Take a fresh look at your lifestyle.

Helpful Content Update Analysis, Google Antitrust “Leaks”, SGE Lite, and Weird Niche Sites



It’s Friday, and that means it’s time for another episode of the Niche Pursuits News podcast, where Jared and Spencer break down the latest SEO news into bite-size pieces. Buckle up people, because they cover many, many topics on today’s episode!

It’s been one week since Google announced its core update for October, plus the spam update was rolled out, so Jared and Spencer first talk about how to know which may be affecting your site.

For those affected by the HCU, they look at a very exhaustive, data-heavy resource and highlight some of the conclusions, including what type of content performed better, how new sites fared in the update, how article length importance has shifted, and conclusions about AI content. They also share their opinions on some of the author’s theories and touch on the future of updates.

Watch the Full Episode

Moving on, Spencer and Jared address Google’s antitrust lawsuit and the recent news reveals that Apple was in talks to buy Bing or DuckDuckGo. They talk about the money that was exchanged between Apple and Google and how Apple negotiated the deal.

The next topic is how a recent op-ed on Wired claiming Google alters users’ search queries was recently pulled from their website for “not meeting their editorial standards,” and Jared and Spencer discuss why it might have disappeared.

In the latest SGE news, they talk about how Google appears to be testing a lite version, where the results are taking up less space at the top of the page. They share a few theories about why this is, how the experience has changed over recent months, and what this means for website owners.

In other SGE news, the conversation shifts to Google’s AI-powered search and its ability to generate images and drafts. They also talk about how AI images are something we should be paying attention to.

The next topic is how Google is using clicks to rank websites, contrary to what the search engine has said for years. They talk about the implications for privacy data and how this information may change the way we operate. 

In the Shiny Object Shenanigans portion of the podcast, Spencer talks about a website in the news niche he created with his kids, the goal of which is to get traffic from Facebook. He built up a Facebook page to 30k followers and the website has traffic of 27k over the last month. His goal is to get into Mediavine in the coming months, and he’ll report back as he and his writer make progress.

Jared then talks about a spreadsheet he created to help website owners, SEOs, and content creators with prompts relating to the HCU. It’s free over the weekend and then he may start charging for it, so grab it now!

You can download the HCU spreadsheet from Jared right here.

When it comes to weird niche sites, Spencer shares Cake Wrecks, which documents cake fails and adds some funny titles to the cakes. The DR59 site, which has been around for many years, is making money through ads and the author also has a book. 

As for Jared, his site is History Colored, which adds color to traditionally black-and-white photos. What probably started as a personal project to add color to photos eventually expanded into a bigger website that targets informational topics and gets more traffic. They talk about how some pages might go viral. With 515k followers on Facebook and 355k subscribers on YouTube, this is a topic that does well on social media.

And that brings us to the end of another jam-packed episode of the podcast. See you next week!


Spencer: ​Hey, everyone. Welcome back to another episode of This Week in Niche Pursuits News. I’m Spencer Hawes. After a week of vacation, actually, I was at a conference last week in Vegas. Um, uh, small business online, digital entrepreneur, uh, type conference. It was a lot of fun. I met a lot of great people there connected with them.

Um, but Jared, you were filling in, um, for me, we had Josh filling in for me last week and, uh, you guys did a great job, of course, but I’m back. In my regular spot. Um, how are you doing this week, Jared? Well, 

Jared: after seeing the agenda for today, I’m feeling a little overwhelmed to be quite honest with you, but other than that, I’m feeling great.

Spencer: Yeah, we have so much to cover here. Um, we were just trying to review, like I’ve got. About a dozen tabs open in front of me and we’re trying to review and it’s like, you know what? We just need to hit record because It’s going to take us another hour just to review everything and so this truly is live guys I mean, we’re just recording we’ve studied but there is so many updates that have happened Uh within the last week that yeah, we’re gonna try and cram it all in here.

Uh for you guys So the very 

Jared: first I feel like it’s sorry I just had to make a little joke because it was what was on my mind, you know The old movie Anchorman where they’re like, don’t you know whatever you put on the teleprompter he’ll read? That’s what I feel like right now because we just got this long list of things we’re going through and a lot of this We’re gonna have fresh takes on that.

We haven’t really been able to discuss beforehand. So it’s gonna be um You know, that’s just the nature of what happens when there’s so much news to report 

Spencer: on. Yeah, and that’s why we do this, is because invariably there is always more information to cover. Uh, which makes this industry very exciting. Now, uh, we are going to try and have time for our side hustle shenanigans and our weird niche sites.

We will definitely get to those. I will just say that while I was at a conference, uh, it sparked a weird niche site. Um, and so I’ll explain. How I got that, uh, I did not idea, but, uh, somebody, you know, shared an idea with me. And, uh, so we’re going to cover that at the end. So let’s jump into the news for this week.

So, uh, it’s been about one week exactly since Google announced the October core update. And so we did want to acknowledge that, um, that that is happening. It usually takes about two weeks to fully roll out. And so it still is. The middle of that. So if you’re seeing fluctuations in your rankings, just know that it might be because of the October, 2023 broad core update.

Um, I, I will say that on niche pursuits, I have seen a slight, uh, downtick. So, um, I do think that the October core update has hit niche suits a little bit. It’s not a lot, it’s pretty minimal, um, not a big deal, but I, but I have seen something. Have you seen anything on your sites, Jared, or any that you, um, are monitoring or helping out with?

Jared: Some fluctuations, some ups and downs, nothing, no, um, often times in an update you’ll see like just kind of a nice down, and then you continue in a new level, or an up, and you continue at that new level. It’s mostly been some fluctuations on a few client sites. One thing also to note, Is that we had a spam update get launched at, I mean, almost simultaneously over top of each other.

That’s right. So, we’ve just come off the heels of an August update of an HCU, a helpful content update, right into a double update, which makes it very hard to analyze. Somebody asked, somebody was Google, probably John Mueller, probably somebody online, How do I know which, which is which when I’m analyzing my sites?

And um, I thought I’d just share this. I’m paraphrasing, I’m not reading it, because we have too many tabs open as it is. But basically it said, if you spam, uh, and you suffer negatives, it’s a result of the spam update. If you don’t spam, and you suffer negatives, it’s from the core update. So, there you have it.

You heard it here first. Very 

Spencer: clear answer. So if you have any questions, just, you know, go through that thought experiment, Venn diagram there of sorts. Thank you, Google. Uh, yeah, so, um, so anyways, that’s about all we have to say from, for, for the broad core update. Just know that it’s in the middle of it. It might be done by next week.

We might have a little bit more to say, uh, once kind of the, the dust settles a little bit, but as Jared mentioned, two updates going on at once, so, um, not a lot of specific data we can share. However, uh, the helpful content update was a huge one. Um, so many sites within, uh, the niche site. Uh, industry, uh, affiliate websites were hit by the helpful content update.

We did a whole episode on that a couple of weeks ago, uh, but we do want to dedicate some more time to this because, um, there’s been, um, people finding out information, doing studies, coming up with an analysis of how you might be able to recover from that. Um, and so we’re going to share one of these. Um, this is from, uh, Eric Lantries on, on page.

ai, uh, and I’m going to share my screen. This is like, if I scroll through this article, like I’m just, yeah, it took me about 10 seconds there, uh, just to scroll through, uh, the article. So it’s, it’s very long, it’s very in depth and there’s a ton of information. Uh, so how do you want to tackle this, Jared, what do you think?

