Ep 239: [Case Study] Core Algorithm Update Recovery To 55% Growth In Organic Traffic

Jaryd Krause:

Every algorithm update is recoverable. Let us show you how we do it. Hi, I'm Jaryd Krause, host of the Buying Online Businesses Podcast. And today I'm speaking with Rad Paluszak, who is a web developer and software architect with 20 years’ experience. He has been a technical mastermind in the SEO industry since 2010, and now he heads up the Buying Online Businesses SEO team with another partner, Rafal. Rad's specializations include international and technical SEO, machine learning, understanding machine learning, and looking at SEO from a business and management's perspective.

Now, in this podcast episode, Rad and I talk about how we helped a client increase their revenue by 373% and increase their organic traffic by 55%, which is absolutely awesome. We talk about what went into the SEO order that we did, why and how this site was disrupted by an SEO attack nine days after they purchased the business, the Google penalties that it went through, five different Google penalties over a period of three months, 90 days, and how we actually ordered the site, found out all of these things, the tools that we used to audit the site to gather all this information, all this data, and how we compacted that and changed that into a growth strategy.

We talk about in this podcast some of the technical SEO things that we revised, worked on, and fixed, and then what our growth plan was. And we added things like another version of the site and another language, and that increased traffic right away—actually, in fact, faster than we expected. We also talk about how we added monetization to the site to increase revenue. And then we talk about the results and the traffic increase towards the end of the show as well.

Now, there's so much in this, and I give you a bit of a brief when we start the podcast episode about how to get more information and more value from this. And this is based on a blog post, and you should be watching this via YouTube or via video somewhere if you can. Let's dive in. This is such a valuable episode.

Do you have a website you might want to sell either now or in the future? We have a hungry list of cashed up and trained up buyers that want to buy your content website. If you have a site making over $300 per month and want to sell it, head to buyingonlinebusinesses.co/sellyourbusiness. Or email us at [email protected] because we'll likely have a buyer. The details are in the description.

Rad, welcome back to the pod.

Rad Paluszak:

Hello, Jaryd. How are you? It's good to be back.

Jaryd Krause:

I'm good, thank you. And yourself?

Rad Paluszak:

Yeah, good, good, good.

Jaryd Krause:

Yeah. Good. All right, well, guys, we’ve got a lot to get through on this podcast because it's a fantastic case study. How to preface this, I guess, is that we had a client buy a great business. Obviously, there were some challenges and risks involved with it. We conveyed that there were risks involved, and they were happy to purchase it. These are the things that I'd need to work on to get business going well.

It did have some seasonality to it, and the traffic was declining through seasonality, and we didn't know how it would behave at the bottom of the season. Something not so great happened. There was an algorithm update, and we needed to jump in and recover it pretty fast. And we did so, right, Rad? I got you guys on board.

As you guys know, if you haven't seen it, we've launched the BOB SEO service so people can basically buy the businesses. We've been doing this for many years, helping people buy businesses. And the next step is that I've been helping people grow with some one-on-one coaching. And I wanted to sort of scale out and help more people.

And we started an SEO service so people can buy businesses and give their businesses to us to help them manage, run, and grow them for them. And Rad has been heading up there with Raf. And this is a really cool case study to share with you guys on how we've been able to do this with just one site that was hitting catastrophe, right, Rad?

Rad Paluszak:

Heading downwards, for sure.

Jaryd Krause:

Yeah, yeah, it definitely hit catastrophe. So where do you want to start with this? Guys, when you're listening through this, there is a blog post that we put out there; it's basically a 27-minute read, but it's very dense. Rad has created another case study prior to this, the SEO Blueprint Strategy, that you should check out. Both of those are on the BOB website. So go away and check them out, and there'll be links to those in the show notes.

But if you go through this case study, which is actually called From Algorithmic Penalty to 373% Revenue Growth (Case Study), that's its title. So look for the title, From Algorithmic Penalty to 373% Revenue Growth. You'll be able to click on that and go through it while you're listening to this episode, if you can. Or if you're listening in the car or somewhere where you can't view things while you're doing what you're doing right now, refer back to this.

Because we've got step-by-step procedures, it's basically a guide to what we've done, tools that you can use, and screenshots and images that you can use as references to help you grow your site. If you don't want us to do it for you, you can do it yourself with a lot of the stuff that we're sharing here. It's out of this world valuable, Rad.

It's really cool. So make sure you guys link to that. And we're going to be sharing a few things in the presentation. We're going to be sharing a few slides. So listen on in and watch it through YouTube if you can, or refer to the blog posts.

So with that admin stuff covered, Rad, where do you want to start with this? Is there anything that you want to pre-frame or mention?

Rad Paluszak:

Well, I mean, first of all, that blog post of 27 minutes sounds like a lot of reading. But like you said, it's so dense, and, I don't know, it's probably like 6,000 or 7,000 words or something like that. It's a book with a lot of images and a lot of stuff—the whole blueprint, how we really worked on this website. So I've got a presentation, which I'm happy to present during this recording. However, the presentation is actually cut short, and I'll tell you where exactly, because yeah, we wouldn't be able to cover all of that here.

