Recently I was invited to TheSEOTalkers twitter space to discuss AI and SEO. It was of course a great conversation and I ended up answering many questions about the topic. I thought it would be useful for people to get to know more about those questions and what are my thoughts on AI & SEO.
That’s why in this blog post, I’m sharing transcript of that twitter space recording and I’ve also filtered it down as questions/answers so its easier for you to understand my thoughts on specific questions related to AI & SEO.
Some information about TheSEOTalkers and What it is?
TheSEOTalkers is India’s first and oldest running Twitter Chat and Community focused on Search Engine Optimisation (SEO). It is an official account that hosts a regular Twitter chat called #SEOTalk, co-hosted by
Malhar Barai – Senior Director Marketing, Affle
Parth Suba – Digital Marketer, SEO & Inbound Marketing
Jaydip Parikh – SEO, B2B, eCom & Digital Marketing
Varun Kumar – Media Buyer Head, Kings Digital
TheSEOTalkers also hosts live sessions on Twitter Spaces, covering various SEO trends and topics. Additionally, they offer premium SEO services to help businesses improve their online visibility and reach a wider audience. Apart from the Twitter chat and live sessions, TheSEOTalkers also engages with the SEO community by sharing insights and asking questions related to SEO best practices, such as the impact of reducing orphan pages on a website.
They also have a podcast called “The SEOTalk Podcast” available on platforms like Spotify, where they discuss SEO-related topics.
Below is a brief summary of questions that I address in Twitter space and just after that I have provided full transcript of discussion as well, also I have linked to other relevant webpages, just in case you need more context to understanding something that I mention – Feel free to click and read through that page and of course come back to this one again.
The Rise of SGE & its Implications
How did SGE originate and how has it evolved?
So the story of SGE started in 2017 when Google’s engineers wrote a paper in which the name was Transformers. And then they figured out that whatever content is read a lot on the internet, we find patterns from that and predict the next word and create a system which will essentially generate content.
And after that, the whole race of AI started because ChatGPT got, I think, 2 million users within a couple of weeks of launch. So SGE sort of started as an experiment and then over time it took off and now it’s becoming AI overviews. And now Google is just like pushing it to everyone in the US. And their plan with recent IO is that they want to push it to over a billion users by the end of this year.
What impact is SGE having on search behavior and user expectations?
So what Google is betting on is the fact that more and more people will search for these long tail queries once they get to know that Google can answer these queries. Right now, people are not even searching these kinds of queries because there is no expectation of the answer. But once there is an expectation of the answer, people will start to search more and more long tail queries.
Why is Google so aggressively pushing SGE and what does this mean for competition?
So what Google is betting on is that over time, all these long tail queries will drive more traffic. With more traffic, they will be able to identify specific intents and they will be able to charge more to show Ads for those specific intents.
How are SEO agencies being impacted by the uncertainty surrounding SGE?
I think they need to update that. That will be the first part. And then the second part will be managing expectation and communicating it directly to clients. So I don’t think that agencies will be able to survive much in this kind of environment unless they figure out a way to communicate AI answers information to the clients in a better way.
AI’s Influence on Content & SEO Strategies
How can AI be used in content creation, and what are the limitations?
I think if brands want to leverage AI, the first starting point can be the content itself because content usually takes a lot of time and then building a content strategy also take a lot of time. But from what’s Google recommendation is that we should not be using AI content extensively, but if brands can leverage that to build the structure of the content and not the actual parts of the content. I think it’s a good tool and a good use of resources of the brands.
How is the focus of SEO shifting in the age of AI?
I think SEO will shift from optimising for search engine to optimising for discovery surfaces. And with any new tool, there will come a new discovery surface and we will need to optimise for that.
Why is it important to strike a balance between using AI and maintaining human expertise in SEO?
Gagan repeatedly cautions against over-reliance on AI for content and emphasises the importance of human oversight and originality. Taking a balanced approach is must.
Google Updates & Their Impact
What was the impact of the March Core Update, particularly regarding “helpful content”?
