Loading summary
Transcript58 lines
- [00:00]
A
The Voices of Search Podcast is a proud member of the I Hear Everything Podcast Network. Looking to launch or scale your podcast, I Hear Everything delivers podcast production, growth and monetization solutions that transform your words into profit. Ready to give your brand a voice? Then visit iheareverything.com welcome to the Voices of Search Podcast. A member of the I Hear Everything Podcast Network, ready to expedite your company's organic growth efforts. Sit back, relax, and get ready for your daily dose of search engine optimization wisdom. Here's today's host of the Voices of Search Podcast, Tyson Stockton.
- [00:43]
B
Global SEO feels overwhelming these days. New market demands require unique approaches. Language barriers create unexpected obstacles. Cultural differences impact search behavior. How can you scale internationally without sacrificing.
- [00:57]
A
On the local level?
- [00:58]
B
According to a 2024 report by Weglet, 75% of marketers agree that localized content significantly increases customer engagement. Yet only 15% reportedly effectively manage SEO for multiple markets. This highlights the growing complexity of successful global SEO. This is the Voice of Search Podcast. My name is Tyson and joining us today is my friend, longtime colleague Sebastian Egger, SEO Consultant at SEO SEB and one of our experts in residence at Pre Visible. Today, Sebastian is going to be sharing his insights into how to effectively manage global SEO across multiple markets. Si welcome to the podcast.
- [01:37]
A
Hey, thanks for having me back. I think it's been, it's been a minute since I've been on, so yeah, good to be back.
- [01:43]
B
It's been far too long, far too long.
- [01:45]
A
How you been? Not bad, not bad, enjoying enjoying life in apac, currently in Tokyo. So I've been doing a few conferences so doing my rounds of finally getting out there. I think when I was at Square I wasn't really allowed to do much influencing or much self promotion, let's put it that way. So now I'm catching up a little bit. But yeah, it's been fun. It's been fun to try to break apart the whole APAC market. Mostly unsuccessfully, but it's a good challenge.
- [02:16]
B
Ah, don't sell yourself short. I mean the chains and shackles are off and you're able speak freely about the work that you've really been doing for quite some time. I mean you have an international background of course, like anyone that maybe is not aware, like you have roots in Europe, spending time living there as well as in the us and then now further expanding that experience into Apex. And I know probably some of the listeners have seen some of your kind of post and information from some of these conferences that you've done in Korea and so I think it's a perfect time to have you back on the podcast and kind of share with our listeners a little bit of your experience. And success is moving, you know, not just on like one market, but really thinking about it from like a global perspective.
- [03:08]
A
Yeah, exactly, exactly. Love it. Yeah, the, the conference I did over in Korea was, was interesting because it was figuring out how to speak to a completely different market, yet with the current SEO landscape that's ever changing. So an interesting challenge. But even over there, even in a landscape that's very different, the interest still remains very, very high.
- [03:32]
B
Which maybe on that point too, how have you been seeing the evolution in kind of global SEO? Do you feel like your approach or tactics have shifted in recent years or has it been kind of, you know, more of a continuation of status quo?
- [03:52]
A
Yeah, I would say, I would say it's a mix of both, but in general, definitely more of a shift. The first, the first thing is it's been a shift because as you work on different markets and actually let's use Korea as an example, their the Google market share is 50% Google and 50% Naver. So there, when speaking to that market, one of the mistakes that I did is I actually, when I first did the conference, I actually didn't know that. I completely ignored that. And it wasn't until I spoke to people and actually the conference owner was like, Sebastian, just FY, when you speak to these people, there's 50% Naver. And I'm like, okay, I probably should have done a little bit more research. So don't be like me, research beforehand. Don't assume that Google has 90% market share. Different countries have different search engines. We all know about China, we know about Russia, but a lot of people don't really know. And common misconception also with Japan is people think that Yahoo is majority shareholder. That's actually not true. Google is primary. So in that case, it's kind of business as usual. But it's weird because with the rise of just large language models these days, that's basically on everyone's mind. And that's basically the main thing that people want to talk to me about, is how to adapt. But then I also work. A friend of mine, he, he works in France and so I've been just helping him and coaching him. And over in France, you don't actually have AI overviews or at least yet on whenever this podcast recorded June 5, AI overviews are not present. Maybe, who knows, by the time it launches, they're in France and then this comment is useless. But so basically you have to be with all the changes happening and the differences in search engine market share, you have to do research and understand these different nuances. You can't come in and expect that. For example, in, in France, AI overviews everywhere, it's like, no, no, you're still dealing with a gold rush of high traffic. Korea is naver. So your optimization is going to be naver Webmaster tools. And now there's Google Webmaster tools.