Now, one thing I will definitely mention—I mean, at least at the time of recording this podcast—is that there’s been a recent core algorithm update on Google. Google just released it recently, and they named it as they usually do with the name of the month when it's happening. So they named it the August 2023 Core Update.

And usually, with these core algorithm updates, they affect different websites. I talk about the little bits of the actual ins and outs of the core algorithm update and why it is different from any other updates that are happening throughout the year. And obviously, we're talking about this in the blog post.

But yeah, I think this is going to bring a lot of value. Because if you were affected by that, I mean, I'm hoping you can just take some of the recipes we're sharing and try to apply them there. And if you can't or are struggling with that, we’re happy to help guide you through it.

Jaryd Krause:

Yeah, awesome. That's a really good thing to preface. Because this is going to be relevant for this update that just happened previously, just now in August. And then there are still so many things. As the new updates do roll out, there’s going to be so many things that are relevant for you, guys, that you can use for future updates as well as changes to the algorithm.

So, yeah, let's get into the meat and potatoes, Rad, of this. So let's get you sharing your screen. Guys, while Rad's sharing his screen, if you can view this on YouTube, be viewing on YouTube. And where do you want to start with this, Rad? Do you want to start with where the site was?

Rad Paluszak:

Yeah, I'll introduce a little bit about the site, then go through what we did, what happened, and then how we took it from there. We also had a growth plan other than just a recovery plan that we made along the way.

Just before I start, I think it's worth mentioning because, I don't know, it emphasized that Google has just released a core algo update, and it might have sounded like what we're sharing here, but it only applies to the ones in the past that happened. But since every single core algorithm update that is being released by Google is very similar in nature, which obviously means I'm going into a little bit more detail in the presentation, those recipes that we're sharing should be pretty universal, and you can use them any time your website is affected by anything that happens in Google. But, yeah, I mean, I think if you're happy with that, without further ado, I can jump right into it. Jaryd Krause:

Yeah, I'll hand the microphone over to you, and I'll jump in intermittently to maybe debrief a few things for people to emphasize.

Rad Paluszak:

Yeah, absolutely. All right. So speaking of the client, they recently purchased a website in the home and garden niche. So, like Jaryd, you mentioned it's affected by some seasonality. However, yeah, seasonality wasn't the only thing that happened here.

Now, a very interesting thing and very important at this stage was that the purchase date of that website was on November 20th, which is basically the date when Escrow released the funds to the previous owner, and therefore the website technically changed hands.

Now, what happened was, and I have a little thing here; I don't know if you can hear it, a few moments later, so for those of you who aren't watching it but just listening, I have a screenshot here from Ahrefs graph showing mainly the number of referring domains. And I put the line here exactly on the 21st of November, which is the next day after Escrow released the funds.

A few days later, on the 29th, for example, the number of domains that this site was getting increased by over double overall. And there was like a huge spike in referring domains. And this is pretty much something that, first of all, drew our attention, but we'll get to that in a second.

Jaryd Krause:

Yeah. So over a nine-day period, nine days after settlement, basically, there was this 50% growth in referring domains, which will break down shortly, right?

Rad Paluszak:

Yeah, exactly. So now, looking at the period of time from when it started losing traffic, which was towards the end of September, up until roughly when we started working on the site, which was mid or the second part of December, comparing that to the previous period of time, the website lost almost 40% of traffic user sessions.

So I'm not trying to put all the blame on the referring domains that just started kicking in in that nine-day window after the site was purchased. So I've got a screenshot here also showing that there was a decline—a visible decline over the whole period of time. So from the end of September until mid-December or towards the end of December, the highest drop was unfortunately around exactly after the settlement date.

Jaryd Krause:

Yeah. Like when the referring domains started, when the increase in referring domains started, that was when it started to dramatically decrease in traffic, right?

Rad Paluszak:

Yeah. And we got some earnings numbers from this website. So around September, to the best of our knowledge, the website was earning around $2,000. And then in December, it was only just shy of $900. So the earnings were slashed by over half. And then in January, which is basically when it hit rock bottom, it was only $669.

So understandably, the client was really worried and kind of freaking out about the investment. Because you're buying a website expecting much higher earnings, and then something happens and boom, everything goes to like one third of what it was getting. So understandably, they were really worried about the investment.

Now, the action plan. The thing with SEO is that a lot of SEOs will say that every site is different. So the strategy will be slightly different. Websites will have a different history and a different link profile. Things will essentially look different. However, what you end up doing is usually, during the investigation and analysis stage, you have a process, and usually you're performing similar things.

And when we approach pretty much any website, be it a modest blog or a huge e-commerce doesn't really matter; we always start with an audit. Now, how much resources on our site go into the auditing? That depends obviously on the size of the website. However, this is the main part that we start with. I like to say, How can you help a website without understanding it first? So this is why we start with the audit because we need to understand it.

Jaryd Krause:

Yeah.

Rad Paluszak:

Sorry, go on.

Jaryd Krause:

Yeah, go on. I'll just say, let's break down what goes into the audit, which we're going to do now.

Rad Paluszak:

Yeah. So the audit usually consists of technical analysis and implementation, Google Search Console auditing, and keyword research. This is part of a growth plan, actually, but we need to look at the keywords, analyze them, and see where we can take them. Link building, planning, strategizing, and then executing link building as well. And the content plan. So these are the usual elements we look at. And I hope you agree, Jaryd, that these are the elements that are pretty standard for every website, and they cover all of those most important elements in SEO for any website.