With the March core update, and then they decided to sort of like merge the helpful content system with their core ranking systems. So I think even Google is not sure as of now what is helpful content. That’s why they keep on changing their own guidelines about it.
Why has it become so difficult to outrank forums like Reddit and Quora?
And I’ll be honest with you guys, I’ve worked with some of the largest news publishers in Australia and they are unable to outrank Reddit. So there is something going on with the Google algorithm right now that is clearly giving boost to ranking of Reddit.
Alternative Strategies for Brands in the Age of AI
Why is an omni-channel marketing approach important for brands in the current landscape?
I think the best thing for them is to sort of like figure out a way where they are less dependent on Google. That’s my personal recommendation as of now. So I think that’s what I’m trying to do right now. Rather than being like an SEO expert for my clients, I’m trying to be like an advisor for them.
What kind of content resonates on platforms like TikTok and how can brands leverage this?
Leverage platforms like TikTok to drive traffic by encouraging users to search for you, not go to your website.
Below is the full transcript of this Twitter Space
Parth: More beam other me cool cool cool. We have the guest speaker of the day with us. I can, good evening. Hey, good to see you around. How are you doing, man? It’s almost 1 a.m. or 2 a.m. for you.Â
Parth: Okay, I can’t hear the gun if you’re speaking.Â
Parth: Okay, you know when you would maybe you can try to join in again. Yeah, okay. In the meantime, welcome all to SEO episode 61. Every week is a roller coaster ride for all the SEOs, for all of us. There are so many changes and case studies and updates. Sometimes John Mueller’s statement comes, sometimes Google’s official statement comes, sometimes some other clarity comes, sometimes Lily Ray comes, Mary Hines comes.Â
Parth: So let’s try to address it and we are super excited to have our regular attendee as a guest speaker today. Welcome to SEO Talk as a speaker and thanks for sharing all the insights so far. But today we’ll specifically try to pick your brain around how AI is changing the SEO landscape, individual bloggers, brands, and so let’s start without waiting anymore because it’s already around 2 a.m. for you.
We’ll try to wrap it up real quick in one hour. We don’t want to extend it, but thanks so much for taking our time and joining at this hour of the night. So yeah, cool. So we’ll get started. And before we get started, like again, a lot of people know you, but do you want to give yourself a formal introduction to the SEO community so that the people who are listening to the recorded session can get to know you better?
Gagan Ghotra: Yes, hello everyone. I’m Gagan Ghotra. I’ve been living in Melbourne for like last six years, but I was born in Haryana. I came here to do my Bachelor of Computer Science because I got a scholarship, fortunately. And I started to do SEO back in December 2019. And I jumped in SEO because I discovered it through one of my web development class.
And so it was totally unexpected for me. I was supposed to become a software engineer, but I ended up becoming SEO. So for the last four and a half years, I’ve been doing SEO in agency side freelancing consulting with brands. And recently also I was working as SEO manager, but I left that job around one or two weeks ago. And now I’m just looking for a new opportunity.
Parth: Awesome. Thanks so much for the introduction. So today’s session will be all about AI and SEO. So I’ll hand it over to the host of the day, Varun Bhai. Do you want to host?
Varun: Sure. Thank you. Welcome everyone. And this is our 61st edition. And I will be more like a listener to Gagan Bhai. So welcome Gagan to our SEO Talk podcast. And as Parth told you about AI and what is happening in AI, what kind of tsunamis are coming, what kind of changes are coming.
And in our previous discussion, we also talked about Google and a lot of things, whether Google is getting relevant or there will be a new search engine or there will be a new update. So we will talk about all these things and what is happening in the AI world, what kind of updates are coming. We will talk about all these things. So let’s start on our first topic. This is a broad subject. SGE and Google is very bullish on SGE. So Gagan, what is your take on SGE and its related things?
Gagan Ghotra: Yes. So the story of SGE started in 2017 when Google’s engineers wrote a paper in which the name was Transformers. And then they figured out that whatever content is read a lot on the internet, we find patterns from that and predict the next word and create a system which will essentially generate content. So that’s where the idea of Transformers started. Then OpenAI picked up that idea and they worked further on it and fine-tuned that idea.
