analysis has become quite challenging over the last few years! When a website sees declines in Google organic traffic in conjunction with an update, it is rare that we can find a single smoking gun to blame. Generally there are many quality issues that need to be addressed.

We feel that with this for update, Google got better at determining what it is that a search was looking for, and also which continent is the most relevant for them. In some cases, it seems that Google is allowing smaller, non authority sites to rank well for YMYL queries when it is clear that the site and its authors have real life expertise that would be valuable to the searcher. As with most core updates, Google reassessed many elements of E-A-T. We also believe that in some cases Google is putting less emphasis on links if those links are not true recommendations of your content. 

If you saw declines with the May core update, we would recommend the following:

  • Thoroughly read Google’s Quality Raters’ Guidelines. Look at each of the examples given to see if the things pointed out as a sign of high or low quality could apply to your site as well.
  • Do the same with Google’s blog post on core updates. Pay close attention to the bullet points that they share to help us determine whether our content is high quality.
  • Determine which pages of yours used to rank well, but no longer do. See who is now ranking well for your keywords. Thoroughly assess whether their content is better than yours. If it truly is not better, look at the types of sites that Google is displaying in the SERPs. 
  • Consider making use of user testing to get unbiased opinions on whether your content truly is the best of its kind.
  • Do all you can to demonstrate any real life expertise that you have that could be considered valuable by searchers. If you don’t have real life expertise, consider hiring an expert to write for you, or making use of well moderated user generated content so that your customers can share their real life experiences.
  • If you have built links in the past on a large scale, consider filing a disavow. Better yet, focus on finding ways to earn links to your amazing content rather than build them yourself. 

If you would like more information on Google updates, our theories, and staying up to date with SEO news, you can sign up for our newsletter. Each week we provide over 5000 words of SEO advice:


If you’re interested in hiring the MHC team to review your website and give you ideas on what you can do to help improve quality in Google’s eyes, you can find more information here on MHC’s site quality reviews.

Were you affected?

If your site was affected by the May core update, we would love to hear your thoughts. Please leave a comment!

Google update newsletter

Want an update when Google makes a big algorithm change or other announcement? Sign up here!

This is a weekly newsletter. We will never send spam. Unsubscribe at any time. Powered by ConvertKit
Share This Article
36 Comments
  1. Hi Marie,
    Thank you for this article. We have one client which has a negative effect of the May core update. Good point to have more focus for elements of E-A-T. We will try that direction! Keep up the good work

  2. Links are importants to calculate rank of a website, but a malicious person can ruin your ranking by using toxic links for your website and gogole sees nothing. Links can kill the SEO of a website even if its content is incredibly interesting. Links should not have as much importance in the ranking of a website: the most important is its content (text).

    • I should probably write an updated article on my thoughts on negative SEO via links. I haven’t seen a credible case of negative SEO link spam actually working in many years.

      John Mueller has told us that if a site has enough unnatural links pointing to it, their algorithms can reach a point where they stop counting *all* links pointing to the site. However, when it comes to link spam like comment blasting, forum posting and other easy to make unnatural links, we believe Google is REALLY good at just discounting those.

      The types of links that may possibly work for negative SEO are actually more likely to help a site significantly before hurting it. For example, we see manual actions these days for companies that have made hundreds or thousands of links via content marketing. At first, those links can work amazingly well to help rankings and then, if Google catches on, the site can receive a manual action or algorithmic demotion. But, not every time.

      Hopefully that makes sense. The point I’m trying to make is that I do not agree that a malicious person could kill the SEO of a website by building links against it.

  3. Hey Marie, I run a couple of different sites (Bony to Beastly for men, Bony to Bombshell for women) focused on helping underweight people build muscle and gain weight. It’s something I struggled with myself—I started clinically underweight and gained sixty pounds. My business partner is a respected strength coach who has also gained over sixty pounds (and has a health science degree and appropriate training/diet certifications). He’s not famous by any means, but he was hired to help our Canadian Olympic rugby team bulk up, he’s worked with professional football players, and he’s trained tons of university sports teams.

    Everything was going well until the June 3rd update, at which point our men’s site lost half it’s traffic and our women’s site lost virtually all of it. We’ve been trying our best to recover it, and some of it has been working, but this latest update has solved everything. Our traffic is better than ever before.