Jaryd Krause:

Yeah, of course. Keyword research. You're going to do keyword research for a growth strategy and in the content plan. But through the audit, you obviously need to look at what keywords are doing what and where they're at, their positions and volume, keyword density, and all that sort of stuff, right?

So as we go through, we're glossing over things here and what we have done. That's why we want you to refer back to the full article on the site and go through it in detail, because that will be far more detailed than what we're covering here. But this is to just sort of showcase that these are the things that can be done, and then you can go away and do each individual thing on your site alone and see it in more detail through that post. So, yeah, awesome.

Rad Paluszak:

Absolutely. I think so. Also, just for a slight clarification, we use a mental shortcut when we say SEO audit. Because, if you think about it, an audit usually results in a document, a spreadsheet, or a presentation, depending on the format, from different agencies. In our case, it's a pretty hefty document that usually specifies issues on the website and then the auction plan.

That's why I can't stress enough that it's not just the audit. Because an audit isn't really executing the strategy. An audit allows you to know the website, and based on that, you're building the strategy, and then you're approaching the implementation. Obviously, without those three elements, if you only have audit and strategy but no implementation, then nothing really reacts on the website because you haven't really done anything other than analyze everything properly.

Now, obviously, for this particular case, since we noticed pretty early on that we're actually dealing with some sort of penalty, we did penalty analysis. And obviously, like I mentioned, since we have this huge spike in links, we performed an analysis of potential negative SEO attacks. And then we were also testing different approaches more towards core algorithm updates, which I'm also talking about in a second.

And this is going to sound like we staged these things one after the other, but pretty much everything was happening roughly at the same time. There was like a month difference between everything. So also, like I mentioned earlier, we had the growth plan for this site because you can't really lose this, let's say, umbrella goal that is above all the other goals that are growing the website, regardless of whether you're under the penalty and your traffic is temporarily down or you're just growing it, and everything is good and dandy. You just need to have some sort of growth plan. And for this particular case, we also looked at monetization.

And I could sit here and talk about auditing and how we did the whole audit in detail, but that would probably take us half a day.

Jaryd Krause:

And you can look at it through the post as well.

Rad Paluszak:

Exactly, exactly. Yeah. Auditing takes a little bit of time and is very mundane. So what I would like to do instead is to focus on leveraging the main things from the audit. So I’m mentioning most of the things we did. However, I focused on a few that really gave us leverage so that we could execute and implement them faster.

So with the SEO audit, our best tool—hands down—is site bulk. We use that crawler for most of the analysis that we do. After we crawl the website, it gives you a nice, visually appealing audit score, and everything is nicely laid out there.

And I think you said that in the SEO blueprint article on your website, and I think it's also mentioned in this whole article of the case study that even if your technical prowess is not that great, if you did run the audit with Site Bulb on your website, you can basically just export a PDF report from it, give it to a web developer, and the web developer will already probably fix a lot of these things. And this is how great Site Bulb is.

Now for the audit elements, in this particular case, we looked at mainly or found duplicate content, images with missing alt text. And now it makes me laugh about that a little bit. Because even the cheapest tool on the internet does something similar to SEO auditing, most of the reports that you're getting from people who are only salespeople trying to get you to buy SEO from them will include something like, “Oh, your images are missing alt text.”

And the thing is, we usually don't react to these things. However, as SEO's job, your main thing is to assess how severe the issue is versus how difficult it's to implement it versus the potential benefit. And in that case, when we were looking at images with missing alt text, you can imagine the website was in a home and garden niche, pretty visual, a lot of equipment, a lot of guides, and needed images.

So if we looked at that and only 10% of the images on the entire website, and that was over 1,000 of them in total, had alt text, then we actually assessed that it's pretty important to do that. And actually, by just fixing alt text and adding alt text to the images, we gained some traffic through image search, which was great.

We also looked at page speed and core with vital things, duplicate or multiple H1 tags, and missing meta descriptions. Now for the last two, so H1 tags and missing meta descriptions or meta descriptions in general, we found recently a crawler called Horseman, and it's a very simple, very cheap tool. It's just a simple crawler, you would think, but the beauty of it and its power lie in the fact that you can connect Horseman with, for example, ChatGPT or other external tools.

And what it does is, if you know a little bit of JavaScript or you can actually amend the codes in the examples that they give you, you can basically crawl your website once, like you would usually do with Site Bulb. But Horseman, at the same time as it's crawling the site, can actually take and fill in all the missing meta descriptions, rewrite your H1 tags, add your H2 tags, and give you a list of all of the pages where your heading structure isn't how it should be. So it does a little bit more logic, and it's most powerful when it connects to external tools.

Like I said, this is my favorite, and this is just an example of how we're using Horseman to rewrite the text of the article to do some summary of the content, which is really great. And after it runs through your website and does that, you can export all of that to a CSV file. And when you have the CSV file, if your website is on WordPress, you can use the WP All Import plugin, which takes in the CSV file, and then you can just use a drag and drop to match whatever you have in the CSV file to the fields in WordPress and then upload like hundreds of records.