And then we got ChatGPT in November 2022. And after that, the whole race of AI started because ChatGPT got, I think, 2 million users within a couple of weeks of launch. So that was the start point. Users liked it because it was just giving clear answers, but it was also spitting out wrong information as well.
But just to not let it grow so far and to assert its market presence, Google started to come up with new ideas and started to build similar tools. And the first one that they launched was SGE, Search Generative Experience. And this was launched in, I think, 2023, I think around that time. And this was just in beta for some US locations and some countries in Asia.
And this was just like taking all the information from the internet and presenting it as a short snippet, just like a feature snippet which we used to see. But this was essentially like picking up the information, rewriting it, summarising it, or sometimes simplifying it. Or even in some cases, it was just like copy-pasting the information. So it took like a paragraph from my website, a paragraph from your website, just combined it together, simplified the language, and just presented it as the answer.
The challenge for SEOs and business owners came when it started to become a big thing because Google started to show it for more and more test users over time. And it came to a point where site owners were questioning the fact that if this is what the future of search is, then how they will get clicks and how the traffic will come to their website. And they will end up losing all their potential customers. Even if it’s like a service side business, they will end up losing all their clicks to these random article websites or random spammy websites.
Or if it’s a site that makes money essentially through ad clicks, they will also end up losing all their traffic to Google because they will be getting lower and lesser click rates. So SGE sort of started as an experiment and then over time it took off and now it’s becoming as AI overviews. And now Google is just like pushing it to everyone in the US. And their plan with recent IO is that they want to push it to over a billion users by the end of this year.
So the impact of that for the businesses will be severe because as a community from its experience, we already know that the future snippets were stealing a lot of traffic. And now we have a bigger future snippet which shows up on the top of the search results.
Varun: So as you said, it started as an experiment but now Google is pushing it. So what is the future you are seeing? Especially in terms of Google, like their revenue is also going up. And you also tweeted about it. Although it’s a funny part, the serious part is that Google’s revenue is also going up. So what is the benefit for Google to push AI and SGE?Â
Gagan Ghotra: So essentially, if I am using Google to look for the information for a really long tail query where I am saying that I am visiting India and I will be visiting these states and plan the hotels and restaurants near me, that is like a really long tail query which if I am using Google now, then it will take me like 30 minutes to figure out all the information and put it together as a Google Doc. So what Google is betting on is the fact that more and more people will search for these long tail queries once they get to know that Google can answer these queries.
Right now, people are not even searching these kinds of queries because there is no expectation of the answer. But once there is an expectation of the answer, people will start to search more and more long tail queries. So what Google is betting on is the fact that there will come a new set of queries which are really long tail and the intent for those long tail queries will be really, really specific. For example, if I am saying the same example of the query, then the intent is really specific that I am actually interested in visiting Delhi, Mumbai, and Bangalore.
And the hotel star ads will be really, really specific. Even if in SGE, Google shows me a hotel ad in Delhi, I am more likely to click it because where I am coming from is a really long tail query where the intent is clear. I am already planning to go. There is no like yes or no. There is no exploration. So what Google is betting is that over time, all these long tail queries will drive more traffic. With more traffic, they will be able to identify specific intents and they will be able to charge more for those specific intents. So an ad placement in SGE will cost more than the usual top ad.
This is my expectation if the query is really long tail because the intent will be really, really specific. So this will impact people who are running ads to get clicks. But it’s gonna be like their CPC will be higher, but there will be lesser impressions that they need to be short for. So this is my thinking that over time, the longer tail queries, they will try to figure out a way with the ads to monetize it better.
Varun: So if I summarise this, it is eating the informational website. SGE is showing its information and capturing the long tail queries so that it can convert higher intents to ads. It is targeting them too because it is losing quality traffic.