- [05:54]
B
And what's like, how do you think about the balance? Because you know, at least at the enterprise level, you're not focused on one market.
- [06:03]
A
Yeah.
- [06:04]
B
And so from like the executive or leadership level, you're looking at what's our global strategy? Whether it's like specific products or services, positioning, what have you. But then you have like how that translates to the tactical local level. Like, how do you look at balancing the uniqueness of a market but then also creating consistency or alignment to like a global vision or strategy of a business?
- [06:35]
A
Yeah, it's an interesting question because a lot of American companies, while they might be successful from a product perspective, product penetration, very successful internationally. I mean, the top companies in the world are mostly American companies and their products are used worldwide. But also to my surprise, a lot of these companies do a very poor job at localization. So I always think that if only they would do an even better job, I can't even imagine how much better their market penetration rate is. And over at Square, when I led SEO over there, what was interesting is we put a huge focus on localization and we were mostly competing with local products, local companies rather than the global companies. So I can see here a just high advantage of the localization. But I think going back to initially your question, I think I veered off a little bit. One of the things that I would do is I tried to understand the total addressable search volume. Basically when we enter new markets, people would do like TAM analysis, total addressable market analysis. And that, that looks at just a myriad of different factors, but it never includes organic search, never in these, in these analysis, it never includes how people are searching. So what I did is I stayed very close to the people who did these researches and I translated just normal search volume into total addressable search volume by products. Because you do the TAM analysis from a product perspective, what you do, you want to do the exact same thing. Whatever they're doing, you're copying everything, but you're just changing the KPI to that total addressable search volume. And first of all, you Want to see does it match up or not? It doesn't match up. And very often you'll see some, some small mismatches. It gives you an idea that maybe what people are looking for may be a little bit different than expectations. And that's where it requires a slight difference in strategy per market.
- [08:34]
B
Interesting. And I don't know, this is probably a large question, but taking that information, I feel like you go two different directions with that. Where you have managing up with it, which I think could be an interesting thing to look at. But then also, and I think maybe more straightforward is like okay, great. That then shifts how we prioritize on a local level. I think that's probably where most SEOs feel more comfortable. What recommendations or advice would you have? Because again I feel like SEOs like will grasp and get how to capture that. Like what you said, what to do with it is probably the more ambiguous or tricky piece. So like would you have any advice for SEOs out there into how to kind of manage up with that information? Because you know, I would, I would assume too that's really valuable and it may be a difficult conversation to have, but if you are able to have that conversation, you're really going to elevate the importance or like value that search can bring beyond just revenue of the channel.