This is where we amended like 322 records in like two minutes through WP All Import after running some prompts with Horseman on the site. So it's great automation. It saves a tremendous amount of time, and it's actually pretty simple. When you get used to Horseman’s little scripts, it's very simple.

But moving on, I also mentioned... Okay, sorry, you wanted to ask something?

Jaryd Krause:

I was just going to say that's the order. Yeah, let's move on. Yeah. I mean, guys, you can check out all this in the post. So there’s so much in the post we're just going to refer to. Continue.

Rad Paluszak:

Absolutely. So another thing we did during the auditing stage was a Google Search Console review. And we found one of the biggest issues that you can actually find in the Search Console under the indexation. If you ever see "crawled—currently not indexed,” and let's say you have 1,000 pages on your website or 1,000 articles, and then in this case, we had 758 pages that were "crawled—currently not indexed” by Google, that indicates that Google has a problem with something on your site, that is, it's not so eager to index the content after it sees it on the site.

Now, usually, and this was our observation on this site, this problem has something to do with the site structure, and this isn't from any tool. I just made a visualization with the ChatGPT. But to give you an idea, on the site we had over 500 blog posts, and I think they were located in five or six categories. So I have a visualization here.

But you can imagine you have 500 blog posts in five categories, so it's roughly 100 blog posts per category. So Google will spend a lot of time to actually get to those individual posts, and you can logically see that it’s not perfect from an information architecture standpoint.

So what we did was introduce 20 more categories, and it really helped with spreading out the content on the site and making it more understandable for Google. And Google started crawling it much better, and that actually helped with the indexation and solved our problems with the site structure as well.

Jaryd Krause:

I guess it's good to point out with the site structure that if you've got five categories and you've got 500 posts and 100 in each category, you've only got a certain amount of crawl budget that you get from Google when they come to your site to crawl it and find your articles and rank things, right?

So if you've got one category that's got 100 posts in it, for example, and you've got five of them, it may expend a lot of that budget on just one category trying to find what's going on in there when you've got 100 posts. But if you spread them out to multiple ones, then it makes it so much easier, and you spend less on the crawl budget, and you can get Google to look at your site far more effectively without it having a huge load on the crawl budget. Is that right, Rad?

Rad Paluszak:

Yes, that's a perfect explanation. For this, I actually like to use an analogy. So if you imagine you have a big crate of, let’s say, colorful coins, and each crate is like one ton, So obviously, if someone tells you, okay, find me one golden coin in that crate, that will take you probably hours looking through and taking out bit by bit from that crate and looking through all of these crazy colorful coins to find just one that is golden, right?

However, if you took, I don’t know, 30 buckets, and each bucket would get a little bit of those coins, and you started looking through them, and there were a few people looking through them, you would be likely to find that one coin much faster.

And this is how Google operates. So the crawl budget, if you guys don't know how to measure it, is basically the amount of time that Google is allowed to spend on the website. The more popular a website, the more time Google is allowed to spend crawling it. And so if Googlebot runs out of this time, like you said, if it has that big category to go through, if it runs out and hits the limit on the crawl budget, it just goes away and comes back later this week or later next week, whatever the schedule is for your website.

But if you make those categories smaller automatically, Google is probably much faster to discover all the posts within each category. And then it sees that, oh, hang on, there is actually one more category. So it might increase the crawl budget, seeing that there is more to see on that website.

Because if the only thing it sees in the category is pagination and not a blog post, you can think about it like it's getting boring for Googlebots and it has no reason to try to explore more because it knows, okay, it's just the same old, same old, and I can discover it at a later date.

Jaryd Krause:

Awesome. Awesome analogy of site structure. It is so good for people to understand. So, yeah, let's move on to that penalty analysis. Because this is what I really wanted to chat about around these.

referring domains and why that may have happened or maybe not, but certainly what the reaction was from these referring domains. So let's dive into that.

Rad Paluszak:

Yeah, for sure. So we started working on the site around mid-December last year. We had keywords dropping and traffic dropping; everything was looking pretty dire. And like I said, our initial suspicion was that there was no SEO attack because we saw this huge spike in the referring domains that drew our attention because it was just so unusual.

Jaryd Krause:

Let's talk about the negative SEO attack. That can be where somebody refers to a bunch of spammy links to a site and its pages. Explain that to us and how the site can react to that.

Rad Paluszak:

So before I start, a little disclaimer, just so someone from Google is listening to that, according to official comments from Google representatives, John Miller, and these folks, something like negative SEO doesn't exist. However, practically looking at that, we've seen that happen quite a few times on different websites.

So we have an inclination to think that Google might be claiming that something like negative SEO doesn't really work or doesn't exist; however, we think it actually does and can affect your website.

Now what it is, it's basically someone or something; automated tools and automated websites are building vast numbers of links to a domain. And those links might be made with the intention of being punitive to the site's traffic. That's why it's called a negative SEO attack because they're meant to basically derail your SEO strategy and affect your keywords and the rankings in a negative way.

Sometimes it happens that something looks like a negative SEO attack because all of a sudden you're getting a lot of referring domains, but it's just an automated system that is maybe, I don't know, collecting your images and then linking to those images to give you credit and to generate traffic for those websites that are showing your images. We've seen that. It does look like a negative SEO attack, but it usually doesn't affect your rankings or traffic. So it's not really an attack. It's just a coincidence.

