Google’s Core Updates: Prevention & Recovery | JP
There’s a lot of advice out there on how to recover from and even prevent your site from being hurt by one of Google’s core updates. Some of the advice is pretty good… some of it is a bit cliche. So let’s go down the rabbit hole a bit by analyzing the data on the core updates, some site-level patterns that the updates have produced, and what you might want to consider for your site as a result. Here’s a data-packed (yet holistic) look at what you can do to prevent ranking and traffic losses at the hands of a core update.
How to Keep Your Site Safe from Google’s Core Updates
I want to share my SMX West presentation with you because I feel that throwing my slides up on SlideShare doesn’t really do it justice. So today we’re going to untangle the drama that is the core updates, look at some of the data, how big the updates are, some site-level patterns, and some recommendations so that you don’t get clobbered by the core updates.
The first thing I want to go through is how big these updates are. Are we really blowing the core updates out of proportion? Are they really any more impactful than your average unconfirmed update? I want to make sure the SEO world is not blowing things out of proportion. Let’s take a look then at what the data says about how big the updates actually are. The answer? They’re really, freaking big.
This trend chart here shows you the top position on the SERP for the Health niche from the middle of 2015 all the way through the end of 2019. Spikes up are increases in stability and spikes down are decreases in stability or increases in volatility if that makes more sense to you. I’ve tried to highlight where the first core update begins. Do you see that first giant spike down? That’s the Medic update, which kind of makes sense why we call it the Medic update because again, this chart is for the Health niche’s top results on the SERP. There has never been before it nor after it such a heavy hit to that first position on the SERP for the Health niche. Of course, this is just a trend chart, which is meant to shock and awe you but it doesn’t really tell us exactly what’s going on. So what I want to do is compare some of the core updates to some of the unconfirmed updates, look at the levels of ranked volatility for both, and see which ones are really that much more impactful.
But first things first, I want to point out one of the interesting things which I am about to show you about the core updates is that they very much hit the top of the SERP (positions 1, 2, and 3) more so than the unconfirmed updates. And to me, part of the identity of a core update is how impactful it is to rankings at the very top of the SERP. So let’s have a look at rankings at the very top of the SERP. So these are volatility increases for the top three results on the SERP for the core updates including the Medic, March 2019, June 2019, September 2019, and January 2020 core updates versus four unconfirmed updates that were in February, April, July, and December of 2019. Let’s take a look. The Medic update saw a 30% increase in rank volatility. The March 2019 update saw a 21% increase, June 2019 – 31%, September 2019 – 26%, and January 2020 saw a really big 38% increase. Compare that to the February 2019 unconfirmed update which had just a 6% increase in rank volatility, April 2019 14%, and July 2019, which is a bit of an outlier and I’ll get to that in a second, saw a 23%, and lastly December 2019 saw a 17% increase.
I know, it’s too many numbers and I get that, but what you need to know is the core updates were 93% more volatile than the unconfirmed updates. The core updates had more rank volatility than the unconfirmed updates at the top three positions on the SERP. That’s pretty freaking crazy. Now if you noticed the July 2019 unconfirmed update saw rank volatility increases that almost matched some of the core updates and that’s because sometimes an unconfirmed update can be really impactful so don’t discount them just because Google didn’t confirm them.
Now, let’s have a look at rank volatility for the core updates versus the unconfirmed updates, but this time for the top 10 results. The Medic update saw a 90% increase in rank volatility, March 2019 saw 66%, June 2019 saw 62%, September 2019 saw 73%, and January 2020 saw 88%. Compare that to just 30% for the February unconfirmed update, 51% for April 2019, July 2019 saw a big 72% increase, and December 2019 saw a 55% increase. For the top 10 results overall, the core updates were 60% more volatile than the unconfirmed updates.
Since we’re talking about core updates, I feel like I have to talk about YMYL. What I did was I looked at the Travel niche and the Retail niche as my non-YMYL sites, and then I looked at the Health and Finance niche as YMYL sites. Now I know what you’re saying. You’re saying, “Well, retail, or shopping in specific, is part of the YMYL category that’s listed in the quality raters guidelines, except that Google has said that it treats commerce a little bit differently. It doesn’t apply E-A-T to them the same way it does to say the Health niche. And the data here, oddly enough, completely reflects that.
So I’m not going to run through all the data here, but basically what you’re looking at is every single update to the exclusion of the March 2019 update showed way more volatility for the Finance and the Health niche versus the Travel and the Retail niche. The YMYL niches (Health and Finance) were 66% more volatile than the non-YMYL niches for the core updates that I looked at. It’s pretty clear that Google is hitting the YMYL niches.
Just to recap where we are right now, core updates are fascinating in terms of how different they are at hitting the top of the SERP versus the levels that we see for the unconfirmed updates. YMYL does get hit harder. I have a whole blog post on why I think Google’s targeting YMYL in specific. And just because something’s not confirmed by Google doesn’t mean it’s not impactful.