- [09:42]
A
Yeah, yeah. It starts with being in the whatchamacallit, the omnichannel marketing conversations. So for example, let's use, you can use Square. Any other, any companies there's always like a brand awareness team. They do like net promoter scores. And then in our case we had a go to market team and you might have just whatever other teams and very often, and I know you know this like SEO is for, for some reason always on the sidelines and that's, it's weird and because at Square, SEO was about organic search, I should say was about 40% of marketing new profit. Do you believe that 40% of marketing net profit was the channel that I led? But yet when I, when I started it wasn't as part. Well, first of all, when I started it wasn't 40%, it grew to 40%. But in any case. So for a lot of companies it's that case and they're always on the sidelines. And so the first one of the things that I did is then okay, you talk with the brand awareness team because there's tremendous, I mean there's so much you can get from organic search data. For example, the search volume by market. You can see how frequently your brand is being searched. Not just your brand itself, but then your brand and product identifiers are searched across market and that information, I know it sounds like pretty trivial, but that information has to then go to the brand awareness team that they can use it together with what other data sets they have. So then when they even report on it and talk about it, they're able to talk about their own data sets together with organic search. Same thing with the go to market teams, those who go into new market expansion. You want to make sure that you're in these conversations and just push through the door and showcase the usefulness of this data. And ultimately it's the, I think what a lot of SEOs, the mistake that they make. And it's not unique to global SEO, it's just a general, a general thing is they keep thinking that people search for these terms on Google. No, it's before ChatGPT and whatnot, Google was the only canvas where you could type in your thoughts. And so obviously you get all this information in terms of what people are searching for and you know exactly what they're searching for. And that's something that should then be used across channels. Especially when we get into like conversion rate optimization, especially when we get into like the web strategy, web experience and whatnot. How they're searching should be something that's used across the board because if they're searching for it, they're also looking for it when they're going on the website. So then if you use that information and create a sort of global matrix of how people are searching across different markets, then that can also then educate in terms of the web strategy and web experience. So then already you can see that all of the insights that I've mentioned so far, it's all about taking organic search outside of its silo, outside of its box, and just sprinkling in the insights for different teams because it's also helpful for different teams to help them do their jobs and get additional insights on their market. And so I think from a managing up perspective, it really is critical and at least for me, it's worked really, really, really well. And then you get into the whole conversation of which KPIs that you use. Do you want to go into rankings? Probably not. I've done a lot of work into translating that specifically on market share. So you translate. There are a lot of different ways you can calculate market share yourself, like keyword share yourself versus competitors. But that really, really helps because it's again translating these insights to just general marketing insights that people are comfortable with.
- [13:28]
B
Yeah, so you're, you're making it more approachable to those executives, to the audience, people. So then I think, and everything that you're saying, I feel like it's really, really critical for the SEOs to like, listen to and adapt on, because that is what's going to enable your team to then execute on, you know, the uniqueness of the local markets. And I think that's a piece that oftentimes gets lost or missed is like, people can latch onto, like, yep, I understand there's difference in language, especially it being a completely different language. I also can understand there's difference in US English, British English. Probably not a far stretch for people to recognize maybe some differences in, like, consumer behavior. You called out earlier, different search engines. I would also assume two. And you know, from some experience, I've seen that, like, beyond those different platforms, you also have different behaviors within. And from a lot of input that I've gotten is, especially in a lot of Asian markets, you have a very strong social presence, which is probably greater than some other markets. And so that can shift things a little bit. Oh, yeah. But maybe assuming that, you know, we've done our homework, we've been successful in, like, managing up with it when it comes to, like, managing multiple markets and leaning into the uniqueness of those markets. Like, what advice would you have for the global SEOs out there?
- [15:04]
A
Yeah, yeah, I love it. So, okay, we can start maybe with the easy part, which is the partnership with your localization team. And so in that, the partnership, it's a. How can I put this? Like, it's a physical partnership as well in terms of, like, your work streams, but it's also a technical partnership, meaning whichever CMS you're using has to be connected with whatever localization program that they used. So then the machine translation can happen directly in the cms. Now, when I say machine translation, I want to be careful. It's machine translated, but then edited by a human. We might get into that topic later, but that's important to know. So there's always someone behind it. Just checking. But. So that makes sure, to your point that if you're writing something in US English, that it gets properly adapted to UK English. That's. The first part is like just that. Local content optimization, local keyword optimization, obviously, Local keyword research, Local content optimization. And then you obviously you've got like the technical stuff, which is obviously put it in your subdirectories and HRS line tags. I don't know if we'll get into that, but from A purely adapting perspective. One of the mistakes that a lot of people do, and I've seen it with my competitors, it's not adapting for. From the conversion aspect. There's the SEO. There's the SEO aspect, which is like, okay, creating your content into like the right. The right local US English versus GB English. But then there's also the whole, specifically from the conversion perspective. So let's use an example for France. I know by experience and also by research that the French market loves anything that's made in France. So with that.
- [16:53]
B
Wait, wait a second, wait a second. You're saying that the French love French.