Jared: Uh, I think the best way for us to handle it on the podcast is maybe to skip down to part seven and or part eight. And by the way, you’re skipping so much. It’s so, it’s very worth a read. It’s a long one. It’s very worth the read. It’s based on analyzing thousands of sites. So there’s more of a data heavy study.

And it is looking at what the, the really what, what, what was identified and isolated. And moving up and moving down. Now again, a reminder that the Helpful Content Update is actually not a, uh, it does not reward your site. You’re not, you’re not supposedly able to move up from it, you’re only able to go down from it.

And so if you did go up as a result of the Helpful Content Update, That presumably is because other sites that used to be ahead of you went down. So, a bit of a, a detail. Uh, but, at the bottom of the article, the most intriguing thing I thought that he shared is, some trending things that he found, you know, and, and uh, uh, some of them are similar to what we talked about.

Spencer, uh, two weeks ago. And then there are kind of a few new things that we could point out as 

Spencer: well. Yeah. Did you want to go to seven or eight? Uh, 0. 7 or eight. Um, I think 

Jared: it’s number seven there. Yeah. 

Spencer: Okay. Analysis and trends. Number seven here. Yes. Um, but yeah, like you mentioned it, it’s like a site wide, uh, penalty.

Right. And so if you, um, further up, I, I like how he kind of explained that, um, I don’t know if I’ll be able to find it. Cause it’s so long, but, uh, basically, um. If your site, you know, was ranking at number one, sort of your optimization score, your Google, you know, optimization score, uh, is, you know, at a certain point, the helpful content update is like a multiplier to your site.

And so if your site overall was hit and say, okay, we’re going to just apply this 0. 7%. So it’s not a hundred percent helpful, right? It’s like 70 percent helpful. That’s going to, they multiply that by your optimization score and all of a sudden your rankings. Can just drop because of that. Your overall site is not found to be helpful.

So you can be optimized perfectly. But if this helpful content score, uh, is that multiplier that’s like it, yeah, 70%, 60%, 50%, like everything just comes down because it’s because of that. Yeah. Right. Um, so yeah, what do we see here? Analysis and trends. 

Jared: Um, there’s 20 something of them. I can highlight the ones that I think are unique that we haven’t shared again.

You know, if your site was hit by the helpful content update and you missed our, uh, podcast news podcast from, I think two weeks ago, right, Spencer, that’s the one you’ll want to go back and watch like it, it, it turned out to be very popular, um, and we put a lot of effort into it, but some things maybe that stood out from Eric’s analysis that we didn’t discuss, um, one of them is, uh, shorter focused content consistently ranked better and in depth, expansive content consistently declined in this update.

Um, and another thing that I thought was very interesting, new sites were mostly spared by this update in theory because Google hadn’t had enough time to crawl them and accurately assess if they were helpful yet or 

Spencer: not. Interesting. Yeah, that’s very interesting. And yeah, the fact that overall word count decreased slightly.

Um, I think he even mentioned that that’s the first time he thinks he’s ever seen it in recent history. Like maybe in the last decade as they’ve analyzed things that actually shorter, uh, articles are ranking a little bit better. And now, when we say shorter, right, it’s still, I think the average was something like 1, 400 words.

Um, for, for the average article ranking on the first, you know, page of Google or something. So it doesn’t necessarily mean that, um, you know, super short, you know, we’re not talking three, 400 word articles, um, are ranking, but there was a slight downtick in the length. And I, I found that super interesting too.

Jared: AI content. I’ve seen this in many places was mostly unaffected and yet so many people are sharing, uh, about the effects of not having unique content and seeing decline from that. Like. My content doesn’t have unique perspectives, it doesn’t, you know, I’m seeing that content that doesn’t have that declines and yet we’re seeing AI content when it’s a A site that is all about AI, like not declines, I think there’s some very real discrepancies there that no one’s really maybe been able to get their finger around or their finger on the pulse yet.


Spencer: Yeah, exactly. Um, I found that interesting that, um, you know, I have sites that have AI content and the AI content is doing just fine. Um, 

Jared: He had a couple of theories he went on to share in part eight. Um, and, um, I’m not really sure about any of them, to be honest with you. Now, no one pays me the big bucks to analyze these suckers.

I tend to just read and, you know, but I have to say I’m not sure about his theory number one, his theory number two, and his theory number three. But I’m also not sure about the opposite of it either. So I think it’s a very healthy perspective to think about. One of those these views do not reflect the views of the podcast host necessarily, but you know, hey I would like to get your opinion on him.

I don’t know if you want to yeah a little time talking about him 

Spencer: Yeah, absolutely. And I you know, I’ll just sort of echo that thought is that even Eric, you know, he says these are theories He’s not necessarily saying, Hey, I know everything. He’s just saying, based on this data, here’s some theories and, uh, they may or may not be, you know, right.

He’s just kind of throwing out his ideas. And so, yeah, I think it’s great if we can have some back and forth about, Hey, this doesn’t quite jive with what we’re seeing, but, um, yeah, so he’s got, uh, three, three theories, uh, here. that, uh, we can 

Jared: cover. So, theory number one, and, you know, again, Spencer and I are just coming in, coming in hot in all this here, uh, but, you know, basically this idea that the August core update was almost like a first pass in lieu of the upcoming helpful content update, with the idea that Google’s trying to find a way to minimize how many pages they crawl on the internet.

There’s just too much To crawl nowadays and isolating websites in the core update that might be problematic for further analysis would then just allow them to kind of semantically be assumed to get a helpful content negative classifier. Um, and so basically by pre qualifying billions of pages as helpful by virtue of.

where we can talk about this. By virtue of having recurring visitors, Google wouldn’t have to process the world’s largest websites with their AI classifier. This could explain how the August Core Update and September Helpful Content Update worked together to process the web in a more efficient manner.

The August Core Update refined and increased, I believe, I should put that in front of it, I believe, uh, Eric saying this, that the August Core Update refined and increased the weight of these user metrics, direct visitors and recurring visitors, which is why we initially saw an increase on forums, Reddit, and Quora.

And on and on and on. Um, Interesting to think about recurring visitors as a metric. Now that in theory sounds very much in line with some of the things we’re going to be talking about coming up here in the news, right? The result of click data and that coming out, um, user generated. So, very interesting, right?

Spencer: Yeah, I would say it’s very interesting and there’s, there’s kind of, you know, two parts to that. The initial part is, uh, where he kind of says, Hey, we think Google needed to sort of filter out a bunch of sites so that it was easier to crawl, uh, the web. Um, I don’t know how you feel about that. That seems…

Kind of an interesting theory. I mean, I think about Google, this massive company, they’ve been crawling the entire internet. Um, are they really struggling with space to be able to crawl everything, uh, quicker? Uh, we know they’ve had indexing issues over the years. That’s true. That’s true. Um, I suppose they haven’t.

And Google has said, uh, that, um, A lot of articles are never not indexed, even if they’re great, right? I there’s been some talk about that. So, um, so it’s an interesting theory, um, you know, to, to sort of say that. And then, um, and then, yeah, you, you also get, um, the, the other side of the coin that the, of this theory about, you know, People doing, um, sort of navigational searches, recurring visitors, right?

If that’s happening, it must be a helpful website. What do you think about that? I, 

Jared: I, I think in general, this falls in line with a lot of the other things that are getting said, brand have a brand. What does the brand mean? It means that you have a brand, right? People will actually know your brand. They’ll search for your brand.

They’ll look for you. This is all building upon a lot of the other elements we are seeing. In the HCU. So there’s certainly something to say, like, maybe, yeah, certainly, maybe. I mean, um, it’s a very detailed analysis. I don’t know if I would feel that sure about all the components in there, but I can’t argue against it.