In this case, we definitely dealt with something that for sure looked like a negative SEO attack. I've got this here as another suspicion before we go to a negative SEO attack. Another suspicion was the core algorithm update. Because in that same period of time that we were working on the site or while the site started losing traffic, there were three pretty big core algorithm updates and another two big algorithm updates in Google, but these were more related to links. So there was like a whole mixture of...

Jaryd Krause:

Yeah. For people listening, we've got basically dates of updates and the names of those updates and what they were, and we've got five of them listed on our screen. And what each of those updates was linked to, like product reviews, core updates, the October CMS update, helpful content updates, and link spam updates Multiple things happened within this period, which was a cluster storm of things that had happened.

Rad Paluszak:

Enough said, the end of 2022 was very active from Google's side. Within three months, they released five big algo updates—very broad algo updates. So this is quite unusual for Google. We've seen that before, but it's still not that usual. Like you said, it was a cluster of updates.

But yeah, back to the negative SEO suspicion. So I've got a screenshot here where I highlighted two updates that were happening at the time that were related to links. Like I said, there were five, and three of them were broad core algo updates, and two were related specifically to links.

Now when we looked at the dates and then we looked at the number of links that the website was getting in November, we started sort of seeing, and I say sort of seeing because when you're dealing with an algo penalty, you sometimes need to basically take into consideration things that don't really appear to be an issue.

So the more we were looking at that date alignment with when the website was losing traffic, the more we started seeing that roughly two weeks after every link-related update from Google, the website was hit. So that gave us an idea—okay, a negative SEO attack.

And if we, like I said, combined it with that huge spike in the links, plus we combined that with websites with relatively low domain ratings, which some consider a good measurement for website authority, me included, by the way, we started seeing that as a really potentially big culprit of the decline.

So what we did was do a full link review of everything that was going towards the website, especially in this period of time. And I've got a list here. Some of these links you can copy, or just check out the websites that we identified. They did look really, really, really bad. I've got an example here. So for those of you who aren't seeing this, just listening, we have a page that definitely has some bad content, just pun content with added links.

Jaryd Krause:

It basically just looks like a page full of keywords and dodgy links, and it doesn't actually make sense.

Rad Paluszak:

No, not at all.

Jaryd Krause:

And there's multiple ones of these linking to this site that we were working on, which obviously Google's like, “That's not valuable.” That drags down the site that we were working on, right?

Rad Paluszak:

Yeah, for sure. You have no reason for this page, the whole domain, or several domains that look the same. There's no reason for them to exist. And it did seem like it was done on purpose, or that it was built for negative SEO or some really bad type of old school SEO link scheme. But yeah, it does look really, really, really bad.

So what we did, like I said, was identify a lot of these, and then we just added them to a disavow file. We disavowed over 400, or exactly 440, domains. Just to emphasize, I'm not personally a big fan of disavowing domains because I've seen it too many times that SEOs were a bit too trigger-happy with disavowing, and they just disavowed too many good domains or domains that were actually providing some sort of value.

And by the way, I'm always trying to weigh what Google officially says. And one of the things they say is that they don't really allow, like I said, negative SEO to happen, and they're really good at identifying bad links. So there's not much need nowadays for disavowing.

And I can agree with the part that there is not much need for disavowing because for those domains that are so blatantly bad for you, Google is easy to take care of. However, the problem appears or arises when you're actually dealing with something that is most likely a negative SEO attack. In this case, you have a big spike of really bad domains flowing to your website in a short period of time. Google is just not

Jaryd Krause:

A 50% increase in referring domains, which increases mostly from non-valuable links That's something to deal with.

Rad Paluszak:

Yeah, Google is just physically not able to crawl all of these bad domains to notice them and then take action to actually disavow them algorithmically on its own. It's just not possible to find them all. So in this case, we added them to this file to help it out. Now, another thing is, and that's another screenshot that I have here for those of you who are listening, I'm showing Google Analytics traffic graph for this website with added periods or dates when any algo update has happened in that period of time. And this is, I think, something you can assess yourselves if you're watching it, and Jaryd, you can confirm, but it's pretty clear that every single one of those updates caused some detriment to the traffic on this site.

Jaryd Krause:

Absolutely, absolutely. And guys, you can see this on the blog posts that we're linking to as well. We go through all these images in the blog posts too. So, yeah, you can see that there was a storm of the algo updates and then also the negative SEO attacks. So we knew that we needed to do some work. So should we move on to talking about some of the updates and then into what we did to grow this?

Rad Paluszak:

Yeah, yeah. So just quickly, core algo updates are a very specific thing that Google rolls out. If you think about it, it's not just an individual thing that’s happened to your website, and you've just fixed it and that's it. Think about it in terms of the whole playing field shifting. There might be different reasons for that.

It might be something like the name suggests, changed at the core of the algorithm. AI got smarter or updated the weights of how it looks at the parameters. But one thing is for sure, There's no individual fix.