- [16:57]
A
The French love the French. What a surprise. Who knew? But so as an American company, where it gets interesting is, okay, obviously your product is not made in France, so, okay, that's no go. But what you can do is really showcase how your product can help the French market. And so that's showcasing how we can support the French market. And that's done through various ways. My favorite one is imagery. Super underrated. Like, insanely underrated. And this is something that companies to a terrible, terrible drama. And so, okay, so the first one, and this is, this gets a little bit into the CRO conversation of things. But in the test that I've done, number one is if you test people, let's just say you have your product and you test your product on its own versus your product with people using it. The imagery with people will always, in my experience, always, always convert better. And that shows that people want to kind of see themselves using the product. So there's always. There's that personal connection. And this is something that we've seen across markets. So the second point here is think of your messaging. Does it translate across market? I have a great example. This is actually something that happened, I think it was last year or a few years ago. I don't actually remember when with hsbc, the global bank, they had this slogan called assume nothing. Little did they know, if you translate that to a bunch of other languages, you translate it as do nothing. And they lost $10 million through this basically, like, rebranding blunder because they didn't think about localizing their slogan. The proper, like, localization of their slogan. That's. I don't know. I love this example because, like, they just lost 10 million bucks by just not properly understanding local market nuances. Even in previous role, like, previous roles, I had the same issue with the team rebranding their blog to something that is completely nonsensical in other markets. I don't want to state any names or anything, but it's something that was very, very tough to handle because the team was very adamant toward, oh, we're going to name it like this. When all other local market owners were like, the name doesn't make sense. And so a lot of companies, they always focus on like us, the US first piece and don't really take into account the third one is maybe a little bit, I don't know, maybe a little bit controversial. But it's like, don't be afraid of stereotypes. And so by that I mean positive stereotypes. Let's use the French again. Because I'm half French, I feel like I can get away with it. So I remember when we were launching in France, again, like I said, I'm very pro imagery, like using very local specific images to showcase the relevance and that we understand the local market. And so I was asking our creative teams, like, okay, I want imagery for main. For a homepage to basically be the different shop owners, like our product with people in that. In like very French specific. Like not market, sorry, like shops and whatnot. There's businesses, sorry, businesses. Well, what does it mean? It means like bakeries. It means wine shops. It means, you know, you get the croissant, you get the baguette and whatnot. Jacques Cousteau. And the response I got, I guess people didn't, didn't know that maybe Phil didn't know is half French or that I basically, I grew up in France and so they, they came out to me like, Sebastian, that's stereotypical. Like, we're not going to, we're not going to do this. That's overly stereotypical. My response to that is like, no, that's just a fact. Literally, if you go to France, 90% of the business you will see are bakeries. And whenever I go to France, my mom will force feed me croissants. Like it is, it's a stereotype for a reason, but it's a positive stereotype, it's not a negative stereotype. And so things are like this for a reason. And so you want to lead into this, obviously. Would that appear in like a, like a stupid American that goes into stereotypes. But in some cases it is like that for a reason. And in our case, like, no, it's like most of the businesses in France are going to be like small businesses, wine shops, small restaurants and things like that. You want to showcase this because it shows that you understand the market. And so that's very important research. And from a conversion perspective, it really, it really makes a Big difference. You get other things like the logos, the social proof, anything like that, showcasing local companies and whatnot. I'm working with a brand right now, and they're. They have the same logos, like, sitewide. They'll show a German company on their US Website. They're showing a typical American company on their German website. And it doesn't make sense. It's a very small change, but these things really, really. Yeah, really, really matter.
- [21:46]
B
Well, and I feel like to that point too, it's like, yeah, maybe there's stereotypes, but it's also just leaning into giving respect to the uniqueness of that market.
- [21:56]
A
Absolutely.
- [21:57]
B
And I think even on like a micro scale, I feel like the automotive companies do a really good job at this. Like, if you watch any local automotive commercial, they are going to show footage of the car. Like, the format and design of the commercial will be the same across, like, nationwide. But then if I'm in Northern California and I'm watching a whatever brand commercial, I'm seeing them drive the Sonoma Coast, I'm seeing them cross the Bay Bridge, the Golden Gate Bridge. And it's all, like, things that I can associate with, like, and I feel like it really ties into more, I don't know, like, factors of consumer behavior where in that sense they want the viewer to see themselves in the car. And it's like, well, I know that road driving big, sir, and I want to see myself in that new car driving it.
- [22:54]
A
Yeah.
- [22:54]
B
Similarly, in your examples with, like, shops, it's like, yeah, if you show someone, like, in France a Costco, like, that's not really going to resonate very well because they're like, what is this? Like, they won't have any things. And it's like, yeah, it's stereotypical bakeries, but there's a good reason. The bakeries are delicious. And it's amazing in France. So it's like, why not lean into to that?