But there are a lot of AI chatbots. For example, if I am visiting a city and I have to plan my itinerary, a lot of chatbots can create like that. There are so many chatbots. We see in the market, there are 50 plus AI chatbots available who can do these kinds of things. So will Google lose its relevance and ultimately the ad revenue also?Â
Gagan Ghotra: Yes, I think the biggest advantage with Google is the data. I’m not sure about the exact number, but they have like millions of business listings and direct access to that. Then they have YouTube. Then on top of that, they have Google Maps data, which no one has. So if you combine these old data sources together, you can build a really, really good product, which otherwise cannot be built.
So another example is Perplexity. So they can crawl the web, and they can pull up the information from different sites and build the answer, but they don’t have access to like millions of business listings, what their descriptions are, what services they offer. And on the same site, at the same time, they don’t have access to YouTube review videos about different places. What Google can do is use all these information sources, which are private to them, and sort of build a unique experience. But it’s gonna be like an engineering task for them. But otherwise, I don’t feel like there is any competition for them.
If they fail, if Google fails in this, then it’s on them, their engineering capability, like engineering mistakes, and their size. Otherwise, no one has so much data, which is like their own private data, that they cannot figure this out.
Varun: I agree with that, that Google has a whole ecosystem, they have a map, they have YouTube, business listing is one of the greatest assets they have, which is probably bigger than Yellow Pages and Yelp, Google has data on every part of the world. So if they can control the coding and processing cost, then they have a monopoly, and it will be a bigger monopoly. But if we look at it in a different sense, if different chatbots can grow their business on the basis of that data, then does Google’s business look down?
Gagan Ghotra: Yes, before we move to this question, I just wanted to add one more point, that Google is also working on nano models, Gemini nano models. And with that, they have Android as well. So they can convince all the Android phone manufacturers to have Gemini nano as a default. So Gemini nano can also do a lot of AI analysis on the phone itself, and then they can have a bigger ecosystem advantage will be the Android ownership that they have. And moving on to your question, I don’t think that it’s so easy for any provider to source so much data.
Like we have seen certain examples where OpenAI, they have gone ahead and done content deals with the Financial Times recently, and with the AP as well. But it’s not like they can go ahead and do such sales of deals to get the data of scale like business listings, because Google is the only one that have access to this data, and they will never give access to this kind of data set to OpenAI or any of their competitors. So I don’t think that in terms of acquiring data, there is no advantage like, and no other company has an edge over Google.
Varun: After all, the data is the new oil. Why would anybody give it to their competitors? So just like you said-
So yeah, but it is very costly and very important nowadays. Otherwise, without it, we are not doing anything, even not this access space is also possible. So moving on, like our premise is set, how is SDE and the rest of the AI? So now the SEO is also changing. So how do you see changes in SEO? Like before, you had to work for feature snippets, now you have to work for SDE. And Info website, the way it has flown in recent updates. So how is SEO changing and what should we do?Â
Gagan Ghotra: I think with SEO, it will become more of a discoverability thing. So rather than optimising for search engines, we will be optimising for discovery. And with optimising for discovery, we will just look for discovery surfaces. So one surface can be like Google Discover, which is like, we already do optimization for news publishers to show up in Google Discover.
Then the next surface can be SDE, or then the next surface can be a chat GPT UI, or perplexity or anything like that. So I think SEO will shift from optimising for search engine to optimising for discovery surfaces. And with any new tool, there will come a new discovery surface and we will need to optimise for that. And a simple example can be like, right now, like many startups are working on building chatbots, which are specific for Google Sheets. So that chatbot will be a new discovery surface. We can create content which shows up in that, that is specific to a site which publishes content about like Google Sheets.
So I think with this change in optimising from search for search engines to optimising for discovery, SEOs need to like work on their skillset. That will be the first part. And then the second part will be the unpredictability, like how we can communicate that to the client. This would be a big challenge because right now, if we take certain steps over time, the ranking of course improves and we can show it into search console data or Google Analytics data. But with optimising for surfaces, it will be really hard to show the specific data because AI will make multiple choices almost like every day.