And in fact, you didn't; if you think about it, your site didn't really get penalized per se. It sort of got what it deserved based on what changed in the algorithm. So the overall quality of the site kind of deserved it. So just to emphasize, it's not any one thing you did or didn't do; it's everything. It's like the whole picture that you should be looking at. So with that in mind…

Jaryd Krause:

It's the whole site. It's not just one part of the site; it's the whole site. And then how does that site, your site you've got, behave in the environment of the new algorithm or the updated version of the algorithm? So there are multiple things on the site that might need changing. It's not just like, “Oh, I just need to make sure my content's valuable.” It's a lot of other things as well that could happen.

Rad Paluszak:

Precisely. So that's all of the things that might have happened. So what we did was go through the auditing that we did once again and just go back to the mundane basics. So we rewrote most of the meta tags. We looked at 404 errors and just did an overall cleanup. We made content improvements. We looked at the FAQ sections for key takeaways.

By the way, speaking about content improvements, it's an example of one final takeaway from one of the articles on this website. And I'm just highlighting some really badly written, language-wise content here. But my very favorite of those is, and I'm going to quote it, the sentence says, “There is nothing for a lifelong last.

So there is no reason to be upset about the replacement.” And I think what it was trying to say was meant to say that nothing lasts forever. So whatever part, whatever equipment, sometimes has to be replaced. But you can see how bad this content was. Obviously, it's all in the screenshots on the website.

Jaryd Krause:

It's written very poorly. We highlighted multiple parts, and just like two paragraphs, they were written super poorly. So update that.

Rad Paluszak:

It must've been spun content or something. But apart from that, like I said, we introduced FAQ sections for most of the content and obviously added JSON-LD structured data to that. We introduced key takeaways for most of the websites at the beginning of the article. I've got a screenshot here from the Daily Mail.

All the big guys are now using it, and I think it's very user-friendly. Because if someone lands on the top of the website, they can assess from the key takeaways if this article is something they're looking for or not.

Now, the content tools we used for that were mainly GPT-4. We automated a lot of that. Some of it is using the aforementioned Horseman, some of it is just through the API, and some of it is through ChatGPT. We also looked at people asking, especially for the FAQ sections.

I know what you're thinking. You might be saying, “Oh, Rad, but isn't AI content like a little dodgy area from Google now?” Google says one thing; they say, “Oh, it doesn't really matter where this content comes from,” but no one really knows what is really the thing with it.

What’s the story? So from our perspective, we usually try to find a good balance between automation and personalization. And I think if you think about that from this perspective, it's pretty straightforward that you can use automation, but make sure there's enough personalization in that.

Jaryd Krause:

Yeah. Make sure it's super valuable. Guys, we also—Rad, you, and I—have done a podcast on AI content generation risks.

Rad Paluszak:

Yes.

Jaryd Krause:

And it's Episode 220. And it's titled AI Content: The Good, the Bad, the Ugly, and How to Make It Truly Valuable. So that's with you and I, Rad. If you guys want to know more about AI generated content and how to do it the right way and avoid the risks, then Episode 220 check that out.

Rad Paluszak:

Yeah, that was a very, very meaty episode.

Jaryd Krause:

It was meaty. And the number of links that we have in the show notes for all the different tools and all the different things is huge. And we got some really good reviews from people saying, “Wow, that was a great podcast episode.” And the resources that we have in the show notes are out of this world valuable. So guys, check that out, Episode 220.

Rad Paluszak:

Yeah. I think I'm going to pick up the pace now because it's actually getting pretty long.

Jaryd Krause:

Yeah. Let's move into some of the growth things.

Rad Paluszak:

Towards early March, we started seeing good signals that the penalty was lifting. We were still losing keywords, but we started gaining traffic. And it might not initially make sense. But if you think about it, Google has already granted us traffic, so the trend has reversed.

However, the reason we were losing keywords was because Google was still purging those keywords that it thought were not aligned with either user intent or just didn't make sense for the site to rank for. But that was a really good signal we saw here.

What happened, though, was that, like I said, early in March, we started seeing some early signs of recovery. Then, for those of you guys who are just listening to that, there’s a term we use internally called the bucket of death.

And that is called that because when you look at the traffic graph on the website, if you suddenly see a big decline, and a couple of weeks later, you see that this traffic comes back, it kind of looks like a bucket. That's why we call it the bucket of death.

Usually, that happens. And it doesn't necessarily mean that you had to fight with a penalty or some devaluation. It sometimes happens naturally around algorithm changes that Google releases. It means that Google is reevaluating your website.

So if you're losing traffic for sometimes two or three weeks and then it comes back, usually stronger, then that means that Google sort of took your site down with the new algorithm but then reevaluated the entire website and everything went back.

So we had this little scare here from the end of March until, like, mid-April. But the graph here—for those of you who don't see that—looks very, very, very strong after that bucket-of-death moment.

Jaryd Krause:

Yeah. Huge, huge growth up to the right

Rad Paluszak:

Yeah. So for the growth plan, like I said, we did keyword research, content planning, link building, additional monetization, and a Spanish version. Now I won't go into all of these. Just a few notes about the additional monetization If you have a content website, obviously that's a great way to monetize the site; however, this

Jaryd Krause:

Through ads.

Rad Paluszak:

Yeah, sorry. Through ads. I forgot you didn't see my screen. Some of you might not see my screen. Yeah. So if you're earning money through display ads, then obviously that's a great way to monetize the site. However, sometimes it's difficult to optimize it and earn more money with it.