- [23:16]
A
Exactly. And there's. There's a bunch of other, like, small things that you can do. One of, one of the other ones is that I noticed is, for example, each. Each market has specific certifications. Or there's. So, for example, again, like with square, you have to be certified by the French, the French government, they give you a seal which is like the rooster and like red, white and blue. And a lot of companies just promote. Promote that on their website. Right. And it's something that, for example, square, we never did, even though we obviously had the certification. But these little things, they just, again, they showcase that you are, that you understand the market that you're also like, technically certified in that case. And there's a bunch of these other examples too, whether it's like review sites, having the badge of the local review sites and whatnot. So, like, what's interesting here is we're getting kind of even away from the SEO side. Obviously, there's like, there's a lot more on the content strategy piece that we can, we can talk about after, but from like a first, the first stage, it's. I really think it's a web strategy piece. It's like figuring out the demand in each of these markets and then figuring out how you go to market and how you're going to target the local consumers through basically, like going away from SEO, like focusing on conversion and web strategy and then the tactical piece, which is the local, like, local content strategy and how to go about that.
- [24:34]
B
Yeah, absolutely. No, I feel like we can't talk or we can't completely talk about globalization without really digging into content, because clearly a lot of the technical side stuff, you have some nuance, but it's not, you know, what you would do in one market. It's not going to be drastically different than the other, except for, you know, pointing search engines in the right spot. But how about content? What's like the kind of rule of thumbs or the approaches for kind of giving respect to the local markets and content?
- [25:02]
A
Yeah, yeah, Good, good segue. It's. I think in my, in my experience, this is where a lot of companies don't have a very good playbook. And so the way that I go about it is I first of all try to understand how my current content is performing globally. So I guess there's like, there's two cases, like if you want to expand globally or if you're already in different markets. I guess it doesn't matter that much. But I just look at how it's performing globally and I'm trying to see, okay, or US Content, let's say. Let's just say you have one en page and that's meant for globally, how does that page perform in the US versus how does that page perform in the UK versus Australia versus Canada and whatnot? And that could be super interesting because if you notice that it doesn't perform well in a lot of cases, most likely it won't perform well if you have one em page, that usually means that you're lacking a lot of local distinctions. Obviously, there's going to be a lot of other factors at play, for sure. I don't want to simplify that much. But still for me that's a really good indicator. Especially when we talk about like in the SaaS world, in product pages, this is where the conversation gets interesting. It's like, okay, do you want, maybe you want to have a page for UK versus the us? So that's the type of conversation that you want to get into. So in any case, you start with this and then let's just say you notice you have a specific piece of content, let's say informational, that is not doing well in the uk. Then you can start thinking about localizing this into engb, for example, and then you implement the hrefline tags. I'll talk a little bit more about the hreflang tags because there's some very interesting research that I've noticed on my end. But anyway, so you implement the tags. That's the localization part. Sorry, the localization part is then you localize that into gben. And then there's the optimization part, which is you use a semantic content optimization tool the same way that you would optimize any piece of content. And you redo this process which is basically researching all of the local specific terms to target. So then that first phase is you're not targeting anything that is. Sorry, I meant like the global specific terms. You're not initially researching the local terms, you're keeping it global. So you're trying to understand like if I have, if I'm targeting something in the us, a specific topic in the us, can that same topic also be targeted in au, in gb, in Canada and in other countries and even in non English speaking countries, France, Japan and whatnot. Because what you want to do is for that specific topic, you want to connect it via hreflang tags. So you're creating this just global topical hub where you know that a specific topic holds relevance across the globe and everything is connected via hreflang tags. The reason why I swear by this strategy is because one thing that I've noticed is when you grow globally, most likely there's always going to be a few markets where you don't hold high authority. And let's use Japan as an example. In previous companies that worked, APAC has always been tough. And so by following this framework, what I've actually noticed is authority actually gets transferred over via hreflang tags. And I think I've heard Google say not really, but I don't believe it. I've seen the complete opposite because whenever I've had issues on sites with hreflang tags, the local content always ends up dropping And I think that's fascinating insights if you think about it, because then that means that actually proves the point that when you grow, when you want to grow into specific market, you want to focus on topics that already exist in other markets and you build that authority with the hrefi tags, with the related content abroad and you use your most authoritative market to then help the smaller markets. And this really, really works well. And then the second phase is you can then start targeting the local specific terms and then that allows you to then get into the local market. And it's basically a global, global local approach. You know, it's like a think globally and act locally. Or if everyone, anyone seen the movie up in the Air, it would be the locale from Dirt Clay. So yeah, does that, does that make sense?
