How To Improve User Engagement (It's Becoming Google's "Truth Detector")

INTRO:
It's time for the SEO show, where a couple of nerds talk search engine optimization so you can learn to compete in Google and grow your business online. Now here's your hosts, Michael and Arthur.

ARTHUR: Welcome to another episode of the SEO Show.

MICHAEL: Well, we're recording, we're rolling, and we're rolling with that. Do you hear the way he speaks to me up here?

ARTHUR: Did you actually record that?

MICHAEL: Yeah, we're recording.

ARTHUR: Hello and welcome to another episode of the SEO show is what I wanted to say. Yeah, but you're too busy teasing me.

MICHAEL: But we're back for another episode. And today we're going to talk about something we haven't spoken about that much on the show because We've been banging on, you know, with all this AI talk last episode, our predictions were pretty much all about AI. We realized that last year and this year, our prediction was that user engagement metrics are more important for SEO, particularly for Google trying to figure out. where the content's good. It doesn't matter if the content's written by AI or a human. If it's gonna figure out if it's good, it's gonna be looking at user engagement metrics. Yes, for sure. So why don't we look at them? Let's do it. Why don't we chat user engagement metrics? You fired up? I'm fired up.

ARTHUR: I feel awkward because of the intro now. I'm curious as to what I sound like teasing you.

MICHAEL: Well, seeing as you never ever listened to the show, you'll never know. Well, I'll listen to this episode. So look, as we all know in SEO, SEO is just content backlinks. That's all it is. That's all it is. Or it's always what it always has been. But there's a rise in user behavior signals to combat AI. Because people like us are out there in the world churning out content right now. SEO people creating useless pages on their site with AI content.

ARTHUR: Well, I don't know about you, I'm not doing that.

MICHAEL: They're just churning it out in general.

ARTHUR: In general, yes.

MICHAEL: So not us specifically, but people like us. Same job titles. Yes. And Google has a problem on its hands because whatever you said last episode, 7 billion blogs a day or… 7 million a day. Whatever it was. Which is a lot, yeah. How does it figure out, A, what it should even index, B, what it should show to its users? Well, the answer is, of course, links, always. But outside of that, user behavior signals are going to be more and more important for telling whether AI content is any good, whether human content is really good. So what does that mean? We always say user engagement metrics.

ARTHUR: Yeah.

MICHAEL: But what is that?

ARTHUR: We need to break it down into all the metrics. We are going to break it down. We've got a few here. To the metrics that we think are most important. Because there's others out there, but we feel these are the main ones that you should focus on.

MICHAEL: Our metrics are better than the other metrics, aren't they?

ARTHUR: Well, they're not our metrics, they're the metrics, but we just have chosen the top metrics.

MICHAEL: Yeah, so our metrics are the only ones you need to care about.

ARTHUR: I think so.

MICHAEL: Don't worry about it. If someone's trying to come around and talk about other metrics to you, just say, listen, I know the best metrics, okay? Okay, let's get into the metrics. We've got four different buckets for our metrics. And then within those buckets, there'll be different,

ARTHUR: Aspects yeah, and the idea is as well. We want to talk about it and give you like ways that you can improve these metrics Practical practical very practical things that you can actually apply to your website All right, let's go to all time. Yeah, massive. Yeah, so what's dual time?

MICHAEL: You tell me.

ARTHUR: No, I can tell you. I can tell you too. I'll tell you. No, well, I can tell you. Okay. No, you tell me. That's pretty basic. Draw time is, um, I guess the amount of time that a person is on a webpage. So the whole idea is like previously in analytics, you'd look at average time on page. And the idea is the more time someone spends on a page, the more valuable the content is on the page. So, Having a high dwell time is a user signal for Google that the content of the page, A, not only matches the search query and intent, but B, that the content is relevant and engaging. So essentially, the higher the dwell time, the better.

MICHAEL: So to give you an example of what a bad piece of content might do, I don't know if you've noticed, but chat GPT and AI, if we're writing content about digital marketing, it will love to start an article off with in today's digital landscape, something, something. If I see that, I will immediately bounce because I know it's written by AI and it's just going to be largely a bunch of waffle.

ARTHUR: Right. It's just the way that it kind of starts sentences and paragraphs is very typical.