    What’s kind of interesting is that we almost gave up on our women’s site. With no traffic, it had no interest or income, and so we focused on our men’s site. But even with barely any changes or new posts, our women’s site has more traffic than ever before. We have these two sites with similar content and structure, and it was the inactive one that saw the massive boost. It has more traffic than our men’s site now. Weird, but nice.

    • Well this sounds like an interesting case. It would take me a while to give you a thorough answer as to what could be going on here. Have you read our article on the June 3 update? The majority of the sites that we saw affected by this update were ones that had a significant amount of content that was perhaps contrary to scientific consensus. The article I linked to above gives some ideas of the types of content that we feel was affected.

      • Yeah, I did. Trying to fix things after June 3 is how I found your site.

        We aren’t all that edgy. For the most part, we seek out the scientific consensus and try to align with it. What I found strange, though, was how it recovered by itself. While we were pouring all of our efforts into the website where the traffic got halved, eventually bringing it to 150% of its original traffic, the website that lost over 90% of its traffic now has 200% of its original traffic. It seems like whatever they were penalizing us for they gradually did a full reversal on—every core update has radically increased traffic ever since June 3. Very odd. But good!

  4. Tech, health, finance are some of the industries that are worst affected by this update.

  5. Really great info about May core update.

    Now I am clear what I need to change on my site too.
    Thanks for this detailed article.

  6. Great article. I was hit, medical niche, I have low EAT but articles have excellent quality. Clearly with this update it seems that the Brands have had an advantage over the quality content written with low authoritativeness.

  7. Hello Marie
    thanks for the nice article; I have a question, your client health website that has large user generated content, us using subdomain or subfolder for his blog?

    We have exactly this problem; our blog is in subfolder; do you think change and redirect it to subfolder is good?

    • For this particular client, their UGC is on their main domain, not a subdomain. But this is one strategy that can sometimes work. Sometimes, if you can separate out lower quality content such as unmoderated UGC to its own subdomain, Google can recognize that the quality issues there are mostly confined to that subdomain. Still best to find a way to moderate or only index the good stuff though!

  8. I agree with a couple of things with you like expertise. I think Google is picking well on that especially with the medical world to the extent that when I visit a doctor I know what to expect after reading about it.

    And everything goes just as expected; one doctor asked how I knew all about my medical condition but it was because I had read about it on Google. And I had to him tell that stuff is written by doctors like him

    But then Google is not getting so smart, I don’t know why people think that whenever there is a Google update it is always the webmasters at fault, and Google can never go wrong.

    I think Google became a little stupid with the May update and to me, it failed at relevancy

    Google failed to draw the line between authoritative sites and meeting search intent.

    I Googled some song lyrics and some of the search results include Google books just because the words in the lyrics do exist in some books.

    What exactly do the song lyrics have to do with Google books and how exactly does authoritativeness and expertise in song lyrics come in because all I care about is finding the right lyrics from whichever site has them and at least Google books don’t for a start?

    Before I used to find the song I am searching by just typing in a few words I heard and Google world give me the lyrics. In fact, I didn’t even need to add the word lyrics

    But right now I can’t find what I am exactly searching for because of authoritativeness. And I am not the only one, a couple of regular internet users that don’t know much about EAT, SEO, SERPs or webmasters let alone Google core updates are complaining due to results that don’t meet their intent after the update.
    Here is a good example of google failing to meet search intent after the May update according to regular internet users on Reddit https://www.reddit.com/r/NoStupidQuestions/comments/goiaa5/has_google_search_gotten_less_accurate_in_anyone/

    At the end of it all, the results are very poor because the site that actually has what I want is not an authority site and the end result is having search results filled with Pinterest, Google books, Reddit, Amazon, Facebook, Youtube, Google books, Quora and Instagram all of which are authority sites that don’t meet user intent at times because if a person wanted something from any of those sites they usually visit it directly, they don’t come to the search engine to look stuff they can directly find on any of those authority sites they already know

    Which means Google is failing to understand what I want and thinks BBC or Reddit or Facebook must be having what I want just because they are authority site with some of my query words on their sites and if BBC does not have what I want then Google will provide a Google book or youtube video because some words in my query do exist their which brings me to my next point