- [29:37]
B
Yeah. So it's like basically you're using, let's call it like your primary market as your foundation to then translate or just localize if it's the same language. And then from that you're like using that as the foundation for each local market and then you're adding nuance in addition to it from there.
- [30:00]
A
Yeah, yeah, exactly.
- [30:01]
B
With some of like the localization aspects, like, so let's take like one language and you made reference to English and then differentiating, you know, UK versus US what would be your rule of thumb of where to draw the line into diminishing return for it? So obviously US, uk, obvious ones, Australia, probably another obvious one. But then you also have ones like Canada is probably another good one. But then you have Ireland, Scotland. What's the kind of point that you're like, okay, yeah, it's probably not worth having some separation or some differentiation between, I don't know, Scotland and Ireland.
- [30:45]
A
That's a really good question. What's interesting is in their rankings, what I've noticed is there's a correlation between geographical closeness. So we actually had a lot of issues over square with US content and Canadian content or GB content and Ireland content. And so for a lot of companies, it's going to depend how different. I don't want to say how different the culture is because then I'm going to get burned where they're like, well, obviously Canada and the U.S. is different, which it is, in all fairness. Obviously it is. I'm not saying anything different. But then it's how different is your product, its marketing, its narrative to the local culture. Like when you're targeting something you can have, like SaaS, for example, if the industries between where you're targeting from the US and Canada are Drastically different. Then I would say most likely the SERP is also going to be different and then that would. You would then need a local specific page. But if you notice within the SERP there are a low amount of local competitors, the market and the audience is fairly similar, then most likely not really and you can get by with one page. So I guess it comes down to just reverse engineering what you're noticing and the amount of local specific competitors. And usually if there is a high amount of local competitors, that means, well, there is a need for that Hyper, hyper localization. And based on again, like the work I did at Square, that is something we noticed. Like in the UK there's a, there's a fair number of local payment providers. And in the end when I looked at the global, the global competitors like Shopify and whatnot, they didn't do as good of a job in localization as we did. And the data showed that in the end, like we weren't competing against the global guys because we were doing better than them. Our goal was like, hey, let's break through the local, the local folks. And that's where the money was at. So I guess that's the way I would go about it.
- [32:43]
B
Fair.
- [32:43]
A
Fair.
- [32:44]
B
And how about as far as, you know, using more, you know, LLMs or generative AI from like a translation standpoint like I know before, Google would always go on record and saying, you know, human translated content wouldn't be kind of qualified as duplicate content regardless if you have like hreflang tags or not. Where's kind of the point that you would say you can use some of these automated tools versus needing the human element at this stage? Is it something that's more of use it as your first draft as if the same for content creation and then add an editing layer or a piece on top that's human.
- [33:29]
A
I think that's the right way of going about it. Ultimately it goes, it's so much faster to go about it. The funny thing is I actually have a degree in translation and interpreting and international business. And I keep thinking if I were to do in five or 10 years from now, is that degree still going to exist? Because I would spend hours upon hours translating word for word and figuring out the grammar and whatnot. And now you have tools like DeepL or DeepL, however you pronounce it that really, they've mastered that. And so yeah, it's, there's no reason not to use it. I think there's always going to be a need for the human element, especially in localization I think this is something that people think that like a localization program manager and localization team are going to go away. I really don't think so, because there is. The nuance is something that's always going to matter and like the originality and uniqueness of content is always going to matter, but it's about the efficiency. And right now we can produce and translate content way faster than before at really high quality, but it only gets you maybe 60 to 70% there. And that's not good enough. Like 70% of the way there is not good enough. And you always need someone to manage this, edit it and just give it that specific touch. But I had nothing against it. We should use it and it gets you there faster as long as there is the human, the human element that's, that's needed. And I think that that type of content, that type of translation will be the one that will win not just from a ranking perspective, but even just from a conversion perspective.