MICHAEL: Yeah.

ARTHUR: Yeah. There's patterns, essentially, that is just kind of across all the content that it's produced.

MICHAEL: So if you start off with a joke or some sort of quirky statement or angle, it's going to be a bit more engaging. Or a question. Encourage you.

INTRO: Maybe a question.

MICHAEL: Yeah. So, um, anything that you can do to make the content seem like it's written by a human and it's actually fun to read because it is fun to read is going to be a good thing. So what we like to do, the big one is breaking your text up into very short paragraphs. I often, my paragraphs are one sentence.

ARTHUR: Yeah. I hate, I hate wall of text, right? When you go on a website and you just like, bang, here's like a thousand words in an essay form kind of like format.

MICHAEL: I never read it.

ARTHUR: I never read it because it's just too, first and foremost, hard to read and just overwhelming.

MICHAEL: Yep. And AI does tend to do bigger paragraphs. But breaking it up, even if you have written your content with AI, spending a bit of time to edit it, don't just copy paste it, bang, straight on the site. But break it up a bit, maybe tweak it when it's using dull and boring language like in today's digital landscape.

ARTHUR: Yeah, or train it as well to not write like that. To try to get it as, I guess, website ready as possible without you having to edit it. I mean, obviously still read it and edit it, but fine tune it so you're not writing the same generic gibberish all the time.

MICHAEL: Yeah, correct. And then using subheadings throughout, instead of just, as you said, a wall of text. Pretty obvious, but not always done, particularly if you're just copy pasting AI content onto a site. I think a big one is having a personality.

ARTHUR: So like, it's not going to be applicable for every business. Yeah, but like your own tone and style of writing.

MICHAEL: Yeah, put some jokes in there, be a bit quirky, have your own tone of voice, personality.

ARTHUR: Well, like you said, you can't be like a law firm and then writing jokes and stuff, but I think it's just, you're right. Like trying to make your content different than everyone else's in a good way, then make it stand out. Make it feel like it's been written by a human, which is, I guess, more difficult now that everything's AI written. Maybe not more difficult, maybe easier.

MICHAEL: Do you know there's actually tools out there that specialize in turning AI content, making it seem more human? No. So like, it's an AI tool that works on AI to make it seem more human.

ARTHUR: I don't know whether you need that. But to me, that's just prompting, right? Yeah, it is. The whole thing in, I remember, I remember when we went to Chiang Mai, it was about prompt engineering. and how important it is, because essentially that's what's gonna create good content. And just investing as much time as you can, engineering your prompts and like building GPTs to actually churn out that content properly. And that will give you proper like tone of voice and stuff like that. So I don't think you need a secondary AI.

MICHAEL: Yeah. It's a good idea. Kids are using them when they're doing their uni essays. They just get it written by a chat GPT and then use this tool to make it human so it can bypass, turn it in.

ARTHUR: I feel like unis have given up on trying to crack down on AI.

MICHAEL: Yeah.

ARTHUR: Wasn't there like ones, I know this is slightly off topic, like a school or uni that partnered with chat GPT or something, or was it the department of education? I can't even remember. I think it was the department of education. They've got their own like AI. So I think they've kind of just accepted that people are going to be using this. There's no point trying to fight it and discourage people. Just find a way to make it work.

MICHAEL: Well, they got it because the old approach of like handing in an essay on a topic is dead. Really? But anyway, that's not what we're talking about. Um, I think a good thing as well is starting like, like any TV show that you watch or movie or something, or if you think of like a Quentin Tarantino or Scorsese or something, they might start with, showing, Tarantino does it a lot, showing something that happens later in the movie, and then working back to the start, you know, three days, so they'll show some big dramatic moment. YouTubers do it all the time. Three days earlier, yeah. So they'll just, yeah. Yeah. A little taster, right? A little hook. A little compelling hook at the start. Something that's going to, I guess, capture their attention and make them keep reading. Yeah. Because that increases dwell time rather than just landing on something so dull that it makes you fall asleep as soon as you read the first sentence. 100%, yeah. And then of course it's not just the text. It's like jamming in images, videos, case studies, videos, like a relevant video could be a way of increasing your time. If they start watching that video on your page.