Now that we know that the core updates are some big bad boys, let’s take a look at some of the patterns. One of the great things about working for JP is I have access to a ton of data which means I can see a lot of the sites that get hit and don’t get hit by the core updates. To preface, finding a pattern in any update, unconfirmed or core, can be a large pain in the rear end. That said, core updates are not as big of a pain in the rear when looking for site-level patterns, which is really interesting because Google has said, “There’s nothing to see here. Just write great content.” How do you reconcile that with so many patterns coming out? I don’t have time for it right now, but find me on Twitter or find me wherever and I’ll love to explain that to you. I don’t think Google’s lying because it’s true you should write good content to deal with the core updates, it’s just a question of what that means. But again, a topic for another time. Just know, there are a lot of really interesting patterns coming out of the core updates.
So what are those patterns? Instead of just telling you what I’ve seen, I’m going to walk you through it step by step. This is draxe.com and it is a health information site that got slammed by the Medic update and a bunch of other updates actually. They saw a bit of a reversal for the January 2020 core update, but not an entire reversal, not even close, and that’s because Dr. Axe actually scaled back on some of the things we’re going to show you here. Before they made the changes, you would see this on every single page. You would go to a page, the top of the article had a banner ad, scroll down the page had another ad, you finish the article you get a consolation prize of two banner ads, and of course, the related articles underneath is another banner ad, all for different magical powders he is selling.
I know what you’re thinking, core updates are about having too many ads on your page. No, I am not saying that. It is not about quantity, it’s about quality. It’s about what that implies about your site qualitatively which is a very different concept than the quantitative problem with ads.
I want to show you what I mean by that. Let’s take a look at spinehealth.com. It’s another one of these health information sites that got clobbered by the Medic update. It starts off with a series of informational content. You scroll down the page, you have your sponsored content box over here, there’s even a box where you can enter your address or zip code to enter to find a doctor. Of course, you get this page about this doctor and of course, it’s not free so that’s also sponsored. And at the very, very bottom of the page and on every page on this website, you get a box full of sponsored articles.
Let’s click on one of these articles. Here’s an article about why you should get spine surgery. Yes, it’s an ad about getting spine surgery. In fact, you can send them your MRI and they’ll see if you need spine surgery. I mean, what are the chances that you don’t need spine surgery? And it’s not a banner ad, it’s a whole full article outlining all the benefits of getting spine surgery. Now you have to wonder how much pain you really need to be in to think it’s a good idea to get spine surgery from this ad. This goes to point out that this whole thing is kind of sketchy. It gives me the heebie-jeebies and I think it gives Google the heebie-jeebies, which is sort of my point here. It’s not about the number of ads, it’s what the ads say about your site. What kind of site would have full-page article ads about getting spine surgery?
It comes down to identity. What is the identity of your site? Does your site have a conflict of identity? In this case, are these sites really about offering me really good health information or are they about having a forum to push their ads to buy whatever magical powder they’re selling?
I call this site profiling and it’s basically Google asking literally who are you? What is your site? Who is your site? What is your site’s identity? If you want to be technical the question is what’s your core intent profile? Do you have a well-defined identity? Are there conflicts in this identity? Are you really an informational site or are you an informational site that has an undercurrent of commerce to it.
Identity equals authority. That is true in life and now it’s true for websites. Imagine you walk into a doctor’s office, you speak to the receptionist and ask, “May I see the doctor?” and the receptionist answers back, “Would you like to see the doctor about getting spine surgery or would you like to get car insurance?” You ask in confusion, “You’re a spine doctor and you sell car insurance?” “Yes,” answers the receptionist, “And we also offer falafels.” At this point, you’re running out of there. Singular identity equals authority and a lack of singular identity lacks authority. If you’re a doctor who sells spine surgeries, car insurance, and falafels, that’s not very authoritative.
I want to get into this a little bit more granularly. I want to see how good Google is at doing this sort of thing. Back in September, during the September 2019 core update, I noticed there were a lot of loan sites that got hit and I thought it would be smart to look at the sites that were hit versus the loan sites that were not hit to see if I can find a pattern. And lo and behold I found a pattern. I forgot to mention this, but the same pattern we saw with the sites with so many ads all over the place was the same ranking pattern I saw for sites with too many CTAs, sites with too many internal links going to product pages, or too many affiliate links. Again, it’s not about ads, it’s about that undercurrent of a different identity or a conflict in identity.
I couldn’t figure out the pattern at first. I first thought maybe these loan sites had similar patterns to other sites with too many CTAs or too many internal links to product pages, but that wasn’t it. So you know what? I started to read the content on these pages and turns out that was a good idea. Look at this page here. This page is meant to inform you about different types of loans, but instead of doing that you get, “Why a Small Business Loan is Perfect for You.” That’s a little more like marketing than informational.
Here’s my favorite and when I say favorite, I mean it was the worst because it was for sure written by a used car salesman. Again, the same thing, this page is meant to be showing you straight up information about loan types, but instead, you have, “A business term loan never goes out of style. We all love a classic…” That’s a little bit too salesy for me and it was too salesy for Google.