- [35:03]
B
And I would assume too, like, you know, I remember as well, like we used to take like a similar approach like that where we would take the content, we'd have it usually in like some sort of Google sheets, use kind of the function for translating and then have the human editor over the top. I would assume though, with LLMs like use like you could probably get even closer with prompts versus like more traditional translation tools. Like is that something you've seen also kind of gets you, you know, maybe instead of 60, 70% there, it gets you 80, 90%.
- [35:37]
A
You can refine the prompts as much as as you can to get something. I read this thing the other day which I thought was really interesting. I, I guess I thought about it in that way, but not as much. Someone said that creating content in AI will be unique content, but it won't be original, which I thought was very interesting because I completely agree. So you can refine. So with that thought, and per what you just asked, you can refine the prompts as much as possible. And that's true. You will get something that's unique, unique to you, that no one else has maybe asked that way. But because large language models are based on existing content, things that already exist, you're never going to get something that's purely original. You're never going to get an idea that's completely out of whack, that's completely came out of nowhere because it's doesn't have your experience. Like us humans, we're experiencing things that might be brand new all the time versus LLMs, it has to be read somewhere. Now what's interesting is with ChatGPT, you can change what you call the parameter setting. So you put a prompt, you put in parentheses. Temperature equals anywhere from zero to one. Zero is the wackiest one is the most conservative. So you can play around with this and you could do something like create a article about this and this and that. Parentheses temperature equals like 0.1. The default I believe is 0.3 or 0.5. I'm forgetting. So you would do like a 0.1 and then you would get something that is very wacky, that basically like the way it would work is that it'll just generate a bunch of the tokens that are the least probable based on what it's coming up with. So that is how you would get something that's pretty unique and fairly original. But again it is based on anything that's already existing. And that's why I believe that regardless of how you're doing, like prompt engineering or you're refining it, at the end of the day there's always going to be something missing. And for me, I don't know, like 80% is still not good enough. I would want to get as close as possible to something that is based on human experiences. I don't know. And that's why, if you think about it, right, that's why ugc, Reddit and Quora, they're skyrocketing in demand and valid people are demanding. Ugc, not aigc. I guess I'm coining a new term.
- [38:02]
B
Yeah, it's been interesting too to watch that kind of pendulum swing of the favorability of ugc. I remember, I don't know, seven years ago or something, UGC was just really tough from like a quality standard and like now obviously it's a lot more favorable.
- [38:20]
A
That's a good point actually. Yeah.
- [38:22]
B
And then I agree with you on the originality versus uniqueness. It's like, it is kind of a fun thought process though of like, well even within, you know, humans, like how much of our thoughts are truly unique versus previous experiences and things like that. So it's like you kind of go down that like slippery slope of like, okay, well then I learned this from someone else piece. But I think as far as guiding in like a writing standpoint, I would agree with you that it's like the main piece or at least what I kind of like push people into is the first hand experiences. It's like opinions, firsthand experiences, like those are things that are like more human in nature. Which is, you know. Yeah, you could say people regurgitate that as well, but it's maybe not as, like, extreme, I guess.
- [39:12]
A
Everyone, they have their own way of writing, and so that's something that maybe they'll appreciate. And same on the. On the global side, whatever you're writing, especially if you think about. I mean, think about British English or American English, it's not just a spelling difference. Obviously, there's word differences too, but the way that the sentences are structured. You read an article from the Economist versus something from the Times, you'll notice, like, there are differences. The Economist being British, Times being American. And so when you localize the content. I don't know if that's true. That's more of like a feeling. I feel like someone who's British would notice, like, oh, that sounds like. That sounds British for America. I don't know. I'm making this up. I assume they would.
- [39:51]
B
No. And you have, like, little nuances just even within one country and stuff. Like, maybe it's. Again, there's always a point of diminishing returns, but, I mean, you can tell by the way someone talks where in the US they're from.
- [40:08]
A
Yeah, very true.
- [40:09]
B
And like, even on a micro scale, it's like you could pick up on little nuanced details that would differentiate someone from Northern or Southern California even 100%.
- [40:19]
A
Yeah.