ARTHUR: I know we were using memes and stuff and trying to, you know, like you said, having a personality. So just like inserting relevant memes or even creating your own memes that are like, I guess, relevant to whatever the content is.

MICHAEL: Yeah.

ARTHUR: To just keep people on the page, keep them entertained. Break it up.

MICHAEL: Giphy.com. Yeah. Pull out some good GIFs. For sure. GIFs. Did you notice GIF?

ARTHUR: Yeah, I know, but I think everyone calls it GIF. I call it GIF.

MICHAEL: GIF sounds wrong, but that's right. Custom images is a big one for me.

ARTHUR: Yeah. Ah, I was going to say AI generated images. I'm so sick of it because they're just everywhere now.

MICHAEL: Yeah.

ARTHUR: We've got them on our site. I know, but I love a good, like a good custom image, like a good real photo. It makes you appreciate it a lot more nowadays.

MICHAEL: Yup. And, um, like I'm just sort of thinking to blogs that I've read by, let's say digital marketing people in the States. And when they have, start off with like a joke and they have some on topic meme or pop culture gif. And then the copy is a bit sort of lively and fun. They're the only sort of thing that I would even give the time of day these days. Like when it comes to reading a blog.

ARTHUR: I was just thinking, I don't know if it's just me, but if I saw like a blog post with a AI generated, like featured image, my assumption will be just that the whole blog post is going to be AI. And it kind of, it's a deterrence to me. Like psychologically, I probably won't read that blog because if they're using AI images, they're definitely using AI content. And if they're that lazy that the AI image is not even like branded or it's just like whatever, I know that the content is going to be rubbish. So I think that people will get more wearied like to it and switched on and do the same. Yep, I agree. So if you have like a proper cover image designed like to match the article that's real, people might get attracted to that and actually see that this is a real article.

MICHAEL: Yeah, well our general gist from the predictions episode is like this AI onslaught, it's out, it's happening and people are going to seek out real human stuff more and more and more as that happens. Yeah, definitely. I agree. Take the time. Use a designer. It doesn't have to be expensive, but make it a bit more tailored to the article. And I know the appeal of scaling things up quickly and just pumping out AI contact with AI images and ticking the box and look at all this stuff is good, but is that stuff useful if it just sits on the site and gets no traffic and dwell times under 10 seconds because people hate it? You're not going to be in Google's good graces and it's going to end up with you, your whole site traffic going down because of it.

ARTHUR: Yeah. The other one here we had was like calculators and stuff like that. So I remember working with air conditioning dealers, having like calculators to figure out what type of air conditioner, what size of air conditioner you need within like embedded within the content as well. So stuff like that.

MICHAEL: Yep, where applicable. Tools? Funnily enough, you could create your tools with AI. So an example in our space is Broadmatch, not even a thing anymore, but Broadmatch Modified Keyword Type in Google Ads. We used to have a tool where you put a list of keywords in, and it would automatically add the plus signs to the words to make them Broadmatch Modified. That was a tool we had a developer build, but now with AI, you could easily have it give you the code to do that, put it on your site. embed that tool in certain pages or even links, because we're going to get into it in a minute, but like internal links and having people, the depth of their visit on your site, not just dwell time is important. So there's so much stuff that can be done, ROI calculators or mortgage repayment calculators, whatever, depends on your niche, but.

ARTHUR: Let's not dwell too much on dwell time.

MICHAEL: I think we've dwelled enough, haven't we? We can move on. Let's go away from dwelling. Let's get into depth. Okay. You can tell me what interaction depth is. Well, for me, it's how much of your site they use when they're on it. Like, are they just going to one page or are they moving from page to page or scrolling more and more down the page? You know, if you have a big, long article, are they going all the way to the end? Or if your article is paginated across pages, are they going across all of them? Or do they come in on a blog post but end up on your service pages? Let's say a scroll depth, which is like scrolling down a page like a big long blog post. Easy hack that I like is the indicator bar that shows you how far along you've read it.

ARTHUR: Yeah, like a progress bar.

MICHAEL: Progress bar. That's what I was looking for.

ARTHUR: That's cool. Yeah, I like that. Or even explaining a lot of blog posts now will have read time five minutes. Yes. So you basically know how much time you're investing into this article.