So in the morning, for the same question, it will pick up my site and in the evening, it may pick up your site. And the next day, it may pick up someone else’s site. There is no certainty. So I think the challenge is that we need, SEOs need to shift to the discovery, optimizing for discovery surfaces. And then the next one will be to build a sort of like a framework, which they can use to communicate to the clients that this is what is happening with your site. And this is why it’s not predictable to predict what can be the traffic trend for the next six months or something.Â
Varun: Okay. So here is a counter question. Because you just said there is no certainty. If there is no certainty, how the SEO agencies will work? Because a client will say, I was ranking in the morning or I was ranking this keyword for the yesterday, but I am out of the top 10 ranking or something like that. Malhar bhai, please go ahead. Malhar bhai, Opath.Â
Malhar: So thanks Gagan for setting up context of the conversation today. But I have this question like yesterday only, I think, so today morning, SEMrush came up with a study that at this point of time, like less than 1% of the queries are being solved with AI overviews. So do you think Google is going to cap these, the percentage or the diversification of queries through AI overviews?
Gagan Ghotra: Part, I don’t think that they will cap it. What they are trying to do right now is like, they are not providing the opt-out option. So they are pushing it. And whenever Google pushes something, it is to train the users. It’s not to get like an immediate result. They don’t need immediate results in terms of revenue or anything. What they need, what they are doing with this one is training the users that this is possible with Google.
Because nowadays, users just think that searching really long tail queries is only possible with chat GPT. But with this, without providing any opt-out option, they are pushing it to the users and training users that you can reset the expectations from the Google. And once that expectations are reset, then they can look into increasing it. So it’s almost like if I am searching something, I noticed that, oh, generative AI showing up. I will share it with my friend saying that, oh, look, this is cool, cool feature. You should also try it out. So this is what Google is doing.
They are pushing it to a small subset, but the goal is to spread it out. But it should be like verbal by people, not like Google pushing it out to all the users. Because they don’t need to like hit any revenue target from it in the next quarter or something like that. So I think over time, they will release it to more queries. But what will happen is for short tail, they will stop it. Once they get enough volume for the longer tail queries, the example that I gave you earlier. But for the short tail, it’s always gonna be the regular feature snippet, I think.Â
Malhar: Okay, got it. Yeah, that makes sense. Thanks for answering.Â
Varun: Okay, back to the question. Like if there is no certainty, how the SEO agencies will survive in this Google AI apocalypse or something like that?Â
Gagan Ghotra: I think one of the things that I noticed is like many SEO agencies use like standardized dashboards that they have built either using Google Studio or maybe Google Analytics, or maybe they are using some other tool to build a dashboard to report to clients. I think they need to update that. That will be the first part. And then the second part will be managing expectation and communicating it directly to them.
So I think it will be like a communication challenge to help client understand that this is what is happening and this is what we can do as maximum because clients over time will understand, but the initial challenge will be there. So I don’t think that agencies will be able to survive much in this kind of environment unless they figure out a way to communicate it to the clients in a better way.Â
Varun: That is a very interesting perspective because many people will lose their job.Â
Gagan Ghotra: Yeah, I think many people will lose job, but if they figure out a way to communicate it to the clients early, then they will win. Then they will get the jobs which other agencies which are unfortunately will shut down because of that. So some agencies will win who will be just like first movers in the industry, so.
Varun: Definitely. And it’s always interesting to think like, what can we as a SEO, if I’m reporting to a client who does not know much about SEO, unfortunately like many small to medium sized business owners they don’t have much knowledge about SEO. But with the Search Console Analytics 4, it was possible to like communicate it to the clients and show the month over month or quarter on quarter growth.
But I think with SGE showing the data, that will be the number one challenge because Google is not providing like any AI overviews data in Search Console as of now. And then how we can communicate it to the client that makes sense to them. And they does not feel like, oh, SGE is giving me a hundred clicks this month, but I am paying you same amount and next month, it’s just 10 clicks. So I paid you same amount for 10 clicks only, what I am paying for like, like what will be the answer from an SEO if a client is asking this?