Spencer: Yeah, and then, you know, here in red is sort of his finding of, hey, how do you recover? He says, I believe that one way to recover from the update is to improve your site’s metrics by increasing the quantity of navigational queries. Uh, queries seeking your website. So people type in a niche pursuits, right?

We’ll find nichepursuits. com, right? That would be a navigational query. Um, alternatively, the other way to recover would be to enhance each page of your website so that the AI classifier deems the majority of your pages as helpful. 

Jared: So let me give you an example of this I saw in the wild recently. Um, uh, Nathan Gotch, who is on Twitter and, um, has been publishing, you know, content around SEO for a long 

Spencer: time.

And we’ll be on the Niche Pursuits podcast, uh, less than a week from today. Oh! 

Jared: A little reveal there. you go. He will

be live next, next week, right? I didn’t check. Next week on Wednesday. Next week on Wednesday. So there you go. Here’s something he’s doing, and I think it plays nicely into what Eric’s talking about here. He will release. A YouTube video and he’ll tweet about it and he’ll say, Hey, here’s the video. And then in the comment he’ll say, go over to YouTube and search Nathan Gch.

And it’s the first video you’ll find. Mm-hmm. Yep. That’s an example of him playing the algorithm to increase YouTube’s trust of Nathan Gch as a brand by asking his people to go search for his brand name to find the video, rather than just clicking a link. 

Spencer: Yeah, I think that’s smart, right? Um, it does, it sort of invokes this trust, Hey, people know this guy.

Um, and I would imagine that, yeah, Google search works the same. More people are searching for this thing, so, Uh, maybe it’s actually a helpful website. Okay, whew! So that is just one theory, right? There’s two more. There’s two more 

Jared: theories. Should we dive into it or should we leave those there, I think?

Maybe we’ll tease and let everybody else kind of go. 

Spencer: Yeah, let’s sort of, you know, generally review what these other two theories are and, um, see where that gets us. 

Jared: What’d you think of theory two, narrow versus wide scope articles? 

Spencer: Uh, so… Uh, I, I think it’s kind of generally, um, the way we’ve always thought about niche websites in a, in a way, right?

The theory has always been, hey, if your website is focused on a very specific niche, like it, it should do better. That hasn’t always played out right. Because we always see these general sites like Forbes or you name it that do well on all topics. And so this takes it like at a more microscopic level, right?

Within the page itself, like, okay, this article is very narrowly focused on, uh, Fleece ferret hammocks, right? One of my favorite niches ever that I went into you know instead of just like Much more broad, right? It’s 

Jared: an example of how to clean a bathroom floor and within that article you go off topic Discussing the history of bathroom floors.

This could potentially be negatively impacted by the helpful content update. Exactly, 

Spencer: right? Yes, exactly 

Jared: And we’ve seen this we’ve talked about it in the podcast before with recipe blogs, you know, and it’s like Hey, you want the recipe? And by the way, he’s not talking about where the answer is in the content, but just how much of the answer is actually in the content.

With recipes, we talk about both, where it is in the content and how much of it is there, you know? Like, I need a chocolate chip recipe. The recipe takes up 200 words, and instead I get a history of the family tutelage of Grandma and Great Grandma and all that, you know? And so, how authentic is this and how relevant is this?


Spencer: exactly. So, staying on topic is, is a good thing, right? I mean, that’s sort of the, the end theory here. It’s like, do your best to really focus on the subject at hand. 

Jared: Theory number three, which he says is the least likely, um, uh, of the scenarios. But basically, like, your website gets classified as a certain type of website and then judged accordingly.

And, um, Uh, I might be over summarizing that a bit, but after categorizing the web, you could reduce the visibility of every single informational website that is below a certain user metric, for example. Like, if you were an informational website, you got categorized there, you’re going to be judged differently than if you’re a, uh, job posting website, or a review website, or something like that.

Spencer: Exactly. Yeah, so, uh, I, I think it’s interesting, right? I, I don’t know if it provides us a whole lot of actionable data. Um, to go off, right? It just, um, make, make your website helpful, engaging, make people stick around as long as possible, no matter what type of website it may be, right? 

Jared: I think we’ll see a lot more of these types of analysis and articles.

I think we are also seeing that it’s hard to determine, you know, a lot of factors went into this, right? Like long gone are the olden days where it was like, Panda’s out, we know what it is. Penguin’s out. We know what it is. Oh, you know, um, even like the medic update. Well, we have a pretty good idea what this is.

You know, those days are probably in our rearview mirror to some degree. 

Spencer: Yeah, I mean, just this one update is, uh, you know, as I scroll through this article. Of course, these are all theories and data and… You know, we don’t, we don’t even know if this is perfect, but he’s doing his best. Eric here is doing his best to analyze this update.

I think he makes some great points just in general about, um, great SEO strategies and how websites should be built. Um, but I think it’s going to just get more and more difficult in the future to pin down like, Oh, doing this will help me have, you know, uh, the helpful content. Content update won’t impact my site as much because I did XYZ.

Jared: One other thing you can really get, I thought, from this article that’s nice is if like maybe you’re a little newer to SEO and SEO is more about keywords and more about word count and these sorts of things. If you want to get like more into the mindset of how algorithms kind of work from just a binary level, I think Eric attacks the helpful content update data from, from that standpoint.

And so it kind of gives you, Oh yeah, that, that would be how an algorithm might think about that or might look at that or how a bot might. It might run through that, so it’s kind of, um, you know, very nerdy, very techy. 

Spencer: Alright, well Jared, we’re, uh, we’re twenty minutes in and we’ve covered topic one. Can we get a 

Jared: break now?

I mean, I feel like I need a break for the next seven topics we have on the agenda. 

Spencer: No, no break. Uh, we go until I hit stop. Uh, so yeah, we’ve got… We’ve got another, uh, yeah, six or seven sort of news topics. So we’ll probably make them a little quicker, but, um, we’ll roll with the punches here, how they come because they’re, they’re all kind of big topics, uh, to be honest.

So, uh, the, the next one, uh, that we did talk about. Um, Google is facing an antitrust, uh, lawsuit. Um, and we, we talked about this, you know, whenever it started three, four weeks ago. Um, and we haven’t done a lot of updates just because it is ongoing, but people that are following this case, um, are finding a lot of interesting tidbits about how Apple, Google other search engines all sort of interact and and do business with each other.

Uh, and of course, that’s where Google and Apple are getting into a lot of trouble as hey did Apple make Google kind of the default Search engine how much did they pay to do that? Did that squash all the competition? That’s kind of the crux of The case, um, but recently just, uh, that came out, this is just, I don’t know, a day or so ago.

Um, I guess it’s been almost a week. It says this article came out, uh, that Apple at, uh, Oh, over the years, um, a couple of different times has been in talks with Bing and DuckDuckGo to. Uh, either use Bing, uh, as their default or acquire Bing, right? So Apple had, in the past, has considered acquiring Bing or, uh, or DuckDuckGo.

And so this is unsealed, uh, testimony in the Google antitrust case. I just thought it was really interesting that, uh, that Apple at one point considered purchasing DuckDuckGo. 

Jared: Not only that, but they analyzed Bing, and it came back, they evaluated Bing’s search quality results, search result quality against Google’s.

While Bing apparently generally performed worse, it tied with Google for desktop English searches. Oh, 

Spencer: is that right? That’s what it said, I mean. I missed that one. That’s interesting, yeah. Yeah, um, and so yeah, they basically came back. Oh, they analyzed the results and decided, you know, we’re gonna just go with Google.