And I just have an example here, because you can obviously place more ads on the site. But then in that example here that I have, it looks like there's only, I don't know, 30% of the screen that is actual content from the website, and then everything else is an ad. So this isn't

Jaryd Krause:

In fact, we did a study that the less ads, less means more sometimes. Well, quite often, actually.

Rad Paluszak:

Yeah. Conversion is better, I suppose.

Jaryd Krause:

User experience. Better conversions and a better user experience on the site So the site gets better SEO rankings, all that sort of stuff. So, yeah, let's talk about what we added in terms of monetization on top of ad revenue.

Rad Paluszak:

Yeah. Just one thing to add, too many ads on the page can actually lead you to get penalized by Google because Google has this algorithm that detects how many ads there are. And if there are too many, your site might get penalized.

So we recommended this client, in particular, Amazon Associates, as the easiest program to start working with in terms of the additional monetization. Because if you're talking about the products, you might as well just include some contextual affiliate links and make some additional money with that.

For that, our go-to plugin is AAWP. I think it's called something like an Amazon affiliate for WordPress. And we have two examples of what a product box looks like that you can implement with a very simple one-line short code from the plugin. It's very easy to add products from Amazon, whichever you want, and add those nicely looking, very aesthetically pleasing product boxes into your content. It just makes sense.

Obviously, you need to remember that if you're going with the Amazon Affiliates Associate program, it’s also oversaturated, has relatively low commissions, and has a pretty short cookie window. So there are upsides and downsides to this program. I think it's just very easy to implement, and that's the biggest thing about it.

So if you only monetize through display ads, the way to start making money with that is very easy. And then you can, for example, change it to some other affiliate programs. I have some examples here, like Impact, ConvertKit, ShareASale, and Partnerize. These are very popular. They sometimes have larger commissions.

Jaryd Krause:

Commission rates and longer cookie durations.

Rad Paluszak:

Yeah. So you might essentially earn more money with these. However, with Amazon, it's a quick test. So if it starts bringing you some money, then you can obviously motivate yourself to jump ship to something else.

But obviously, there are plenty of things you can implement for additional monetization. You can create a newsletter or direct ad sales. You can just approach some brands and offer them ad space on your website directly. You can delete generation apps, cross-sell, drop shipping, and so on and so forth.

I mean, I don't know what's your view on this, although I'm guessing, similar to mine, Jaryd, that the more ways to monetize websites, the better. Because it not only gives you more revenue streams, but it's also more stable if something happens with your traffic, for example.

Jaryd Krause:

I do have caveats to that, depending on how much work is involved for the ROI that you actually want as well.

Rad Paluszak:

That’s true.

Jaryd Krause:

That's involved with multiple monetizations. Yeah. Let's move into Spanish. We're talking about adding another language here.

Rad Paluszak:

I'm just going to go through it very quickly. So we decided to create a Spanish version of the website because the site was just in English. I mainly use ChatGPT to be my technical assistant. I'm a programmer, but I don't program in WordPress directly. It has its own caveats, which I don't really like. But ChatGPT was very helpful.

We connected that with DeepL API translation. I've got an example code that I used here and the output when the whole thing was running. So after I built it, and by the way, it's important to mention that we did that on the site as a proof of concept,

So we didn't charge the client anything for the additional work that we spent on building this whole tool, connecting it with an API, or doing the translation. We wanted to prove that it's going to work, essentially.

So we did 590 articles translated from English to Spanish within, I think, two hours overall. Downloaded from WordPress, translated, uploaded back, all published with all the content. And so it was crazy. But the crazier thing, for me at least, was something that really surprised me.

After just a few days, Google actually indexed 427 of them out of 590, which is very fast indexing. Something you don't really see that often. So it must mean that all of the changes that we made to the site were in Google's favor.

I've got a screenshot here. So since we put this Spanish version live, within three months—roughly, I think, roughly three months—from April to July, it got over 5,000 clicks and 209,000 impressions. So yeah, the API translation was $100 or something.

I think it paid back double so far. So it may not have generated that much traffic. But if you're watching this, or if you have a look at the article later, you'll see that it's on the constant rise. So I think this traffic will pick up more over the next few months.

Jaryd Krause:

For sure.

Rad Paluszak:

So it's great, yeah. And just to summarize, obviously, speaking of the results, we did the recovery. We made the recovery. Year over year, sessions and users were up 55%. So that's great. That is looking at the period of time from April to the end of June. Because in all this battle with penalties, we actually forgot that Universal Analytics was retiring.

Jaryd Krause:

Exactly.

Rad Paluszak:

Yeah. We looked at April until the end of June, and I've got some screenshots for that here, obviously, so that was a 55% increase overall in sessions and 57% in users. But then, if we add it on top of that, the next period, from July to August, we still had a 9% increase month over month. So that is just to show that year growth continued to grow after that period when Universal Analytics was retired.

Oh, and sorry, revenue numbers. So you remember how I said in January that we hit rock bottom at $669? Since then, because obviously —

Jaryd Krause:

In February?