MICHAEL: Yep. The other thing you can have is like, obviously we spoke about content being broken up with images and that sort of stuff. So you can break up your posts with like stats or questions or little teasers or quotes that are styled differently to the rest of the post just to make it easier to read and get through. And then, like a YouTuber might do, where at the start they say, I'm going to fix this car that I hope to turn a profit on. I'll tell you how much I paid for it at the end, but first. And then the blog post equivalent of that would be teasing something later on in it. Yep. Sticky element, like table of contents. Yeah, table of contents is good.

INTRO: Yeah.

MICHAEL: So all that sort of stuff can help you get people scrolling down the page more. And the other thing would be clicks. Yeah, you're trying to get the most of.

ARTHUR: Yeah. There's tools that people can use to figure out their scroll depth and clicks and heat maps and all that for free like clarity because like if you don't have that sort of software installed, you're not going to be able to tell. I think GA4 tells you. Some scroll depth info? I can't recall. I don't really use that. I avoid GI for it. I don't use it because we use Clarity, and it's superior. But it's a free tool that you add on your site. And you can basically go and analyze any page, see how far they've scrolled. And you'll find that on a home page, most people won't scroll past the first whatever, just below the. pretty much below the fold. Yeah. But that's pretty standard on a homepage. But if you've got articles, you don't want that. That's going to be a negative ranking signal to Google.

MICHAEL: Yeah. Um, clicks as well. I think things like, um, content behind tabs and accordions, read more content or FAQs. It's encouraging clicks. Um, do you reckon it's encouraging people to read more or discouraging? I'm talking about clicks. So if we're trying to encourage people to click on a page because that's the user engagement signal. Yeah. Let's say you have your FAQs. Yeah. Not just having the question and answer displayed. Like an accordion. Yeah, understood. Because each of those clicks is a metric. Yeah, yeah, yeah. That's going to say to Google, well, this is a human engaging with the page. Yeah. Let's say you have a slider, your before and after photo. You know those sliders where you can slide left to right and it reveals the before and after? Yes. Stuff like that will encourage clicks or image galleries that you slide across. On our site, for example, we have all of our client video result, what do you call it? Testimonials? Testimonials. But it's on a slider, so each video has a video and a quote from the person, and you have to click through it to look at them all. All of that stuff is engagement that Google, although they say they don't track, they do, and it's come out in their antitrust cases that they have been doing it for a long time. despite swearing to the SEO community that they don't use user engagement metrics. It's been proven they do. So anyway, that's all stuff that through Chrome, through pixels and tracking and all the other ways that Google knows everything that everyone's up to at all times, they're able to sort of take that as a user engagement metric. What else? Tool tips. You know, like a little eye icon next to something where if you hover over it or click, it expands with more of an explainer on what that thing is. It's called a tool tip. Including that sort of stuff on a site. I don't know what else. Your reviews, testimony, like Google reviews on a slider. Carousels and sliders can be good for user engagement metrics.

ARTHUR: Yeah.

MICHAEL: As a general rule. And then internal linking. Very important, yeah. Of course, it's a big one. Lots of people don't do a good job of their internal linking.

ARTHUR: No. This is like interesting because it's very like, to me, I'm just thinking CRO and like UX and like basically doing a CRO audit on your site to see how people are like using it and how they're engaging with everything.

MICHAEL: Yeah.

ARTHUR: Which is very like, I find it super interesting.

MICHAEL: It's interesting because CRO wants to generate an outcome, like a lead or a sale. Whereas here, we're sort of one step before that, where it's just doing anything with the site is good.

ARTHUR: Because the CRO audit is you're basically looking for hypotheses. But this is essentially not looking necessarily for hypotheses, but you're just doing the audit. You're just trying to see how people are interacting with your site. And then finding, for example, with the scroll depth, how you can improve it. Maybe this content must be rubbish because people aren't reading past this point. So it's basically a CRO without the O. In a way. Conversion rate. Actually, it's not CRO at all, is it?

MICHAEL: It's just UX. It's many of the same principles as CRO. Without the actual… But you're not focusing on that conversion outcome. Yes, correct. That's what I meant. It's more just about… I was about to say tricking users, but it's not tricking users.