    BERT, as long as Google bases its search results mainly on keywords, BERT is a lot useless. I think Bing does well with BERT because to understand a query it doesn’t necessarily need to have certain keywords in it. You just need to understand the query naturally at least from a non-algorithmic perspective

    Actually, if you have an authority site and you copy and paste content from a weaker site as it is without even changing a single word, the authority site will rank higher for the same exact content from a weaker site because it has gotten trust from Google but you would not expect this to happen because it is plagiarism

    And for some people, their social media accounts rank higher than the actual sites to which the social media accounts belong

    So I think Google didn’t get any smarter and they made some mistakes this time even though some people want to believe it’s always their fault their sites tanked and Google is always right and perfect or most times Google knows they are right and it is the webmasters doing something wrong

    But since there are always minor updates, some of these may be fixed in the long run

    • I see what you’re saying. Personally I believe that Google’s philosophy when it comes to ranking YMYL queries is to prefer authoritative sites. That way, they have less likelihood to be presenting searchers with harmful info. But, with the May core update I feel that they got a little bit better at recognizing that some not so authoritative sites still are great to return on searches. I am betting that as they further develop abilities to understand language and the meaning behind a query, it will get better and better.

      Re lyric searches, I bet that some people who do type those words are looking for a book, or another reference. But, I am also guessing that over time Google’s algorithms will learn that the majority of people who type those words are looking for a song, and the SERPS will change again.

      • That’s true, that Google decided to stick with the big guns for uncertainty but again that shows Google is not so smart to tell what content is fake and accurate so one can literally manipulate the algorithm with backlinks

        And that’s why they couldn’t risk because the algorithm cannot tell which site could be accurate and good well as when they stick to the big site, even if the SERPs don’t meet the user intent at least the information is not detrimental

        But then again they shot themselves in the foot if I share my fake content that Google is trying to filter out on Facebook, Reddit or Pinterest, it will rank on the first page with a link to a Pinterest account or of Facebook, so much as they don’t show the filtered site in the google SERPs directly, Its indirectly showed via social networks links which even makes the user experience worse as the user will have to first visit the authoritative site first then go the desired site that was filtered out in google but appears on an authoritative social network

        And again today I read something, Google might not understand personalized queries anymore, and when it fails, it will always throw in Google books even when totally irrelevant to the query

        One query that interested me was “How to babysit a toddler when my wife is at work” Am sure you do understand the query but the algorithm isn’t so smart to understand which implies BERT is not working as intended,

        When I compared the same query with Bing, it wasn’t so smart either to understand that query but it at least provided 1 relevant result from wiki-how which Google couldn’t do as it is obsessed with its big sites and even worse Google books, I don’t think that query requires a book really

        When I tried the same query with duckduckgo, it was on point and it provided the desired results.

        Though I think there could be a chance some of the small sites filtered out by Google update could have the relevant results or the algorithm has a bug & google is too proud to admit so they throw the blame to webmasters

        And certainly, this is the new normal, better make sites authoritative

  9. Hello, this is all very right and interesting, but I see some sites that clearly miss EAT signals, and yet, they are getting a huge boost from the last update.
    for example, check this one:
    https://www.powerofpositivity.com/
    it doesn’t feature author names, has no ‘about us’ page, and doesn’t even have a contact form!
    how would you explain this?

    • This was really interesting to investigate. You’re right…this site is lacking in so many elements of EAT. My guess is that this site is actually getting some benefit from unnatural linking. It didn’t take me long to find in-article links that were likely made for SEO purposes. (https://www.worldwidehealth.com/demo/health-article-How-to-Know-if-You-Are-in-a-Healthy-Career.html and http://chrisryanfitness.com/exercises-for-workout-partners-to-do-together/).

      It looks like there was a big spike of new links just a few weeks ago. My bet is that soon their traffic will start to decline, or possibly with a future link related update. Those links are likely working to prop up rankings right now, but I don’t think the benefit will last for long though.

      I may come back to this and investigate further as this is definitely an example of a site that doesn’t appear to be the highest quality according to what Google has shared with us over the years.

      • They are not ranked for any YMYL content. They have really strong presence in social media for “new age” content. I cannot see how E.A.T is related to that.

        • As far as I know, google assesses a website for its overall content, not only for the ranked pages. So if a site has YMYL content, google still considers it a YMYL website. Am I wrong?