So it will be very scary time for the agencies and the SEO freelancer also, who are not good in communication or not good in updating themselves. Malarbhai, can you take? Looks Malarbhai is celebrating RCB win.
Malhar: No, no, no, sorry. Yeah, it took me a while to unmute, but continue the conversation. It’s an all interesting point.Â
Varun: So moving on, there are, like we have seen the AI reviews, AI overviews and the hallucination of information and there is also bias kind of things in the AI. So how do you think this will affect the SEO and search overall in the near term?Â
Gagan Ghotra: I think the first challenge will be for the company itself. So Google, if they are presenting wrong information directly in their search results, they may get sued for it. And then they have to do the litigation and everything. And which means that Google may be sending some traffic to spammy websites on the internet. And what we as an SEO, I think we can take that as an opportunity and write content about the same queries. So for example, if I tomorrow in the morning, I see that a spammy site is ranking for a query, I may write the content for it and outrank it.
And if you guys remember it back in like 2016, 2017 time period, SEOs used to do same thing for forms. So if for a query, Quora used to rank. SEOs used to write like a detailed blog post about the same keyword and they used to outrank Quora. But that is no longer the case nowadays because Google is just boosting those forms nowadays. So I think hallucination, Google will try to control it over time. But meanwhile, if any SEO sees any spammy site ranking, just write an awesome blog post and you may outrank that. I think you will, of course.
Varun: I think Google is just doing that. Let’s rank Medium one year, let’s rank Quora one year, let’s rank some other kind of Web 2.0 this year, and then let’s rank Reddit again in 2024. This kind of thing is happening in the Google headquarter one year after one year. So now let’s talk about the SEO update and the March Quora update. So many things have been changed and there are content in search results. So what do you think, how the SEO is changed and what people need to do? Because most of the queries that Reddit is ranking. So how to outperform that?Â
Gagan Ghotra: I think right now with the March Quora update, Google promised a lot of things, but I think they did not deliver on any of it, unfortunately. Rather, they just ended up boosting ranking over Reddit and Quora. And I’ll be honest with you guys, I’ve worked with some of the largest news publishers in Australia and they are unable to outrank Reddit.Â
Varun: Oh, come on, man.
Gagan Ghotra: So there is a clear sort of like something in the algorithm, which is clearly saying that no matter what, just rank Reddit. Now, even as an SEO, if I am honest with you guys, I don’t have any answer to outrank forms like Quora or Reddit. But in the future, that may change because right now Google is clearly giving some algorithmic boost to Reddit and there is no sort of like seems like a way to beat that boost even with the high authority sites like The Age. Like The Age is one of the oldest newspapers in Australia and they cannot even outrank Reddit. So there is something going on with the Google algorithm right now.
Varun: And has the Google stopped giving the revenue sharing to the Australian newspaper? Was it confirmed?Â
Gagan Ghotra: No, they again struck a deal. So earlier, they had a deal for three years and then they negotiated again. And now, I think last week or the week before that, I even tweeted about it that they now have a new deal. So that is not stopping.Â
Varun: Yeah, so I miss that because they have said that they don’t want to restart that deal again.Â
Gagan Ghotra: So yeah, that is with Meta. So Meta also, when Google did similar deal with Australian newspapers in, I think three or four years ago, Meta also followed and they also did the same deal because in Australian parliament, they were about to pass a law to force these companies to pay. So they were like, oh, don’t pass the law, but we will do the negotiation with news publishers and we will do it voluntarily by ourselves.
And I think Google’s new deal is finalised around two weeks ago, but Meta is still negotiating. And then there was some news around that Meta is threatening that they will not do a deal and publishers can just not show up in Facebook or something. So maybe in one month or two, we may end up in a scenario where Australian newspapers are not showing up in Facebook, Instagram, or anywhere.Â
Varun: That’s interesting. Malharbhai, please go ahead.
Malhar: Yeah, thanks Varun. So again, in terms of, say, what we just mentioned, see even beyond the media sites and Google’s last update, which they said helpful in core update, et cetera, where do you see this impact of AI coming in? Because even at this point of time, we are still not sure what helpful really means and is Reddit equal to unhelpful content, if we are going to say that.
