Simon says: “Remove the complexities of SEO and get back to the basics. No one is going to be able to keep up with AI in 2025, so stop trying.”

There's a lot to unpack there. Shall we start off with what you mean by saying that no one is going to be able to keep up with AI in 2025?

“A recent study by the California Institute of Technology has suggested that our brains can compile information at about a throughput of 10 bits per second, which is not a lot. That's not reactive, where somebody hits you and you punch back. That's actually thinking about bits and pieces. AI is going to be a bit faster than that. It going to be a huge amount faster.

My guess is, this year (because there's been hints, etc.), people like Google will be making use of AI to change things on the fly. We're kind of seeing that a bit already. I think that's going to come in a lot more.

We always know that search depends upon the things that people are asking, and we know that 15% of queries every day are new, so how do you manage that as a business? I think Google's just going use AI and they are already, and we're just going to be keeping up with how they change the SERP.

I don't know what half the features on the SERP are now. Yesterday, I was asking what one thing was, and nobody came up with the answer. It was a grid of 9 boxes with a bit of AI in it. Nobody knows what it's called. It said, ‘It's from other sources around the web.’, but what's it called then?”

Even Barry Schwartz hasn't come up with an answer.

“No. He didn’t come back. It's not new. It's been there since 2022, but I didn’t know what it was called. I did it for a client, and I'm like, ‘What is this? What am I going to call this? I’m going to have to screenshot it.’”

Maybe you can name it. Maybe it hasn't been named yet. It's like, ‘Simon Cox, the first man on the moon.’

“I don't want to name it. It's not that exciting, but that's the point. I think things are going to change so much. The SERPs are going to change so much. Nobody's got time to sit there and just keep watching what the SERP looks like for one keyword, all the time, every day.”

Perhaps, to keep up with AI, we need a hardwired connection to our brain just so we've got that bandwidth.

“Going into Musk territory there. I'm sure some people would love to do that. If you want to be a slave to that, fine, but it’s not for me.”

There's no end point, really, is there?

“No, but maybe that is the way. But then, if you are wired up like that, you're not going to get paid.”

Wired up. No pun intended. What about embracing AI. Are you using AI at all in your day-to-day SEO activity? Are you saying that SEOs should still be doing that?

“Yes. I think that there's a difference between using AI and trying to keep up with it. Using AI to speed up your processes or make them more accurate (although we know AI is not particularly accurate), but certainly for research and helping stuff on the way – AI as a tool.

I'm seeing it in in more and more places, and in software. I spotted one in Chrome DevTools this week that I didn't know was there, which is an AI tool to help you understand why certain CSS classes aren't or are working.

We're going to see that pervading everywhere. Outside of the trust of whether it's accurate or not, it's pervading everywhere, so we're going to use it whether we like it or not. I mean, I use Grammarly all the time because my spelling is terrible (my wife’s is brilliant and she shouts at me). That's a good example of AI that you don't even think about because it's not going into ChatGPT and saying, ‘How do I say this better?’”

So, to summarise: don't bother actually trying to take the time to determine how Google is using AI to form its SERP but, in terms of your use of AI as an SEO to enhance/make your activities more efficient in what you're doing on a day-to-day basis, absolutely keep on trying to do that.

“Yeah, absolutely. I don't think we've got any choice. More and more tools have got AI built into them. Screaming Frog's adding it in now. We've got to embrace it and see where it can improve our ability to provide SEO services to people.

But trying to keep up with the SERPs, no.”

Trying to determine which AI tool to use is challenging because an SEO, like any savvy digital marketer, is drawn to bright, shiny, new things and experimenting with things. But, obviously, you want to be doing day-to-day tasks that are profitable and relevant to you and your business, or your client's business.

How do you actually determine which AI tool to embrace and use to enhance what you're doing and which tool not to even look at?

“I think, this year, what we're going to see is AI going into our everyday tools as opposed to going out to a specific AI tool to use it to do something.

