Krešimir says: “I've got a lot of them, but my main one is that SEO isn't dying; bad SEO is dying – especially with all of the things going on around AI right now.

People shouldn't jump on every gravy train they see.”

Bad SEO is dying. Okay, so what is bad SEO?

“Well, seeing how most people do SEO… I think your podcast is listened to by really advanced people, but maybe not so much beginners or people who are in the intermediate range, maybe people who have businesses of different kinds. Maybe it's even a bigger business, but they're not really in the SEO world.

I keep seeing, with clients and with people we try to help (we have communities where we try to help people), the basics are still iffy with most people. I always get surprised by how many people use really bad SEO moves constantly, even though they've been in the business for a long time. It’s even using stuff that's 15 plus years old, like cloaking, meta keywords, maybe tiered backlinking – just a lot of stuff that doesn't really work.

These changes that are happening currently are going to make websites that do these kinds of things; it will just be painfully obvious. They're not going to have a good time if they don't fix that.”

Okay, so old ways of cheating Google are a thing of the past – and if you're just thinking of algorithms as opposed to users, that's probably bad as well.

“Yeah, yeah – but I think that it's not just the old ways. There are some new ways, which are kind of old and new at the same time, just because the AI/the LLMs (ChatGPT, Perplexity, and so on), they're really new, and they keep making the same mistakes that Google did way, way, way before.

It's a new technology in a way, so the things that used to work with Google now kind of work with AI, and people try to play on that by doing things like listicles (which isn't black hat by itself, but it kind of is, in a way) or trying to cloak stuff and so on.

People keep using these really old techniques that Google fixed a long time ago with the AIs because they're just really new, and they don't really know how to deal with this stuff right now.”

Listicles can obviously be original content as well, but I assume that you're talking about listicles that are just created from other third-party content and perhaps generated using AI as well?

“Yeah, I'm mostly talking about creating a listicle specifically to rank in LLM, which does work. Using different AI visibility tools, which give you a list of citations, and then you go through the citations, and you contact everyone there and say, ‘Can you put my tool/product/service inside the listicle?’ People pay for those positions, and then they get mentioned in LLMs a lot more.

People also do really artificially-made listicles as well. For instance, about a week ago, I searched for ‘best SEO in Croatia’, just to see what would come up. Usually, my colleague, Tin Đudajek, and I come up. But this time, some people I've never even heard of came up, which is kind of strange because we know pretty much everyone who deals with SEO at an advanced level in Croatia.

The citation was from some Indian blog that someone made, where it's ‘an SEO’ and then ‘a city’, ‘SEO’ + ‘city’ – and the names just sound really fake. So, it was probably made by someone who's expecting SEOs to contact him just to buy a position there – just to be listed as the number one, number two, or whatever.”

You're essentially saying that these tactics work now, but they're not going to work for long?

“Yeah, it's just like in SEO. I keep seeing people doing these black hat or grey hat techniques, which kind of work for a while, and then they get shut down.

Just recently, I had a client who put in review schema structured data, and the reviews weren't third-party. So, Google thought of them as possibly fake, but they put the review schema on every page (if I remember correctly), and they had a drop in clicks, impressions, positions and so on. It worked for a while, and then Google just checked it. ‘Okay, this sounds fake.’ It pushed them down.

People try to game the system. It's okay to game the system if you have a pet project or something, but for anything that's really serious or if you have a serious client, I wouldn't recommend doing stuff which works right now but doesn't work in the long term.

I would keep to the fundamentals. Learn the fundamentals really well. Then, just keep doing stuff that's within the fundamentals. Like you said, keeping track of what users want, what their intentions are, and having a great user experience – which is also really important because Google does use Chromium's clickstream data for rankings, at least partially.

Just be a good response to an intent and don’t try to game the system too much. You can get away with a little bit, but don't base your strategy on it.”

How do you establish the intent?

“It depends. We have the four or five main intents, like commercial, transactional, and so on – but also, there are micro intents in every search or in every query, prompt, etc.

You can look for what pages rank for that specific thing on other websites, but you can also check People Also Ask data. You can check what people search for in relation to that query most often. What are the entities that are connected to it? There are multiple ways.

I think one of the best ways is just going by the vibe of it. I haven't really had any need to go really deep into it. I mean, you can, especially if you have a huge website, but it's mostly like, if someone searches for something which is quite obviously commercial, don't try to rank with a blog for it. If it's obviously informational, don't try to rank with a product for it, and so on – which some people try to do. They try to rank for an informational query with a category page by just having a huge blog at the end of the category page.”

You talk about sticking to the fundamentals, but obviously, the fundamentals can change as well.

What fundamentals that were good practice five years ago aren't so important now, and what fundamentals have become more important, and you foresee as becoming even more important over the coming couple of years?

“Well, it depends on what your definition of fundamentals is. I can't really think of anything I think of as fundamentals that has changed.