          And also, YMYL content doesn’t only include health or finance topics but also things like self-improvement and relationships.

          How is the presence in social media connected to SEO? Some sites have huge FB pages and zero organic traffic.

  10. I strongly believe that ggogle did not find a way to better understand content. I was on 2nd page of google for the “best synthesizer for beginner” and after May’s update i m at 9th page. The top 3 websites have less word count than mine, older updates, and the first ranking website recommends prodcut that is not even a synthesizer.

    • Sorry to hear that your site is not doing well. I think somehow in SEO we’ve made a decision that high quality = high word count and that’s not always true. I did a quick search and the sites ranking well for that query seem to be pretty relevant to the query. What matters most is whether this content is what is helpful to readers.

  11. So we put in a lot of money into fixing all of the issues with SelfHacked
    https://selfhacked.com/

    We literally fixed up every post to make sure each was compliant with medical consensus. Outside MDs have said our content is the best and most thorough medical content on supplements, etc…(after the changes.)

    These changes were finished at the end of Dec, around 5 months ago, and since then there were 2 CORE updates since then and we are where we started, even after adding a fair bit of content. We would be getting 30-40X the traffic if there were no medic updates restraining us.

    We fixed all technical issues as well at the time. We are exactly where we were.

    I’d challenge anyone to come up with a single post that isn’t authoritative or comprehensive or goes against the medical consensus on SelfHacked. It’s all medically reviewed and written by writers with credentials, with the proper hedging language. Much better credentials than other sites such as WebMD and healthline as well.

    We also took out all affiliate links from the site after some suggested that that might be the cause.

    The only sites doing well are the mega sites. I don’t think there is any way to improve a site that has already been hit with MEDIC updates.

    Google is not giving smaller sites a way to improve.

    • Wish I could look into your issues further Joe. It’s hard to add more than we’ve already suggested without doing a more thorough review.

    • Remove the adblocker warning Joe. Nobody will turn off ads the first time they visit your site, leave instantly and your rankings will drop.

    • A lot of citations in the few articles I read.

      126 citations: https://selfhacked.com/blog/fish-oil-top-22-science-based-health-benefits-of-fish-oil/

      Are you providing anything unique and beneficial that user’s may not be able to find elsewhere?

      Although the articles state they’ve been reviewed by medical experts, the language used is not that of a medical expert. The articles seem to be a directory “copy n pastes”, with no natural flow on the topic.

      There are 3 – 4 ads in each fold of the page on desktop, covering as much as 80% of the screen. Its often difficult to distinguish the main content from ads.

      I hope this helps a little, UX is quite important.

      Yianni

  12. Hello!

    I was reading through this and wondered if Google’s core update affects email deliverability. For example, before the core update was completed, we were inboxing in Gmail and getting 23-25% open rates at my company. In the days after the core update, our open rates were slashed in more-than-half. We went from 23-25% to 11-12%.

    When conducting spam tests across two different tools on two different campaigns, it seems these emails are now landing BACK in the Promotions tab. It’s not undeliverable, it’s just re-categorized yet again.

    I was hoping if we could have some light shed on that.

    Thank you!

  13. I s it a good SEO practice to delete old content and replace it with new content? I feel i need to improve on my content to achieve relevancy accepted by google.

  14. Thank you Marie for in-depth analysis! We were hit 25% down. And as you’ve mentioned there could be multiple reasons why – need to work on overall quality and E_A_T.

  15. “it appears that in many cases Google is now inserting a site or more into search results that is not a massive authority, but appears to be extremely helpful.”

    More disturbing is the fact that Google now inserts “simple” sites without any added vlaue (thin content like short product pages) and no backlinks (!) for into SERPs for competive keywords (and some are even ranking for months). In other words, you now see more “garbage” domains in SERPs than ever.

  16. Our blog post about “the best synthesizer for beginners” went from 2nd page to 9th after Google core update on May. The first ranking page on the same term, review a product that it is not even a synthesizer.
    We update old content and add new posts every week, while my competitors have really old reviews, weak content, and suggest products that are not produced or sold any more. Me and my team are experts on the field with many years of experience.

    Google core update on May, makes completly non-sense. We see less authorative sites, with weak copy-paste content of 600-700 words, and bought non-relative backlinks gain 5x traffic, while we lost 80% of orgnic trafic, and it keeps dropping.