Jared: And I, what I, what I thought was, as I was reading this article, um, and I think it’s in there. Shoot, I should have the number in front of me. Maybe you can find it. Um, I’ll try to find it while I’m talking. But the, um, you know, when you look at the amount of dollars that are on the table here. Um, and how much money is, is going around in terms of billions of dollars.

Um, you know, apple decided to continue on with using Google as the search, uh, uh, browser of choice for an estimated $19 billion 

Spencer: annually. Annually. 

Jared: I’m like, mm-hmm. , I’m blown, but hold on. Now, I did a little research here. I did have time to look this up because I was so blown away by that number. And then I thought, well, is that number really a big deal?

I looked it up. Uh, so Google reported a revenue off of advertising, which Google’s ads businesses is really funneled through Chrome of 147 billion in 2021, 2020. Sorry. Wow. So this is just over 10 percent of the revenue they generate. But 19 billion to have your, your search browser as a default. 

Spencer: And what’s super interesting.

You can kind of see how leverage is used here because it says internal Apple emails revealed that during, during the trial suggested that the company Apple was leveraging being as a negotiating tactic to extract more money from Google. Uh, and during his testimony, the. Chief of advertising at Microsoft, uh, confirmed that that was the case.

So they’re really like, they’re getting a ton of money from Google. They, they tested out being, and they, you know, um, did a lot of negotiation with being, but really it was just to extract more money from Google. So there’s some cutthroat business for you. That’s just how it 

Jared: works. Yeah. I mean, it looks like they did somewhat of the same with duck, duck, go and duck, duck, go, I mean, 20 meetings and phone calls.

That’s not just a, uh, A cursory check in, you know, to check the box off with the board or 

Spencer: something. Yep, exactly. And this has been recently with DuckDuckGo. I mean, they’re very privacy focused. Sort of, one sort of chink in their armor is, um, you know, the same thing that Apple discovered is that he expressed, they expressed concerns over DuckDuckGo’s reliance on Bing for search information, suggesting this could compromise user privacy.

And I know it’s been maybe a couple of years since that kind of came out as a big sort of hit against DuckDuckGo is like, well, you’re, you’re using Bing and there’s some privacy implications there. Um, and so Apple discovered that same thing and maybe that was one reason among probably others that they didn’t go with DuckDuckGo, but boy, what a payday that would have been for DuckDuckGo to, uh, have been acquired by Apple.

Jared: It’s just, I mean, I cannot wait. It had to see how much more we’ll learn from, you know, the antitrust. I mean, we don’t know which way that case is going to end up going for Google, but just what we’re able to learn along the way has been fascinating, even, you know, without the results, uh, 

Spencer: withstanding.

Exactly. Yeah, we’ve been learning a lot of interesting things about Google. That kind of takes us to our next subject here. Now, was it through the same antitrust case that this story originally came out? I believe so. 

Jared: I believe so. I was asking Josh last week, like, where the heck are they getting this information?

But this was from Wired. We talked about it last week in the podcast about how emails between Google’s ads team and search team had been revealed showing that the ads team was behind on their quarterly quota and needed a little help from the search team to make the ads do better. Uh, I’m paraphrasing.

Um, and that’s what Wired reported. Well, guess what this week, folks? Wired pulled the article down. It’s no longer 

Spencer: around. Yeah, and I’m showing this screen. I just, I went to the article URL, and all there is is an editor’s note that says, After careful review of the op ed, Uh, how Google alters search queries to get at your wallet.

Um, and relevant material provided to us. Yep, following its participation, Wired editorial leadership has determined the story does not meet our editorial guidelines. They didn’t deny any of it. They didn’t necessarily deny any of it. Um, and we don’t know, yeah, we don’t know all the reasons that it was, uh, Pulled down, but it is interesting that you know, you guys covered this whole story and then the the article got pulled And so we wanted to mention that that hey We don’t know which part of the article was, you know on point in which of it wasn’t gotta be 

Jared: careful here Spencer Careful, you’re who’s gonna happen to you know, we cover too much of this stuff.

We’re gonna get in hot water here 

Spencer: We’re gonna be testifying before Congress Podcast 

Jared: Spencer you were not present with 

Spencer: Josh and Jared That’s right. Where were you? What happened? Um, so, anyways, I think that was noteworthy that, uh, we should mention that, um, since we had covered that story. Um, another, uh, sort of update, uh, with, with Google, again, of course, is, uh, SGE, right?

So we, boy, how many episodes have we talked about SGE, uh, in the last… Six months, something like that. Uh, a lot. Um, and, uh, so, there’s been a few people, including Glenn Gabe, on Twitter. And, um, you know, a lot of other people that have kind of analyzed what’s happening to the SGE results. I use SGE quite a bit, testing it out, kind of seeing what it looks like.

Thinking about, okay, am I going to get more clicks from Google? Am I going to get less? Uh, and how glad Glenn Gabe is calling it S. G. E. Light, um, or the incredible shrinking S. G. E. Right? Uh, because as Barry Schwartz tweeted here, Google is testing a light version of S. G. E. Right, which basically, uh, if you’re looking at this, you can see that the results are getting smaller.

Um, you know, it’s taking up less real estate at the top of the page before this was fully expanded. Right, and sometimes it was taken up like, gosh, almost the entire above the fold, uh, area of your website. And, uh, or of, of Google, of your browser. And, um, now they’re shrinking that. Um, and so it looks like there’s less and less real estate.

And, uh, why do you think they’re doing that? What, what, what are you 

Jared: hearing? It’s because it’s not working. I mean, this is only the beginning. This is only one of many things. Like, I don’t know, Spencer, if you’ve noticed, like on the vast, first off, you got early, you got access earlier than me to SGE. So I was sitting here every week, like watching you pull up all these cool screenshots and seeing what the future of search was and not able to do it.

Then I finally got access. And like you said, the majority of SERPs were getting big full layout SGE boxes and we were watching them and you can go back and watch podcast episodes. Where spencer’s showing the screen and it’s got like eight remember we were called the the super featured snippet, right the super snippet And we were saying yeah now there’s like eight spots to get in the snippet, right?

And then we were excited about the quotations. They were adding and now all of a sudden First off, the size of it’s reduced significantly. And second off, it doesn’t generate very often. It says, press this button if you want, and we’ll generate an SGE result. But it stopped automatically generating on a lot of queries as well.

And um, so, the other side, just to be balanced. You asked my opinion, I think it’s because it’s not working very well. But the other side could be, are we starting to see them find a way to make room for ads? You know, that could also be another theory there, so, you know. But boy, it certainly has changed dramatically and quite rapidly.

Spencer: Yeah, definitely. And I would call this good news for, for SEOs, right? Like, Hey, uh, maybe there’s some less uncertainty about the future. Maybe SGE is not going to be taking up as much real estate or taking as many clicks, right? I’ll share my tab. Um, I just did a search for best smartwatch. Right. And you’ve got tons of sponsored ads, uh, at the top.

And then it did generate an SGE experience, um, SGE result. It’s like two sentences is, is, is all there is. You have to click show more. Right. And then you’re, you’re already into the SERPs, right? Um, so that is significantly less real estate. I wish I could remember some of the. Um, queries that we had done in earlier episodes, right?

So I could compare them. Oh, you had 

Jared: like a best Bluetooth speaker for something, and yeah. Yeah, 

Spencer: maybe you’re right. Um, 

Jared: best… Like an outdoor party or something like that, wasn’t it? 