So how does one prepare for such kind of scenario when the instructions or probably the indications from the search engine is still not clear in terms of what helpful is? So how do you see this basically, or how do you think a brand can balance that entire AI impact along with its core updates coming in?
Gagan Ghotra: I think with the especially March core update, and then they decided to sort of like merge the helpful content system with their core ranking systems. And also they launched helpful content system back in August 2022. And last year in September, they also did an update. So there was something wrong with the previous system. That’s why they decided to update it. So I think even Google is not sure as of now what is helpful content. That’s why they keep on changing their own guidelines about it.
And in terms of what brands can do, I think the best thing for them is to sort of like figure out a way where they are less dependent on Google. That’s my personal recommendation as of now. Even if a brand is coming to me and they are like, oh, Google, we want to rank on Google for these kind of generic keywords. And then I’m like, oh, okay, we can, we can do this, this, this. In like a small retainer, I can help you guys in this much per monthly for the next six months. But the best like ROI for you guys would be to start a TikTok or start Instagram or start YouTube.
So build an audience over there and push those audience to search for you and then build amazing pages. So I think that’s what I’m trying to do right now. Rather than being like an SEO expert for my clients, I’m trying to be like an advisor for them. And essentially I am doing like CRO optimizations on the website. So building audience on the TikTok and other social platforms, pushing them to search for the brand itself.
And then when they land on the brand, giving them amazing experience. And I think that Google is picking this up, picking this chain up and they are rewarding those kinds of brands. Because I’m like for the last five, six months, I’m working with like a clothing brand in Australia. Like it’s, it’s like really big, especially in women category.
And they are, I recommended them the same strategy and their traffic is like double, triple. And now they are beating like highly authoritative, really, really old, large brands they are building their site now for clothing related keywords. And it’s not like I build, go ahead and build like thousand backlinks or something like that. It’s just that I recommended that you start TikTok and you keep on posting, but make sure that every now and then you mention that people to search for you, not go to your website. Just mention that you can look it up, look at, look us up on Google.
So I think this strategy is working out really well for the brands. And especially in India, I think there is not enough, but YouTube shorts can be leveraged by brands to do the same.Â
Malhar: No, but again, so if you go back on the topic part of it, where the AI impact is coming in. So what you mentioned is an overall strategy, which is good for a brand, right? But when it comes to the building the brand with the entire impact of AI coming in, where do you see that this balance to come in, right? Because otherwise see the brand can get bigger on different platforms.
And then you’re kind of looking at traffic coming in from those platforms or that kind of driving search, et cetera, on Google, right? So where do you think that AI impact is coming in? Because otherwise it is more like a, you can say an integrated marketing strategy kind of thing that you mentioned.
Gagan Ghotra: Yes. I’m not sure what your exact question is, but if I am understanding it correctly. So you’re trying to say that if site owners are using AI, Google is like maybe saving their content.
Malhar: No, no, no, no. So what I am trying to say here is what you mentioned was an integrated marketing approach where you could leverage different channels, et cetera, to drive traffic to your pages or you can say, get people to search for your brand.
But if I go on the other side of the table and say, okay, how can brands start leveraging the entire AI mechanism which is coming in? Is there some kind of strategy there on how brands can start building content or even building their SEO strategy from that point?
Gagan Ghotra: I think if brands want to leverage AI, the first starting point can be for the content itself because content usually takes a lot of time and then building a content strategy also take a lot of time. But from what’s Google recommendation is that we should not be using AI content extensively, but if brands can leverage that to build the structure of the content and not the actual parts of the content. And I think it’s a good tool and a good use of resources of the brands.