ARTHUR: User behavior. User behavior and analyzing how they're navigating and using your website and trying to tick all the boxes. Yeah. That's going to help in engagement.

MICHAEL: But see, putting what we spoke about before, like an FAQ, putting the questions behind a read, like an accordion, a toggle, arguably could be a worse user experience because they have to click to see the answer. But from a user engagement point of view, that's good because they're clicking on the site. And that's probably not going to hurt conversion. So you do need to walk about. The ultimate goal is to get lead sales with the site. So that should always come first before all of this. We should make that clear. Probably goes without saying. But some of this stuff is more just ways of getting people to engage with the page more. without even converting or doing anything, just before they, on their path to conversion. Correct, yeah. There we go. Nailed it. All right, here's another one. So we've done two. The first one was dwell time. The second one is interaction depth. So you're scrolling and you're clicking and stuff. One that we've banged on about is brand. So return visits, people coming back to the site more than just seeing it once and then disappearing, never to be seen from again. And brand searches. As we spoke about last episode, brand searches, there's been studies that have shown helpful content has hit sites that don't really have brand. How do we improve return visits and brand?

ARTHUR: Short answer, really, marketing, right? Just marketing your brand, marketing your website.

MICHAEL: Yeah, actually be a brand. Yeah. Yep. Google ads, Facebook ads, YouTube ads have multi-channel ads running. People are going to see them. People are going to, at some point, search your brand. They're going to search things like your name, reviews. So making sure that you have your review profiles claimed and that you have links from them back to your site so that people are finding your site that way after searching reviews. I heard a good one from friend of the show Matt diggity. Mm-hmm where he's just done a video recently talking about some hacks that he's been using in particular to drive brand traffic. And his was if he's hiring a job, like hiring someone, post a job ad on pretty much every job ad site that exists, which in America, there's a lot more of them. Yeah. OK, cool. That's a good idea.

ARTHUR: Yeah, because the people then search about the brand. But the thing is, is he putting a job ad up looking for someone, or is it just the hackers just to get traffic to the site, not actually looking for someone?

MICHAEL: Well, the end game is that people will have job ads up when they're not necessarily hiring, if it's proven to drive brand searches and search volume.

ARTHUR: But I mean, it's not like people always have, like, always on job ads.

MICHAEL: Yeah, at a certain size, a business is probably hiring all of the time anyway. For sure, yeah. Interesting, that's a good tactic. Yeah, so he's pumping it out on all the job boards. Like, you could do it even as an Australian. Yeah. Put your job ads on US job boards and build a bit of interest in people searching for you, Bram. Yeah, interesting. Another one is good old-fashioned media stunts, which not everyone has the appetite for or the ability. If you're just a local plumber, it's a bit hard to do a media stunt, but… We should do one. I've thought about it, but like… Anyway, it's not to be discussed on the podcast, but yeah, just trying to get PR coverage because that's building a real brand. And then building your own authority in your niche. So using PR, trying to become an expert source. So when journalists are putting together stories, reaching out and getting featured in them, reaching out to journalists that have covered your niche before to try and get found, covered in future articles, All of that sort of stuff takes a lot of work. Takes a lot of time. You got to chip away. It's not something that you just do in an afternoon.

ARTHUR: You have to be- It's going to take a lot of time to build it all up. Building social profiles and then like posting your content on there and then encouraging people to click onto the site as well. Organic like posting on Facebook, Instagram.

MICHAEL: What we're really talking about here is being an actual business and doing actual marketing and advertising. Which every business should be, but not all are. You know, and even a lot of professional or sort of home service businesses, they're very much bottom of funnel focused on people that have that problem there and then, and then they forget about them. But you could. One idea that's just popped in my head now, you know like electricians or plumbers do mail outs where they send out magnets and that sort of stuff to homeowners?

ARTHUR: We have like a hundred magnets literally covering the side of the dryer. There you go. Constant. We had one last week.

MICHAEL: Actually, I've just thought of a better one.

ARTHUR: Oh, wow.