Spencer: Uh, there was… something. Look at that. Best Bluetooth speaker doesn’t even… Generate. 

Jared: That’s what I was talking about just now.

You have to hit the generate button. 

Spencer: Yep. Yep. Interesting. And then even when you generate, I think it’s gonna be the small… Well, we’ll see, actually. Um, I haven’t done a ton of these. Who has time for that? Okay. When you hit generate, it generated the full SGE, 

Jared: right? That’s what we’re used to seeing right there.

Wow. Good grief. This 

Spencer: is bigger than what we’re used to seeing. You know, this is, this is, yeah, in our early testing. Um… That’s like the one 

Jared: style, like the first week you got access right there. 

Spencer: Yeah. Let’s do an informational query. Like, how to… How to clean something. How to clean, um, A TV screen. It popped up on my, um, okay.

It is generating something. Um, here, of course the, I mean, okay, that is much smaller and significantly. And if you saw this, what I’ve always complained is how slow it is, right? Like if I weren’t on the podcast, I probably would have scrolled through and already clicked on this first result. You know, the old school snippets, um, this took, you know, a few seconds to generate.

And so, 

Jared: What happened to all the images that were always present on the right hand 

Spencer: side? Ah, I gotta, I gotta click the open button for those to get generated. Okay, there they are. You don’t get 

Jared: those unless you hit open. Which, that had a visual appeal to it that did kind of engage you and draw you into the SGE, 

Spencer: right?

Mm hmm. Exactly. So, this is, uh, very, I would call this very interesting, uh, because, boy, if they weren’t getting the results they wanted with this huge real estate, right, of taking up the whole above the fold, like, I don’t think they’re gonna see as many results, you know, with this just small. Right? And so we don’t know exactly what metrics they’re looking for, but they’re testing things out, changing 

Jared: it.

It kind of feels like a company going big and having a big ad spot at the Super Bowl. And then the next year, just, um, you know, basically putting up a little, uh, like a little, a little, uh, a little logo in the corner. That’s, you know, says this, this, this, uh, the Super Bowl is brought to you by one of 700 sponsors.

And it’s just your little logo there. You know, it’s like it. It feels like a big retreat in terms of, you know, uh, you know, taking everybody back all the way, like Google’s an ad company and so they’re trying to figure out how to continue to make money by delivering more and more of what people want. And it’s this careful balance between giving people what they want so they keep coming back and generating money.

And if you are, you know, depreciating these new features, it’s, uh, it’s hard to imagine it’s not because people are, are, are loving it, right? 

Spencer: Yeah, you know, I think about it kind of how companies make these huge decisions when there’s a new wave, right? There’s a new technology. This case, of course, is with AI.

ChatGPT just took the world by storm. And Google said, uh oh, we better respond and respond quick. And their quick response, and kudos to them, right? Is this amazing technology that can generate results. But now, maybe they’re dialing it back a little bit. And I do think about… Uh, other companies, notably, uh, Metta Facebook, right?

They changed their name to Metta. You may, because of this whole NF, NFT, um, you know, virtual reality, metaverse, Trend, like it was hot for like six months and Mark Zuckerberg said we’re going all in. This feels good. We’re changing our name to Metta and now it’s like it’s people off. They keep laying people off and it’s kind of like a laughingstock is because like, oh man, if you bring up NFTs and the metaverse, like it’s not a thing, uh, anymore.

And so, um, you know. Anyways, it’s, it feels somewhat akin to that as like a company makes this huge dec decision, quick decision based on a, a fast growing trend. And, um, the, the, uh, AI trend is not dying, but maybe they’re realizing, oh, you know, it’s, it’s not quite as great for our users as we wanted it to be.

So something to keep your eyes on. They gotta, 

Jared: you know, they gotta, they call it in business, they call it product market fit, right. And you just went through how it doesn’t fit the market. Well, they didn’t even figure out how to get their product into it, which is ads. Like they got to pay the bills in the meantime, you know, so like there’s a lot of things that I don’t want to say they aren’t working but boy talk about rush of blood to the head Like they didn’t really get around to Even figured out the market fit let alone how they’re gonna get their product into it.


Spencer: no no ads on this query Although I don’t know if how to clean a TV screen usually had ads sure But sure every query should have something right? They got ads everywhere Exactly. They make what did what did you say? 139 billion 47 billion in 20. Wow, so they know what they’re doing and if SGE Cuts that by even five percent, like the market will not respond well.

Yeah, remember Spencer, 

Jared: we’re supposed to reduce the number of ads we have on our site in order to rank, but Google can continue to cram ads all over the SERPs. 

Spencer: That’s right, well because they’re scraping our content to put the ads, right, in their search results. Oh man, now 

Jared: we’re definitely going to get something from Google for this podcast.

Spencer: Come on Google, we’ll testify before Congress, bring us, uh… As long as they get a backlink from it.

Jared: Oh, man. Did you have another SGE topic? 

Spencer: Uh, I feel like we did. I think it’s at the bottom. Um, oh, here we go. Here we go. I did. Um, so this was actually trying to keep us on target here. Thank you. What were we talking about again? Um, so yeah, I’m sharing this article from the verge and this just came out today.

And I believe this, this announcement just came out today that. Google’s AI powered search experience can now generate images. Uh, and so, again, the same SGE experience, so they’re not killing it completely, but, uh, they have added the ability to make images, and it says, and written drafts. I have not had time, uh, to test this out.

I actually did try and do a query, but it didn’t generate an image for me. Um, so maybe I, I’m not seeing that yet. But they did have this, um, gif that, you know, they basically show how it works. It says, draw art image or draw an image, um, of this, uh, this guy with a chef hat. And look, he looks good, you know, that’s how it is.

Um, I feel like I’d try it again live, but I’ll, I’ll probably get it all wrong. But, uh, so this is rolling out. Um. And so, yeah, it kind of cool that, uh, you don’t necessarily have to go over to mid journey or whatever image generation tool you can now do it potentially directly in Google. They’re at least testing this out.

Anything to add to that? 

Jared: No, I mean, okay. Yeah. No. Yes so much, but no. 

Spencer: Yeah, kind of cool. Play around with it if you have access It also says and again, I didn’t test it out But it says and can do written drafts written drafts of what I’m not quite sure Exactly. I’m just seeing the image generation itself so image has 

Jared: been Getting its way into AI recently, we haven’t had a chance, it’s been on our agenda for like four weeks, but it just keeps falling to the bottom, like ChatGPT Vision, which is the ability to upload a, uh, image, um, and then have it interpret results, almost the flip side of it, this, what Spencer has on the screen right now is Google SGE creating an image for you, but then ChatGPT launched the ability to kind of read images and understand what’s going on in them, some fascinating results were shared by, uh, Thomas, uh, who had been on the podcast before on his Twitter account, where he had it read a, Uh, basically like a poster.

for an event in the area, and it pulled out all the semantically relevant details. So, images, I think we’re going to be talking about images a lot more in the coming weeks. It’s something to pay attention to, not only in image creation, but in image interpretation. Yeah, 

Spencer: exactly, great points. Um, okay, so let’s move on to the next, uh, news topic here.

This is also information that came out of the Google Antitrust. Lawsuit and this came out a little while ago about a week ago. I think again in some leaked documents um that uh That google actually uses clicks In the ranking data, right? So for years and years, when people asked Google always said, Hey, we’re, we’re not using clicks in ranking data.