It's going to be embedded in the tools we use. I don't think you will be going out to say, ‘Well, I'll go to ChatGPT for that one or Claude for that one’, etc. There are loads. Perplexity is quite good as well, but they're all slightly better at one thing or another. It's difficult to work out which one I should go to for those things. I think what'll happen, this year, is that people will use the AI tools that are given to them within a tool.”

While we're on the subject of tools, how about talking about how often we should be reviewing the tools that we use? Because, obviously, the tools that we use nowadays aren't necessarily the tools that we used 5 or 10 years ago. If you were involved in SEO 15+ years ago, there are many tools that come to mind that don't exist anymore.

How often should you review tools and determine if that is the tool that you should be resigning with and definitely committing to use over the coming year?

“I like to keep an eye out on what's going on in the tool market, who's doing what, and what added features are on tools. I will generally look every year, because of subscriptions, yearly subscriptions, etc.

Most of the tools I've got that I use, I've had for many years, and they've just improved – mainly because I email them and say, ‘Oy, that's not working right’ or ‘This could be better’, so I tend to stick with them. But I'm always looking at new tools.

You're right. There are absolutely tonnes and tonnes out there. I've looked at huge amounts of tools, and some of them are very good, and I like them, and I think, ‘Well, maybe I would use them.’ It depends upon the workflows I'm doing at any one time.

I like to have a bank of particular tools but be able to add and take away from that, as each project needs. I think the approach that people need to look at is, what do you need to do the type of SEO you're doing, and what suits your clients or business the best?”

So, ideally, as an agency, if you have different types of clients, you would have access to different tools and use different tools for different clients depending on their specific needs.

“Ideally, yes, but most agencies will have one tool, or one or two tools, and they'll have a cookie cutter way of dealing with clients, which makes economic sense.

I don't do that because I'm an agency of two people, so we'll manage it depending upon what the project is.”

Going back to the tip that you shared at the beginning, you said to remove the complexities of SEO and get back to basics. What are you saying by removing the complexities?

“If you have a look at social media these days, as we all do (hopefully, everybody's moved to Bluesky by now, as they didn't enjoy Mastodon too much), read through that, and read what's going on, there is so much information about what you can potentially do. A lot of it's repeating each other, etc. but there's a huge amount. You could spend all day just reading SEO articles and not getting any work done.

Jumping back from that, it's good to go out and test bits and pieces. Going back to the basics: just making websites that can be crawled and work well, and content that does what it says on the tin. You are talking to an audience. You're talking to one person somewhere as they're reading your thing, not the search engines.

Don't try and chase what the search engines are changing every day, every hour, or every minute. Try and deliver something to your consumer. That does come back to EEAT: writing good content. It's all there. It just hasn't changed.

The other big one is, spend time building a brand. If you've got a good brand, you don't have to worry too much about where you're ranking and/or ranking drops. We're seeing that, more and more, brands are winning at the moment. They go up and down, and they're doing well at the moment.

I'd step back overall and don't chase all the little fiddly bits of that last 20% of trying to optimize something. Just go back and do the basic things. Look wide, not just in one narrow angle. Don't go, ‘Let's get featured snippets on every page.’ Stand back and look and say, ‘Are we delivering the right content for our business, our clients, and for what the user wants?’”

I was fascinated that your description of what the basics were started off with EEAT and went on to brands from there, and there wasn't a specific mention of technical SEO. Is that in the last 20% that you were referring to?

“No. Technical SEO is an absolute foundation. You need to make your website work, and technical SEO really hasn't changed a great deal over the years. Everything else is changing around it.

If you've got a solid website that can be crawled and that isn't causing problems, then that's it. That's your technical SEO done. That's the massive foundation. That's basic. Get that working, and then the content on top is the bit that really makes the site shine.

You can have a really, really fast website but that doesn't mean it's going to be number 1 in the rankings. It's not going to get loads of traffic if the content sitting on top of it is drivel.”