Malhar: Cool. Thank you. But again, I’ll put like a big note for this one, even if you are using it for structure or even if you’re using it to just get a template for the blog post or something like that, maybe, maybe in future, Google may come up with another helpful classifier and say that, oh, you use it for structure. We don’t like it. So we will penalize you. So the best recommendation about AI is use it as minimal as possible. If you can do it by hand and just do it by hand.Â
Varun: But Google is also not liking content created by humans in most of the cases. So that is also a thing like that. Hi, David, how are you? If you can join us on this speaker, so we will love to hear your side also.Â
Varun: Hi, Odip. Welcome here. So moving on, like you said, and you talk about the omni-channel strategy where we have to leverage the side of the TikTok or the other social media channels, but that is a good strategy, but not every brand is like a pro who can create a content like that, a snackable content, which will lead to them to get some good views on the different kinds of social channels.
So, and there’s a question asked by the Malarbhai, how they can leverage the AI to stay relevant in the SEO section. So do you think that even if, like you said, don’t use for the structure or don’t make a template, but can we leverage AI in some different kind of way where we can help some SEO or do some kind of things which will keep us relevant in the search or in the, you can say, at least in the top 10 results? Do you see any kind of strategies or something like that?
Gagan Ghotra: Yes, I think if coming from SEO perspective, like if I can give a suggestion to another SEO, then it would be that I’ve been testing this with especially CatGPT. So if you see any page, what you can do is maybe think of it as like five or six sections, and then give the content of each section to CatGPT and say that this is section one, two, three, four, five, and then give it a description of your potential audience and mention some queries for which you want that page to rank.
And then ask it, like based upon this information, can you tell me what should be the most helpful order of these sections on the page? So it will come back to you and say that, oh, you have section one, but I don’t think that for this query, it’s helpful for a user to see it first. I recommend that they should see the section three first. So based upon that, you can shuffle the sections or the content pieces on your page to optimize it for your targeted queries and make it more helpful. I have actually tested this for one of my clients, and we have seen like significant improvements on this one. So this is something that you guys can also try.
Varun: Awesome. David, would you like to share some of your views on AI in SEO and SEO?
David: Sure. Thanks, Sam. I appreciate it. So I have three things to say on it. One is, I think AI and SEO is kind of confusing as a topic because there are people who think that there’s AI built into search. So in other words, AI is evaluating search, which I think is impossible. And I think if you look at Perplexity, Anthropic, ChatGPT, Gemini, Copilot, they actually all use PageRank or dumb SEO or dumb search, should I say, as in AI-free search, where pages are ranked according to PageRank. Even for Bing, which is the basis of ChatGPT and Copilot, that’s really a reverse engineered version of Google’s PageRank.
It’s 90% as close. If you rank in Google across 600 terms, you will have pretty much the same ranking in Bing. So that’s the same thing. DuckDuckGo is built on Google. Yahoo is built on Bing. I think there’s a couple of misconceptions about AI and where we are with AI in LLM, and that then feeds into other misconceptions in SEO that don’t really exist. So one is that AI can’t really make decisions like search. Search is actually, and most of the searches we do are very, very subjective in nature.
As humans, we’re very, very bad at telling between objective and subjective. We think a lot of the things we believe are objectively true, and that’s absolutely not true. A lot of the things that are objective are very boring, like what is the boiling point of water? What’s the capital of France? And again, when you dive into some of those things, you can actually find subjective differences within that. So a lot of the things that we believe in society that are highly conceptualised are very subjective, and there are no objective truths. And someone else, Mark Trapegan, or someone I was following, actually said that if you look at perplexity, when it tries to pick an answer, it actually starts to sound fascist.
And that’s the problem with using AI in search, is that you’re asking a search engine to make a subjective decision, which means picking a slant, which means having an opinion. And if you think, if you look at PageRank, Google’s done exactly the opposite of it. And nothing has beaten PageRank, right? And I’m not a PageRank fanboy. I’m an SEO. I compete with Google all the time. It’s the enemy.Â
In terms of where that also feeds into other misconceptions is where people think that content structure, or how you write, is how you rank. And that’s also untrue. That cannot be true, because that would mean that Google has to sort of make a side, pick a structure, or pick a strategy. And companies are free to communicate how they want. There has never been ever a study that’s shown that a marketing strategy, or a communication strategy, or a structure is objectively better than another structure.
Leave a Reply