    • Sorry to hear that Mike. It’s hard to say much without a url. A drop of that size generally means there is are significant quality issues with a site.

  17. My tinnitus forum has been one of the top 3 visited German language tinnitus forums for more than 10 years and always been on page 1 of Google for ‘tinnitus forum’ on Google.de. It is non-profit and completely ad-free. It doesn’t have any manual penalties nor any security issues in webmaster tools. Now after the recent Google core update in early May it dropped even out of the top 100 for the main keyword and overall traffic to about 1/5 of the previous level. Another tinnitus forum I know had the same fate. On the first page, half of the sites coming up for a search on Google.de now are in English, some of the rest are commercial forums requiring paid subscriptions. It doesn’t matter much for me personally as I have never gained anything from running my forum, but tinnitus sufferers looking for help and advice in forums will be hardly hit. Some tinnitus sufferers tend to be even suicidal, and without having anyone to turn to, they might take their own lifes. Google should have a serious look into their update and remedy this completely tragic result for my (and other) forums.

    • Sorry to hear that your site did not fare well with this update. Investigating this drop would take more time than I can commit right now, but here are a few quick thoughts. First, if I put myself in the shoes of a potential new searcher who has typed “tinnitus forum” and ended up on your home page, there is no information on that page to extol your E-A-T. I have no way of knowing who runs the forum and whether I can find trustworthy info on there. Tinnitus talk is not much better, but does give me a few more clues to show the searcher that they are a legitimate site and not just a hobby forum. It looks a *little* less dated than your site as well.

      If I go to Tinnitus-liga.de, I see more information to explain what kind of advice I can get and the credentials they have in regards to tinnitus.

      I’d start by sprucing up the look of the forum so that it does not look so dated, and then adding information to show why you are qualified to teach on this subject.

      There’s likely much more to the story though. If we were investigating this as a site review, we’d be looking at particular queries that dropped to see if we could note any other things that Google has described in the Quality Raters’ Guidelines as low quality. We’d also be looking to see if you have a decent number of mentions in places recognized as authoritative in terms of tinnitus too.

      • Hi Marie, Thanks for your reply.

        ‘Quality’ of a website is relative. If a suicidal German tinnitus sufferer not understanding much English wants urgently someone to communicate with, there is no point if Google points them to an English language forum, nor if they point them to what is actually a print magazine called ‘Tinnitus Forum’ (the page by tinnitus-liga.de) that is only available to paying members of the organization. Some people just want an uncomplicated way of getting help and advice without being asked for money (and even the tinnitustalk forum is asking for donations to support them). And my forum has been providing this for more than 10 years.

        From a technical point of view, my website passes the W3C validator, the other you were referring to fail with 48 and 15 errors/warnings respectively. And my website loads in under 1 second, while the other take about 3 seconds (as determined by webpagetest.org)

        On https://www.semrush.com/blog/google-may-2020-core-update/ you can asee that online communities in general have been most severely affected by this update. Considering that my forum has never been substantially affected by any Google update before, I would like to think that this was actually an unintended effect of the algorithm change, and hopefully this will be reversed with the next update.

        • I hear what you are saying. For years, for many SEOs the main purpose of our work would be to fix technical errors and make our sites look good to Googlebot. Just this year though, Google gave us Core Web Vitals in GSC and has told us that eventually user experience, as displayed in these metrics, will be a ranking factor.

          I am sure that you are correct in saying that your forum has excellent information. But, I really would recommend focusing on making the presentation of the website that a new user sees upon first arriving, more of one that inspires trust. Adding more information to extol E-A-T, modernizing the look, etc. could really go a long way.

  18. I guess google just reminded us that they own the SEO universe. I mean, everyone should follow their rules.
    Site content is always changing, and also, everyone must adapt.

Leave a Reply

Address

300 March Rd., Suite #301
Kanata, ON
Canada
Website: https://www.mariehaynes.com
Email: Contact the MHC team at help@mariehaynes.com

Privacy Policy
Referrals

HIS Web Marketing

Marie Haynes is the founder of HIS Web Marketing, formerly at www.HISWebMarketing.com. In 2015, she rebranded the company to Marie Haynes Consulting Inc.
Recommended Reading