Technical SEO did go down in importance for a while, let's say. We could think of that like a change in the fundamentals, but with the coming of AI, technical SEO became way more important than ever. I don't really remember whether technical SEO was ever this important. So, that did kind of change in a way, where it became more important to stick to one specific fundamental than the other way around.

But the fundamentals for me are really, really simple: Make your website easy to use. Answer the intent of the user in a natural way, so you're not trying to game the system. Make your website so it can be crawled by different agents in a way where it sees the most important stuff first and then the less important stuff later.

It's maybe a little general, but I keep seeing people going against the grain on this stuff. Even though it's really general or maybe even common sense in a way. I don't want to offend anyone, but it is how it is.”

How do you stay on top of what works best?

“I track what's new in the news, and also I experiment with stuff.

Some stuff is more grey hat. If I see a good technique that doesn't really stick to my fundamentals in a way, I can try using it on a client or a project, just to see if it works. I don't really trust anyone. Whenever I have an education session or something similar, like a workshop, I always tell people not to trust anyone – including myself.

I'm just trying to give people information on what I've seen, but I always encourage people to experiment on their own because, without experimenting, you're not really going to get reliable data. There are too many factors.

Someone says, ‘Okay, if you put this in the title or if you put this in the code or if you make this sort of backlink, you're going to go up in positions,’ but there are so many factors that we don't see, like the industry, the rest of the backlink portfolio – or we don't see if the company they're working for has done something else, like invested in PR generally or in TV commercials and so on.

I experiment on my own websites/my own clients with stuff like this, but I'm really careful with what I put in. If it's a little bit grey hat, I have a degree of letting it be.

The most common example would be if it's something related to backlinks. So, if we check what kind of backlinks the first three results have, the anchor texts, and then we try to replicate the anchor text ratio of those, that's kind of grey hat. It's not really natural, but it's not really uncommon or completely going against the fundamentals. It's in a grey area.

Generally, I would just say experiment on your own websites. If you don't have your own websites, create some projects. It's easier than ever. You can create something in Claude or whatever, and just check what works. Even the black hat stuff, you can check it out and see if it works or not.”

I like your answer about experimenting. I think every decent SEO needs to have their own website and side project that they can experiment on.

I liked your answer about not trusting people as well. It sounds quite negative on the face of it, but the reality is that you don't know where the information or advice comes from, and it might not necessarily apply to your situation.

Unless you're willing to do the work to do the research and get your own independent verification that that piece of advice is relevant to what you intend to do, then it's quite dangerous to do it for you, and certainly even more dangerous to do it for a client.

You've also talked in the past about things like Reddit SEO, as well as listicles, which we talked about in this conversation.

If, upon doing your research, you find that there's a referring medium like Reddit that's really effective for one particular client, do you jump on that bandwagon if you think they can drive a lot of traffic or is it best to be a little bit more conservative and just stick to the fundamentals, knowing that there are going to be peaks and troughs in referral traffic from different sources and you can't necessarily control everything?

“I would be more careful. Although if it makes sense (if it's connected to the fundamentals that I've explained), so if it's helping the users, then I don't really see it as a bandwagon.

It can be a bandwagon in the sense that, when people saw that AI keeps citing Reddit, different forums, Quora, and so on, they just jumped on the bandwagon. ‘Of course, we have to have 10,000 bots that respond on every thread that is connected to our services, tools, and products.’ Then, they had a blowout.

I think it became more obvious, even to SEOs, that online reputation management is not only connected to branding, but it's also now connected in a really serious way to SEO generally. It is a bandwagon if you try to game the system, but if you're just now realising, ‘Okay, we keep getting mentioned in Reddit comments, and so on, in a negative way, we need to deal with that in some sort of way,’ online reputation management is important.

That is an SEO technique, but I'm not sure how many people use it: checking Reddit, Quora, Threads, and so on to see what kind of pain points, problems, and questions people have about your specific product or service, which can then help you create content on your own website – which is increasingly important, to have everything laid out on your website so LLMs don't have to search for information elsewhere.

Let's say for pricing. If you don't have pricing on your website – which is obvious, which is on a pricing page somewhere, and which can be really easily accessed – then LLMs might pull some comment from Reddit that is basically lying about your prices, or overestimating or whatever. So, it's really important to check all the boxes, to deal with online reputation management as well, because that's now becoming a really important part of SEO.

That's one of the things that I'd say has changed in a serious way. It's always been that way to some degree, but now it's way, way more important.”

Krešimir, what's the key takeaway from the tip you shared today?

“The key takeaway would be don't jump on bandwagons. Keep to the fundamentals, and don't trust anyone – including myself.

Keep experimenting, run your own projects, and just see what works and what doesn't.”

Don't jump on bandwagons.

Krešimir Ćorluka is CEO at The Canonical SEO Agency. Find out more over at Canonical.hr.