MICHAEL: Why don't you start with that one and go to the better one? Well, then that one, so you could send that out and say, search for whatever to get XR4. Type this into Google to save 25% and it's like your brand and then something else. But there's another thing, a guy in Sydney back in the mid 2000s, he was a club promoter. He also ran a business that did poll posters, you know, posters on electricity poles in advertising events and the like. And he did a campaign where he said, type this into Google, excuse me, type this into Google with an arrow pointing down. And then it was the name of his business. And he put those poll posters up, thousands of them all around like Kings Cross, Eastern suburbs, the city. Thousands of people would have been walking past that and would type that into Google. So that's a pretty good little way, maybe.

ARTHUR: If this was in the mid 2000s, more many people would have had smartphones. Now everyone's got a phone. Imagine how much more affected that would be.

MICHAEL: Yeah.

ARTHUR: You'd walk past it and just search it.

MICHAEL: Yeah, exactly.

ARTHUR: You're thinking about doing it for us, aren't you?

MICHAEL: Well, I'm thinking about Kanye Yeezy. He's whinging about how Adidas is outranking his Yeezy website, and he's whinged about it on socials. So now everyone's going to Google and typing in Yeezy to see for themselves, and a portion of them are clicking his site, sending massive signals to Google that He is a relevant site for the brand. And that's something, I guess that's… 4D chess from Kanye. Yeah, he's a massively effective SEO. What about getting people to return to your site? Outside of brand, you know, people coming back.

ARTHUR: Um, I think just creating content, like content series, for example. So like all content libraries, if someone's like looking for us or working on a specific topic, they'll always come back to you for more information. That's one idea. Yep.

MICHAEL: What about tools and calculators? Well, tools and calculators we talked about, but like, it's another way for not just for engagement on the site, but people coming back over and over again.

ARTHUR: Yeah. Like I see is one that was, um, it was very simple, right. For keyword research, just adding like keyword plus keyword plus keyword. I can't remember what it was called, but I would go to that site religiously. I had a bookmarked. So like, I think one of the points was as well, having like content that's bookmarkable that people always get back to. Yeah. And then that was it. That was my, you know, anytime I did keyword research, bang, I'd open up that site. So yeah, very good point with the tools.

MICHAEL: I use percentage calculators.

ARTHUR: I use it to this day.

MICHAEL: All the time. It's fantastic.

ARTHUR: Yeah. I forgot how to do percentages. You would think by now we would learn how to figure it out.

MICHAEL: But it's just easier. Like if you need to quickly do it rather than pull up your calculator app. Definitely, yeah. The other one is if you have an email list. Well, yeah, true. Sending your content to the list, making it engaging and fun and interesting, compelling so that they A, open the email and then B, click it. That's a good way of getting people coming back to the site. Diversifying traffic sources like we spoke about in the predictions episode, like not just relying on SEO but having people coming back to the site from other ways. Like in our world for this podcast, we have a mailing list of people that have opted in to download our free guide to SEO. But we obviously send these episodes out to that list and those people come back to the site and it's just another way of stimulating brand and repeat traffic to the site really. The general gist is be a brand. Don't just sit there at the bottom of the funnel focusing only on transactional customers there and then and that's it because that's not a real brand and that's the stuff that Google's been at. A, devaluing with helpful content update, but then also B, with AI just becoming more and more prevalent and painful in a lot of ways. It's going to be looking at real brands and these types of engagement metrics moving forward more and more, we think anyway. 100%. It's a 100% prediction. Yeah.

ARTHUR: It's a spoiler. No, it's not a spoiler. It's a prediction. I don't know. What's the saying?

MICHAEL: I have no idea what you're talking about. It doesn't matter. Ignore me. But anyway, that's it. That is enough. Like, I think we are all user engagement metric out. As we said, they're the best metrics. There's more out there. Don't listen. If someone comes to you talking about other metrics, say, listen, I've got my metrics. Okay. But that's it for this week's episode of the SEO show. We will be back. I'm going to listen to the start again because I just remembered I was talking. Okay. Let everyone know if you're happy with it in the next episode. But until then, happy SEOing. Happy SEOing. See you next time.

INTRO: Thanks for listening to the SEO show. If you like what you heard, don't forget to subscribe and leave a review wherever you get your podcasts. It will really help the show. We'll see you in the next episode.

‍