It’s all about relevancy. It’s all about how helpful the content is, how optimized the page is, right? You know, it doesn’t matter if people come and click one website and then go to the other website and the result turns out they actually are, um, they actually are using clicks to help rank, uh, websites, right?

That could be, um. You know, click through rate, right? The more people that, the higher rate of people that click on a result, that might help it improve. Um, it could be what’s called pogo sticking, where you go to one result, one website and then you come back to the Google results and you click on another one and you might do that a few times and then, you know, the theory is that the website you end up on and stay on the longest, that must be a better website than the other two that you came back from, right?

Um, so there’s all these things And that has implications because that’s kind of like, it, it tows the line of, of privacy data, right? Of like, okay, is Google tracking individuals? Why do they know so much information about individuals? Uh, is that really private data? Right? Um, but it, the, the big picture here is again, that Google is using some sort of click data in 

Jared: rankings.

Supposedly the data’s tokenized, which means that it strips all of your personalized information out of it. Um, Also, like, this is one of the, one of the few things I think all of us SEOs can agree we’ve known for a long time, even though it’s never been confirmed. Um, classic example, right? Like, you’ll see an article of yours that was on page two for a while, or maybe a newer article of yours, kind of go into the top page, uh, the first page results of Google, bounce around a bit, and then bounce back.

Um, you know, and, and, It’s kind of like what did Google get like the analysis of my content wrong and these are signs that like, you know, user metrics play a role in it and obviously we’re seeing User metrics play an increasing role. The helpful content update seems clearly targeting poor user experiences, which creates bounce backs, creates shorter time on page, creates less engagement.

We’ve seen them put Core Web Vitals at the front, using data from Lighthouse that they’re able to understand and report back to us. So, this is no secret, but it all makes sense when you look at it from the greater, uh, greater, uh, you know, the large scenario. And then it does give us some new things. I think the Pogo sticking and just some things we kind of had thought about, like, that they obviously pay attention to.

Spencer: Right. And I should give a shout out here to Kevin Indig. This is a blog post on his site. I don’t know if I mentioned that. Um, so you can check that out, um, the implication of Google using click data and ranking. Um, and then he poses the question, how does this, um, change the way that we operate? And, uh, just a couple of the bullet points here that are, you know, interesting to think about is to pay more attention to design and user experience since it does impact whether a user comes back or not, uh, keep optimizing snippets to draw even more clicks, uh, invest in brand awareness.

So users recognize domains and click on our domains in search, keep people on our sites with meaningful engagement and provide outstanding user experiences. So maybe not. Anything new per se about what you should do, but yeah, but reinforcement and just how important that truly is 

Jared: We’ve been saying for a while a lot of times We’ll say all of us are I’ll say guilty of it I use air quotes because too much time.

I don’t have a time to explain my reasoning but Well, we’ve said like a prioritized content. Don’t worry about your brand. Don’t worry about the logo. Don’t worry about the color scheme. Like get the content out of this most important thing. And I, you know, those, those days are this stuff, that’s this stuff’s all starting to kind of coalesce together rather than being as clear cut as maybe it used to be.


Spencer: Okay, I’m taking a deep breath because I think we got through all the news topics. 

Jared: I feel like I need 

Spencer: a nap. Yeah, you can have one in about 15 minutes, Jared. You gotta stick around. Alright, we 

Jared: got, we do have to cram in segments two and segments three in the remaining 15 minutes we 

Spencer: have. Yeah, yeah, and uh, you happen to have a spreadsheet that is excessively large, um, that you may be talking about.

So, so good luck. I forgot what’s next. So… Exactly. So, let’s talk first about our shiny object shenanigans, uh, just a little bit. Um, I, uh, I think I’ve talked about this in the past, uh, if you go back, uh, several months. Um, so I have a website that I created a long time ago, um, with my two oldest children. Uh, it was meant to be like a summer project for them, and some way that they could earn money.

They were excited about it the first, uh, Couple of weeks. Uh, and then they lost interest and suddenly I was left with a website that was sitting there for a long time. But I liked the idea so much, I just couldn’t let it die. And so, um, I’ve built up, uh, the Facebook page to 30,000 followers, maybe 31,000.

Can I, 

Jared: uh, sorry, can I have a moment of levity for a second here I suppose? Sorry, I’m gonna just poke funny in. So just full clarity, like. I, I just, you know, we, we create this agenda throughout the week, and so I just peeked at the agendas that are about ready to start, and so in there, under Spencer’s Shiny Objects and Anikens, it says, Spencer.

Facebook news website, colon, started with children, abandoned, picking up the pieces. It’s like, what happened to the children? What’s going on? They were abandoned? Are you picking up the pieces from your children? It just sounds like My children are gone. Such a terrible story. I’m relieved to now know it has It has to do with the website because I did not understand it for a 

Spencer: second.

No, yeah, I was not the authoritarian boss that totally destroyed my children’s willpower for life. Abandonment. 

Jared: Okay, I had to digress. I thought that was too funny. So you’re getting Facebook 

Spencer: followers. I didn’t even read those notes. Yeah. Um, that was supposed to just remind me what to talk about. Um, I understood it.

That’s what matters. Um, so yeah, this website was originally created to get traffic from Facebook. It was the whole Facebook theory, let’s get traffic from Facebook, monetize with display ads, and maybe over time get this to pick up in Google Discover, to get it even more picked up, because it’s that kind of topic that has posts that can go viral on Facebook and Google Discover.

It’s not a traditional SEO play, although I’m going to add that. If this thing ever starts making money. Yeah, you did 

Jared: mention this. You just kind of teased it. You were going to try to reinvigorate or something. 

Spencer: Yeah, exactly. And so, I finally, um, I finally just today did a call with my writer, um, that is working on the website.

And she, I, this has been such a side project. Um, that I’ve kind of just been letting her run with it. And usually when you just let people run with things, they tend to fizzle. She is like taking the bull by the horns. And I like feeling so bad for like, just sort of letting her run on their own. And I was like, we got to do a strategy call.

Like you’re amazing. Uh, and she is like, she is just crushing it with finding content, interacting on Facebook, growing the community there. Like, okay, we got to figure out how to like make this thing work. Uh, and so I did that today. The current stats, um, it got about 27,000 visitors in the last 30 days.

Almost all from Facebook. Yeah, right. Um, but it’s only made like $148 in ezoic ads, uh, in that period. Low mvs, super low, super low. Um, I think it would do way better on Mediavine or somewhere else. Um, but it is like a low value niche. Um, so I need to like, the traffic here before it starts to get interesting.

Um, and I, I think it will, right. I kind of put a plan into place here, um, with, with my writer. She’s super excited. She loves the niche. It’s kind of got a humorous angle to it. I don’t want to reveal what it is. Um, it, it’s, it, yeah, it’s more, it’s more like news, things that can go viral, but this is like underpinned humor, uh, behind all of it.

And so it is kind of a fun, enjoyable, if this thing takes off, like I would love just kind of working on it and reading it and… Doing that sort of thing. And so, um, yeah, that’s about all I wanted to say. That quick update making 150 bucks a month. I should say I’m losing more than 150 bucks a month, but, um, that much in revenue.

Uh, and, uh, hopefully within the next couple of months, I can be make, uh, be getting over 50, 000 visitors a month consistently. And because I already have a site on Mediavine, I can get this second site added, hopefully, to Mediavine in the next couple of months and, and revenue should go up from there. So just wanted to share that quick update.

I’m still in the weeds, kind of, you know, doing all this thing, uh, with, uh, this, this particular, uh, niche site that is more Facebook, news, Google Discover, uh, oriented. So it’ll be fun 

Jared: to hear on this because I think a lot of us who own traditional SEO websites are trying to figure out how to maybe. Add these types of components onto our traditional website, right?