Here, by the way, is a page that wasn’t hit by this particular update and you can see they do have promotional content on their site which is fine. Google knows you’re supposed to make money from your website. That’s totally okay. But when it came to talking about the actual loan type, all they wrote was, “Secured small business loans.” Here’s all the information about the loan, no marketing whatsoever. This means that Google understands tone backhandedly, which is amazing, and scary at the same time, but that’s not really my point. My point is that Google is sort of picking up on all these latent signals.
Imagine you go to a website and you look at the UI, the UX, the layout, the format, the pictures, the tone, the whole shebang. It gives off a certain feel, it sends a certain message to you if you trust this site. It’s very subliminal, it’s very latent. It’s what this site does overall that sends messages to you about what this site is and if it can be trusted. And Google, just like a person, is picking up on those latent subliminal signals about your site, and therefore, how trustworthy your site is.
To sum this up, identity and authority are synonymous so you need to have a strong identity. Google is profiling your site asking who you are while making sure there’s no conflict in identity and it’s getting really freaking good at doing this.
This brings us to what to do now? The first thing goes without saying but I’ll say it anyway is the idea is that to prevent yourself from being hit by a core update, if Google’s profiling your site and is asking if it has identity and conflicts in identity, the question then is how do you build a strong unconflicted identity?
On one foot, because I can speak an hour on each one of these points, do this. Pick an area, pick a topic, pick a focus, whatever it is. It can be very broad. For example, I write about SEO which includes many different things. It doesn’t have to be very specific or sub categorical, but pick one area and just stick to it. Tackle it comprehensively. Meaning, if you look at authoritative content, you’ll notice it has complete mastery over the topic. It looks at the topic from many different angles, from multiple vantage points, from multiple aspects, for multiple users, with multiple levels, and multiple understandings. Just complete mastery.
At the same time, I recommend writing about it “often.” For the same reason you want to post a blog post every once in a while to show that you’re alive, if you want to write authoritative content, if you want to have an identity, you need to be a part of the active conversation. You can’t be an authority and without keeping up with the trends or without pushing the topic or the focus forward. You have to be an active writer, which doesn’t mean you should write for the sake of writing because Google’s looking qualitatively at things and just writing for the sake of writing would be completely pointless.
Obviously, this means that you don’t target high search volume keywords for the sake of targeting high search volume keywords. That doesn’t make so much sense. I recommend that you don’t target those sorts of topics or keywords that kind of relate to what you do but don’t really relate to what you do. Don’t do it. If it’s on the periphery or the borderline of what you do, writing about it may dilute your focus which dilutes your identity.
The main thing that you really want to do is to manifest something. If you don’t manifest something, Google’s not going to be able to identify it. Now, there is a little bit of a hitch in this because sometimes you can have such a strong identity that the secondary parts of your site, the secondary identities, don’t do so well. When Google has to deal with looking at multiple identities it has a hard time.
For example, let’s say you are a big Derek Jeter fan and you want to create a website about him. You have his bio information, his timeline, and all the information about Derek. At the same time, you’re also selling his memorabilia because he signed baseballs, cards, posters, or jerseys. So your main identity would be the informational content, but there’s a secondary identity of selling memorabilia. Google might have a hard time picking up on the secondary identity of this site if your main identity is so strong.
Here’s another example of this. I want you to meet my favorite person who I’ve never met before. His name is Alan Page. Alan Page is an NFL Hall of Famer, one of the 100 greatest NFL players of all time, part of the Minnesota Vikings purple people eaters, and despite having his head clobbered day in, and day out, he became a Supreme Court Justice for the state of Minnesota. Super cool guy. Now, here’s his SERP and these are the related search boxes that show at the bottom of the search for him. I want to tackle now the first and the third box and then come back to the second box. The first box is ‘Famous Vikings Players.’ That makes sense as he was a famous Vikings player. The third box is ‘Famous defensive players in the NFL.’ That also makes sense as he was a famous defensive player in the NFL. Now, the kicker, the second box, for a guy who was one of the best NFL players of all time, and a Minnesota State Supreme Court Justice, the box “obviously” is ‘Notre Dame alumni.’ That’s kind of weird. Why is being a Notre Dame notable alumni the second most notable thing about him? Why didn’t Google pick up on the fact that he was a Supreme Court judge? I had to look this up, there are many players from the NFL who went into legal careers so there’s a plethora of other people to list there.
When I looked at this nine months ago, how Google picks up a sub identity or secondary identity, Google was only able to pick up on it 25% of the time. Now, it’s got a little bit better. So let’s be generous and say that they can do it now 35% of the time, that’s still not great. In other words, if you’re faced with having a really strong identity and a secondary identity, you might want to simplify things a little bit. Or as my parents always told me growing up when I overcomplicated things, “Keep it simple, stupid.” It may mean that you want to be creative in how you deal with it or it may mean that you want to just pull it out and do a whole separate site with that secondary identity.
To wrap this up, have a strong identity with a strong content focus. Stay away from tactics that will dilute your site’s identity such as going after high search volume keywords and even keywords that exist on the periphery of what it is that you do. And finally, multiple profiles can be complicated. You might want to be creative about how you deal with it or give it a second thought. And that’s all I got. That was my SMX presentation. I hope you enjoyed it. Let me know if you enjoyed it too. Toodles.