- [40:19]
B
So it's like there always is that kind of uniqueness in region going back to HRF lang tag. And so you kind of dropped on that. We wanted to kind of come back there. I feel like this is a common question that comes up in global approaches. And I feel like you oftentimes hear people having issues with the wrong content showing in the wrong market. And so, yeah, maybe there's problems with the setup of href link tags, but what have you seen as being kind of like the most common issues when a website is having issues with maybe like the UK content showing up in the US market or vice versa.
- [41:02]
A
That's. That's an interesting one that. That happens all the time. And a lot of it comes down to this is something that goes beyond the Ahrefs lantags. Oh. So first of all, I shouldn't say that. Actually, the first thing is hrefs land task can be pretty tough to master because you want to make sure that it is relaying from one page to another that are relevant. If you're not doing that, then Google will choose a completely different piece of content. At the end of the day, like, it's a. It's It's a direct. Like you're. You're telling Google something, but it doesn't have to respect it, unfortunately. It's same as like a. Maybe a canonical or something like that. So that's the unfortunate piece. But generally it's pretty good, I would say in my experience it's been pretty good at reflecting it when it doesn't, you know, let's just say you have everything right. Your country code, your language code, you've got your X default. That's something that actually a lot of people forget is the X default. What is the default page that Google should showcase if it doesn't know what to do, or if it's from a country that you haven't specified or something like that. Make sure you have that. Make sure in each country you have your canonical that goes back to the, the main page. So their AU page that canonical back to your AU page with hreflangs to your US and so on and so forth.
- [42:22]
B
Can you elaborate a little bit more on the default? Like I would assume from that it's, you know, you're going to have like a initial or primary maybe market that you're producing from. So if it's like US company, they originally started within US and then they've kind of expanded out, like, would that always be the practice then, that in that scenario, like the US one, US English would be the default? Or is there situations that you might, you know, use a different, I guess, country as your default?
- [42:55]
A
That's. That's an interesting question. I would say whatever has the most authority should be your X default or whichever the content originated from. That's kind of my rule, my rule of thumb.
- [43:07]
B
And would you differentiate that within a company? Like so say. I don't know if this would be the case with Square, but like you have a global company, typically things would come out of one country, but then there's times when you know, someone else might produce something first and then eventually it gets adapted into other regions, like within one domain. Would you always have the same default?
- [43:31]
A
I think so. I think so. I would try to keep that as consistent as possible. Yeah, there.
- [43:36]
B
That makes sense.
- [43:37]
A
Yeah. Another mistake that I think this one is tough, but it's the language selector. This is something that companies do very, very wrong. The language selector, especially when it's. It automatically redirects you. Keeping in mind that Google crawls mainly, let's say, from the US So if you have like an AU page and your country selector automatically redirects, it is possible that it will automatically redirect Google to the US website, which is a big reason why US content ranks above non US content in non US markets. And so a lot of companies do this wrong. They don't do proper caching policy. They just have like a. There's a user, they may not differentiate a user versus a bot. There's a 302 or 301 redirect and then you get screwed over. So that can cause a lot of issues. And depending on the CMS they're using how new it is, how old it is, sometimes you have like a bunch of JavaScript redirects, sometimes you have a bunch of parameters. So that can also cause a bunch of crawl efficiency issues with your site potentially down the line causing ranking issues that can relate to the wrong page being present in the wrong market. So it's important to have technical hygiene in that case. That's even going beyond the technical stuff. Sorry, the global stuff is just your technical hygiene will have an impact on the local findability. Basically.
- [45:05]
B
That's such a good call out too. And it's like one of those things that it's like may seem like a small detail, but then you amplify that out over the growth and size of the website. It's like those things that maybe wouldn't be an issue on the like the smaller scale like can compound into being like a significant point.
- [45:26]
A
Yeah, yeah, exactly.
- [45:27]
B
Well, that's going to wrap up this episode of the Voice of Search podcast. Thanks again Sebastian Egger for joining us. Be sure to check out his information in our show notes. You can connect with him on LinkedIn or go on over and check out his company's website@seoc.com if you haven't subscribed yet and you'd like a daily feed of content and SEO content, be sure to hit that like button in your podcast app or on YouTube and we'll be back in the following day. So with that, that's all from the Voice of Search. Thanks for checking it out and we'll see you in the next episode.