And even though you didn’t start with a traditional website, SEO focused in mind, you are succeeding in terms of the kind of traffic you’re driving to the site from social and in these sorts of signals there. 

Spencer: Yep. Yeah. I think the best month so far was right at 50, 000. Um, uh, In September or, or August. Uh, so I know the potential’s there.

Uh, I got a couple tweaks tricks up my sleeve that we’re gonna try and I’ll report back in, you know, a few weeks or a month and hopeful hopefully things have improved from there. Yeah. Okay. 

Jared: I guess we’ve gotta keep it brief. I wanna ask you more questions. Um, hey, so I took a break from the, uh, Amazon influencer program.

That has, uh, that’s been a slog. I haven’t, I haven’t seen a lot of improvement. Although we’d have those, uh, Amazon like prime day deals happening this week. So there were some uptick there. That’s positive, but still waiting to kind of get that Q4 breakout we’re all hoping for. So I did, I did realize this week that I have spent a lot of time, whether by choice or not.

I’m talking about the helpful content update. I mean, we’ve covered entire podcast episodes here on it. I’ve hosted a couple of spaces on Twitter, uh, which is their new kind of hangout place where we talked about it. Um, uh, you know, lots of analysis, reading lots of analysis in preparation, trying to make sure I can help my clients with this update.

And so, I sat down, and I went through all my notes, and I just went through all the things I’ve been looking at, and all the things that have been discussed, and I created a spreadsheet, uh, of, believe it or not, 124 things. 

Spencer: And can I share this on the 

Jared: screen? Yeah, yeah, share it on the screen, yeah, share it on the screen.

Um, it just, I just published it live yesterday, and I took a little bit of a different approach to this, because it’s not like, go do this, go do that. Number one, go to this. What these are are questions you can ask yourself about your website and then depending on the answer you give might indicate whether you need to go work on that or not.

Um, now this is 124 things, so, you know, I don’t know if any of these are or aren’t specifically related to Helpful Content Update, but um, what the spreadsheet does is, it allows you, it guides you in giving a score of 1 to 10 and ranking each of these questions and then it rolls everything up for you to help you prioritize, like, if, if, if you need to make adjustments and improvements, what to do.

So, it’s um, I don’t know about you, Spencer, but the number one question I keep getting asked is, what do I do? Like, what do I do? There’s so much that I should focus on or that I could focus on improving. Um, and so, I kind of started sitting down to do this to give it to my clients and then just decided to share it with everybody.

So, you can download it if you want and get access to it. It’s free. I don’t know, I said I’d make it free for the weekend. That way it’d give me some space if I wanted to down the road charge for 

Spencer: it because it did take a lot of work. So we’re on Gumroad where people can download that, right? Yeah, exactly.


Jared: even if it charged, it would be nominal and stuff. But it is free now, so I don’t know, should we just include a link for it and if people want it, go grab it. It’s a bunch of great prompts to ask yourself about your website as it relates to a helpful content update. 

Spencer: Yeah, we’ll, we’ll put a link to that in the show notes.

People can, uh, do that. Um, boy, a hundred and ninety something, um, you know, inputs. Or a hundred and fifty, I guess, uh, something like that. Uh, inputs. That’s a lot, um, to go through. But that’s the kind of detail that perhaps a website needs if they want to go through and, and recover, uh, some of this. So, um, boy, they’re…

I mean, typos, misspellings, punctuation, right? Um, content not useful, engaging. You can go through here and really dive deep, uh, into your website. So, boy, you’ve been busy. 

Jared: Yeah, I, you know, this is one of those things where, like, you sit down and you’re like, Oh, I’ll just create like a, you know, a nice little quick spreadsheet for my clients, because I want to help my clients to, um, you know, analyze stuff.

And then, you know, one thing leads to another, and before you know it, you’re like, Oh, I’ve got 125 of these. That’s a lot. Um, You know, and it’s worth saying, like, this, it can be hard to analyze your own website. So, you know, I’ve already had somebody email me back and say, this is great. Thank you. I gave this to my buddy who runs their own website and I’ve asked him to do it for, to do it for my site and I’m going to do it for their site.

And I was like, Oh, that’s a really good idea. Kind of get third party perspective. You do it for their site, they do it for your site and you kind of get a third party on it because sometimes it’s pretty hard to analyze our own content. 

Spencer: Yeah, exactly. So, uh, yeah, very helpful. Uh, this spreadsheet hopefully is very helpful to people.

Um, so, very good. Let’s, uh, let’s jump into our weird niche sites. We’ve only got a couple of minutes here. Um, we’ll try to go over them here pretty quick. I got my short story that I’ll tell about my niche site, I guess I’ll go first. But, uh, yeah, I was at the, um, uh, conference last week. And I was, uh, eating dinner with, with a few people.

And a couple of them listened to the podcast, which is awesome. And, uh, she said, Hey. Uh, the weird niche site section of your podcast is my favorite part. That is the best part of the whole podcast. And I was like, yeah, Jared and I love that. Uh, you know, it kind of brings a fat, fresh, humorous perspective.

And she said, you know, I actually have, I asked her, I was like, do you have any weird niche sites? that you’d like to share. She says, I actually do. Um, So I won’t give her name because she didn’t want to be publicly outed. Uh, but she said I could share this niche site. This is not her niche site. Um, But, uh, it is a good one.

This one is cakerex. com And, uh, it’s exactly what you might think it is. Um, These are actually… Uh, supposed to be professional cake decorators. You know, you go to Costco or you go to the grocery store, right, and, and you get a cake done for you? Like, these are all pictures of cakes basically gone wrong.

Cake fails is what we’re looking at, right? Um, so… It’s just, you know, and then they title it. They’re kind of, Rabid Pumpkins is this one, right? It’s, it’s a pumpkin with like white foam coming out of its mouth, right? And kind of squished up against the side. And you can just scroll through, um, and, and look at cakes.

Uh, there are so many. Some are, you know, uh, yeah, they’re, they’re just terrible. There was a couple here that, Like this one just kind of cracked me up. Um, you know, it’s, it’s supposed to be a big 15. Like, Uh, you know, 15 years old and the, and the cake decorator made it the big 15, super tiny, um, right. It just, it’s ironic.

Okay. This was thought it was supposed to take up the whole cake. Um, anyways, you can go through here, look at all kinds of cakes. They’ve got ads everywhere. So that’s how they’re making money on the site. I was just trying to find, uh, the other navigation. They do have 

Jared: a directory. We’ve been hot on directories ever since Tim joined the podcast last week, which is If you didn’t, uh, if you didn’t listen to that one, boy, that one, that one became very popular very quickly, but he talked all about building directories.

I see this one has a directory, which is ironic because it’s a site about terrible cakes, and then there’s a directory for where to get 

Spencer: cakes. Well, maybe these are vetted, right? Hey, you’re not gonna, you know, these are ones if you don’t want a cake wreck, um… Did you 

Jared: have a cake fail? Go to one of these cake, uh, cake places.

Spencer: Yeah, exactly. So yeah, they’ve got a cool directory. Maybe they’re making some money if they’re following Tim’s advice, right? They’re maybe charging for a listing there. Um, and anyways, you could go through. I mean, there’s like, Oh, that’s a tilting wedding cake. 

Jared: This is apropos. I’m going to a wedding later 

Spencer: today.