What are 1 or 2 key aspects of technical SEO that you find many SEOs aren't getting quite right now?

“It's difficult to say because I work by myself. I see a lot of websites where I see a lot of problems with SEO, but I don't know whether an SEO has been working on that site or not. It's usually developers getting stuff wrong, not SEOs. Technical SEO these days is all there. There aren’t any secret bits or hidden bits.

There's experience that allows you to get in and understand what's happening to a site quickly, and why it's not working. There are no real secrets in technical SEO anymore, but there are some real basics. I was working on a site this week where pages didn't have H1s, and the ones that did were hidden. That's a basic.”

Are we not perhaps moving to an era where Google, or whatever crawler, is getting clever enough to actually determine what a web page is and how it's seen by users, and markup isn't as important as it used to be, or are we always going to be, over the next few years, giving ourselves a better chance of ranking highly if we if we use things like H1s?

“I think content structure is really important – not so much for search but more for accessibility. Where you've got assistive screen readers, etc., having that content structured, with correct heading structure and the content there, will help.

Anything that helps accessibility automatically helps SEO. That may be a massive tip for this year. Go and learn accessibility because it does help with SEO. Yet, you're quite right. Search engines very, very likely don't need all that. They can probably determine what a header is and what have you, but why make them work a bit harder?

When a developer develops a site and then hands it over to another developer later on, it would be nice if that was developed to a standard where the next developer can come along and say, ‘Oh, right. I understand how this is built.’ If you haven't bothered with putting H1s in, people might go, ‘Which bit’s the heading? What do I need to change?’

It’s not difficult, getting all those bits and pieces right, but I think accessibility is the biggest reason for doing that.”

Just circling back to AI in the SERP, if you're working in an organisation where your business leaders are saying things like, ‘Oh, we've seen this in the SERP. We've seen this change. We understand it's not 10 blue links now. What's our SEO strategy for ensuring that we're bringing in as many visits as possible from the changing SERP because of AI?’

“Well, that's the hardest thing when you've got a business team that know a bit about SEO, they know it exists (and who doesn't these days?), and you've got a CEO breathing down your neck saying, ‘Why has my competitor stolen my number one favourite keyword? What do I need to do?’

It's very difficult coming from that point because that's not really where you should be coming from anyway. Really, I think the thing to do there is talk about key performance indicators with the business. Being ranked number one on a search word isn't a key performance indicator.

It'll be a bottom line, like how much we brought in this month from selling our products. That needs to be how our reporting from SEO fits into the business reporting.

You need to work backwards from that, and you need to work with the marketing teams and the developers and the business teams, to understand what they're trying to do, what they're trying to sell, what they're trying to promote, and/or what information they're trying to get across. Then build on that and ensure that you're giving them good advice and strategy for how they can promote that content.

Senior managers screaming at you is never a nice thing. That's when you should maybe think of another job or going freelance and doubling your rate.”

If an SEO is struggling for time, what should they stop doing right now so they can spend more time doing what you suggest in 2025?

“Well, I kind of hinted at it a little bit, but rank tracking. What a waste of time. Chris Green gave me access to his testing a few years back, where he was testing keywords every minute, and the fluctuations in one day are absolutely wild.

I've never really trusted rank tracking since then because, depending on what time of the day you go and get that ranking, or what day of the week, it's going to be different. Two hours later, you're not in the top 10, you're not even in the top 30, because of different factors.

I think AI is going to push that more and more as well, as it's trying stuff to see who clicks on what, etc. So, forget rank tracking. I'm sorry to Semrush, Ahrefs, etc., but I think they know that as well, which is why they're expanding their suites of tools.

But, yes, rank tracking is really a waste. It's nice to have, it's good to look at, but don't report on it.”

Simon Cox is a Technical SEO Consultant at Cox and Co Creative, and you can find him over at SimonCox.com.