Rad Paluszak:

Yeah. I was just going to say that in January, obviously, we were a month into our work. But this wasn't enough for Google to actually index anything. So our work hasn't been seen on the site yet. But just in February, it's gone up. In March, it was already $1.5k. April, almost $3k. And then in May, we hit a 373% increase. And we've gone up to over $3,100. So that was pretty good.

I think after May, due to seasonality, display ad commission has changed slightly and dropped. So we went down with traffic. But then, in July, we also started hitting these sweet affiliate revenue numbers. So that's just adding up. And just compared to the lowest month, we still had like a 262% increase in revenue. So I think that's pretty good for myself.

Jaryd Krause:

Wow. Yeah, a 262% increase in revenue is massive.

Rad Paluszak:

Yeah. That's all for the presentation. For those of you who are watching here, I've got a representation of an AI imagining my avatar. It looks kind of like The Witcher. So, obviously, we can see. —

Jaryd Krause:

It does. You do look like a Witcher in that image there.

Rad Paluszak:

Yeah. We can see that AI isn't really based in reality that much. But, yeah, thanks for this. That was the presentation.

Jaryd Krause:

Yeah. Thanks, Rad. That's so cool to see how much work went into this. And the client is stoked and over the moon with the results. And the business is going to just continue to grow because it's built now to withstand some serious changes. And he's just continuing to stay with us because the results are awesome.

And the reason I want to record this is, so you guys go away and check out this post. I'm going to link to it. It's going to be probably the first link in the show notes. Go away, check out this article. There's so much of what we talked about in it step by step that you can do for yourself.

And if you don't want it, totally fine. We can help you with it as well. And that's why we set this up. That's why we set this service up.

We had been working on our client sites, helping people buy sites for many years. And for the last sort of almost 10 months now, Rad, you and I and the team have been working behind the scenes, working on people's sites, and growing them.

Before we launched our SEO service, we decided to proof and get tests of proof and all that sort of stuff. And yeah, we've got some really good case studies. And this is one of them. So that's why we launched it.

And now people don't need to fuss Rad on “All right. If I buy a site, what do I do with it then next?” Well, you can check out some of our content and our case studies that'll help you grow if you want to grow it yourself. Or give it to us, we can grow it for you.

Rad Paluszak:

Yep, precisely. Exactly.

Jaryd Krause:

Absolutely. Anything else that you wanted to mention before we wrap this one up, Rad?

Rad Paluszak:

I think the word of advice is, if you are hit with an algo penalty, don't panic. It happens. It happens. I've seen that happen to some really good websites. Even though quite often there is something, some things, there are some things you can focus on, but overall, I've seen that happen to pretty good websites that require a bit more creative approach to trying to lift that.

One big downside of a core algo update is it sometimes doesn't release your traffic when you fix all of these things, even. Sometimes you have to wait for the next or even one other algo update to see results. So don't panic, be patient. I know it hurts, it usually hits your revenue numbers, hits your traffic, so throw that your revenue numbers. But I think most of the websites are recoverable.

Jaryd Krause:

Really good point, really good point. Because a lot of people listening are, like, “I wanna buy a content site, but I'm worried about if it gets hit by an algorithm update.” Well, if it gets hit by an algorithm update, you can fix it.

You can spend some of the resources that you're making from the business per month to go back into getting help to fix it, or you can fix it yourself with a lot of what we talked about here.

And it can be difficult, it can be tricky. It's not a local approach. If I fix this one thing in my site, it'll unlock the rest and Google will release my traffic. It is a global approach to multiple things. Hence, why it's smart to have an SEO help you with it. And that's what we do. And you can recover from these yourself. And I've helped people and seen people do it themselves.

And if not, then don't freak out when you're buying a business, a content site, that if it gets hit by an algorithm update, it’s gonna go to zero. I've never really seen that happen ever — a site go to zero. Unless somebody has the site, and it gets hit by an update and they give up on it. That's when it goes to zero.

Rad Paluszak:

Yeah, good point.

Jaryd Krause:

But if you're active and you go, “Oh, I'm gonna do some things to change it,” most of the time these are recoverable.

Rad Paluszak:

Yeah, absolutely. Think about it from a bigger picture perspective. And I know it sounds like a waffle that doesn't really mean anything, but Google looks at the overall quality of the website. So if you unpack that, what overall quality means, it's all of the small things, all of the big things, user experience, your content quality, your information architecture, site structure, and all of that.

So, yeah, sometimes it is tough. But like you said, it's a good point. I've never seen a site hit absolute zero unless it was doing something really, really deeply black hat. But I'm sure, especially people who work with you on due diligence or use your framework, you're not gonna buy a site like that because it will be revealed in due diligence. So you don't have to worry about it.

Jaryd Krause:

Awesome, Rad. Thanks for coming on again. Guys, check out more podcasts that I've done with Rad. We've done quite a few now, Episode 220 in particular, when we talked about AI generated content. And reach out to us whenever you guys need.

Rad Paluszak:

Okay, thank you very much. Take care, guys.

Jaryd Krause:

Hey, YouTube watcher, if you thought that video is good, you should check out this video here on 2 Types of Websites Beginners Should Buy. Or check out my playlist on How I Made My First $100k Buying Websites and how to do due diligence. Check it out. It's an awesome playlist. You'll enjoy it.