There you go. Get me in the mood. So, oh, that cake looks pretty good. Am I missing something? I think it’s tilted over. Oh, maybe it is. Yeah. Um, so anyways, uh, That one’s definitely a little tilted. Um And, uh, I believe that there’s the author has a book, right? You can buy a book on Amazon when, um, cake racks, when professional cakes go hilariously wrong.

I was back from 20, uh, 2009. So super old. So this website is old. It’s been around for a long time. And, uh, if we just share the AHRF stats, it’s not like it’s getting a ton of organic traffic, about 10, 000 visitors a month, it’s a DR of 59. Uh, but if we go over to SimilarWeb, it shows that it’s getting, uh, roughly 76, 000, uh, visitors a month.

Just kind of a fun niche site, making a little bit of money. And, uh, yeah, just something fun to look at. If you want to check out Cakewrecks, you can go to Cakewrecks. 

Jared: com. Ugh, man, I missed my calling. I was a wedding photographer for 12 years. Do you know how many pictures I have of cakes? Most of them done well, by the way.

But I have been there when a wedding cake toppled over before the reception even 

Spencer: started. Oh man. Yeah. And miss all the pictures. Yeah. Oh yeah. 

Jared: Uh, that was from a hot day. I’ve also knocked a wedding cake myself. My flash and my camera, I swung too fast. I took out a chunk of the cake. Thankfully it was before anybody walked in the reception, the caterer figured out how to plug that sucker up.

I have a lot of cake fail stories. 

Spencer: And since that day, since that day you’ve worked online, so you weren’t in person next to cakes anymore. Slowly I’ve moved away and away from those types of jobs. Away from physical objects, very good. Um, alright, what do you got for 

Jared: us, Jared? Today we’re looking at HistoryColored.

com. Um, it’s uh, you know, it fits a weird niche. Uh, nothing weird about it specifically, but I thought it was super interesting. Basically, this site started, and I say started because I’m gonna, I’m gonna expand on that. It started going back and colorizing old photos that were only black and white. Um, and you know, that’s like a thing in the movies, right, like Technicolor and like going back and adding color and whatnot, but, um, I just found myself a bit enthralled by some of the photos, uh, that they were able to go back and add color to.

I mean, you’re looking at the, the, the kind of the core one that I was looking at as well, which is we’re looking at the various presidents of the United States and they got, went back and colored them. Boy, we had some really ugly presidents, I’m going to go ahead and say, um, you know, not trying to, trying, trying to judge.

Could have just been the time of year, but look at that, you know, you can see how they actually took this look how bad it is, look what they ended up doing with it. Like, um, and the author talks about how they did that. So this is like a real personal project for this, um, this person. Now, they’ve kind of done the classic thing that we might recommend they do.

If you actually go over to Ahrefs and pull up HistoryColor. com, what started as a very clear… Yeah, you got a domain name. It’s all about colorizing photos of history, right? This is a very kind of specific silo here. But if you pop over, you can see that they’ve taken their initial success and really kind of gone the route we’d probably all do as niche website builders, right?

They started expanding into informational topics. Um, and you can look at that organic traffic as a result of this, right? If you kind of look like when they started going into more informational topics like history facts and specific facts about different history events, look what happened to their traffic.

And it’s not off the charts, of course, but you’ll see they started ranking for stuff, 

Spencer: right? Yeah, they were writing about interesting facts, uh, about World War II, historical events that took place in the 18th century, right? All sorts of informational articles now is actually bringing in, uh, traffic. And the majority 

Jared: of the traffic, because that’s where they search for, you know, I’ve never thought to go, like, search for a colorized photo of…

You know, uh, Jefferson Davis, uh, or whatever, you know, uh, Thomas Jefferson, or, you know, just name some figure from history, but certainly you’re always looking for like different, interesting tidbits about Winston Churchill or something, you know? 

Spencer: Yeah. Um, very interesting. And I have to wonder if some of these pages just go viral.

Um, I would think so, right. Either Google discover or social media or something, right. These colorized photos, um, you know, if they kind of revealed, Oh, a first time photo of. You know, name the person, um, you know, some, some president, uh, my mind’s going blank here, but, uh, yeah, you can just stick with Thomas Jefferson.

Right. Um, first time ever, right. That, that could go viral. I 

Jared: mean, look at this seven bizarre facts about the only emperor, the United States of America. I mean, if that wasn’t made for a little bit of a tweak for Google discover, you know, um, uh, the Google discover title would probably be like, did you know that America has had an emperor before or something like that, you know, or.

You know, so there’s some really cool things here. And again, kind of go back to like, obviously HCU is on the mind, right? Like, so it’s just, it’s everywhere I look and see and think and breathe. And, um, you know, talk about original content, right? Like if you’re the one who takes something. It’s already out there like a photo of, um, uh, Abraham Lincoln and then you colorize it like that’s a, that’s a new piece of content that the world’s never really seen before and you think about all the content you can make out of that and all the authority and expertise you can build off of that and all the combining you can do and it really starts to open the mind in this, you know, potential new world of HCU that we look at.

Spencer: Right. Um, it’s kind of interesting. Now, I didn’t look at the traffic. Um, that is happening on like similar web, for example, but I just went over to their Facebook page. They have 515, 000 followers. Wow, I didn’t look at that. Uh, and so they actually may be doing quite well on social media. That’s what I was thinking, like this type of site could do really well.

My father in law 

Jared: would land on this site and we wouldn’t see him again for a week. 

Spencer: Yeah, and so it, it looks like they are doing, you know, here’s Twitter. I don’t know if they have that many, not that many on Twitter, but Facebook for sure would do well. I wonder if they’re on, how well their YouTube is.

Yeah, they have 355, 000 subscribers on YouTube. Right, and so, oh, sorry. Yes. I’m now on YouTube. And, uh, so they have all sorts of videos about Oh, I mean that’s just pictures and voiceover Jerusalem in 1897 restored footage Oh, I’ve seen this video actually Um, 1897 restored footage I watched that one, it’s fascinating Yeah, they look all really short Um, to see colored, yeah They take, it looks like old video footage And they colorize it 

Jared: I mean, so history colored in photos, history colored in video 

Spencer: There you go.

That’s so cool. Pranking a gardener in 1896. Right? Um, anyways. Aren’t you glad we don’t have to wear those outfits anymore? Yeah. Um, I mean we almost gotta watch the end. Oh, he’s, he’s stepping on the hose so he can’t get it. Alright. Um, so people can check that out if they want. So cool website. So they’ve probably got a good YouTube channel uh, as well.

Man, making prank videos in 1896. How cool is that? Learn something 

Jared: new every day, Spencer. We learned a lot today, I’ll tell you. We 

Spencer: learned a 

Jared: lot. That was a crash course. I, uh, yeah, I don’t even know what to say. 

Spencer: Yeah, I mean, I feel like I was talking on double speed and we still, uh, took over an 

Jared: hour to that person who commented many months back about how they always listen to us at 1.

5x. I don’t think you’re gonna have to this week. 

Spencer: Yeah. Yep. We, we talked a little bit quicker for you. So, uh, tons of news topics. We went over side hustles. We did our weird niche sites. Hopefully people found it interesting. Thank you for sticking around to the end. Of course you can follow along at niche pursuits, uh, niche pursuits.

com slash newsletter where I send out my email newsletters. But like usual, you can give it a like, you can give it a comment, you can give it podcast. Just let other people know that this is valuable content. And with that, thank you everyone. Have a great weekend. Have a great weekend. Take care. Bye everyone.


Source link

Leave A Reply

Your email address will not be published.