← Voices of Search // A Search Engine Optimization (SEO) & Content Marketing Podcast
Voices of Search // A Search Engine Optimization (SEO) & Content Marketing Podcast
Mastering AI Search with Topical Authority and Links
00:29:42
Loading summary
Transcript40 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
A 2025 SEO industry poll revealed that 86% of SEO professionals have integrated AI into their strategies and 65% report challenges in maintaining topical authority alongside rapidly scaling content. So how can we better leverage emerging AI tools to drive further topical authority at scale at the enterprise? You're listening to the Voice of Search Podcast. My name is Tyson and today I'm joined by Daniel Hurwitz, SEO Lead at Informatica. Daniel will be exploring with us how to master AI integrated SEO through topical authority and strategy linking. Daniel, welcome to the podcast.
- [01:21]
C
Thank you Tyson for having me. I'm excited to be here.
- [01:24]
B
Yeah, I like, you know, when I saw kind of like the topic come through in our, you know, the application of the show, it really caught my attention because you were hitting on two, two areas that kind of spoke to me. One, the topical authority part, but then adding the complexity of scale for the enterprise level. So maybe to start off with the listeners, like how would you characterize some of the greater challenges that you're faced with at the enterprise level that maybe you're not faced with or it's not as present at a smaller scale, I.
- [01:59]
C
Would say it's all SEO, especially at the topical authority level. At the enterprise level is all about cross functional relationships. You're only as good as the people you know working with development teams, product teams, everyone across the board. I've just noticed how the more people I've gotten to know, the more people I've sat down with, the more expertise we've been able to leverage at the organizational level, the better our results are. And because our topic is so complex, like enterprise data management, it really requires bringing and our experts to validate and make sure our content is up to the highest possible standards while still scaling.
- [02:41]
B
Yeah, and I think the piece on the stakeholder management is it feels like the emphasis that the table stakes are almost like what needs to be done. The more complex and intricate problem becomes like how do you navigate the organization how do you build kind of support buy in? How do you get those supporting functions to be executing on your vision and your strategy? But it's like, I feel like more and more, a lot of the time comes towards the now, how do I get this implemented? How do I get this done more than the. Like, what do I need to get done?
- [03:19]
C
Absolutely, yeah. I think often at the enterprise level, the SEO role can often be like janitorial. Like you're just kind of like sitting in a closet, you know, updating some internal links, you know, optimizing some blog posts here and there. But, you know, to really scale in the AI era, you need to do a lot more than that. You know, know, fixing your meta descriptions and title tags is not enough anymore.
- [03:42]
B
Great transition point for that. Because I think, like, then the question is, with all the changes that are happening in the industry and stuff, like, how should we be looking about these kind of pillars and how are we using them to kind of leverage in that? So maybe from that point, like, how are you looking at some of, like the developments and the technologies that we're using, you know, more pointed towards this topical authority of a domain that you're working on?
- [04:11]
C
I think the best way to look at topical authority and, you know, especially when you're working at the enterprise level, where you have an established brand, is it's not going to just be your website anymore. Right. It used to be like, okay, you know, topical authority was like, all right, Maybe we write 10 articles on a topic, you know, we link them together. You know, we have a, like a hub article and then supporting, you know, kind of like this almost like HubSpot 2011 inbound marketing style. When I worked at an agency for seven years, that's exactly what we did for all of our clients. But now it's like, well, you know, people could be finding you on YouTube. So, like, a good example is like, I've been working very closely with the social team to optimize our YouTube presence because there's a lot of different stakeholders who have just uploaded things in different formats and they're not optimized. That's just one example. And then, you know, not to mention people going to, you know, ChatGPT, Gemini, Claude, et cetera, where search is just so fragmented now and there's so many different places that people can get that information. So it's like, how do you show up everywhere? That's the real challenge today.
- [05:14]
B
Absolutely. And with that, how would you describe the different components or pillars? What are the levers that we have to pull as SEOs to build or ensure the topical authority of the site that we're working on.
- [05:31]
C
Well, first of all you want to just like whether it's a small site or a huge site, you'll want to audit your existing content and go into your Google search console data and actually see like okay, like maybe I don't have a ton of not getting a ton of clicks on this topic but like am I getting impressions? You know what search terms are showing up, you know, maybe they're not keywords in the traditional sense, like you would go into semrush, but you know, these could be really relevant to your audience. And in AI search, you know, everything's so personalized so you know that can give you a great idea of things to optimize toward, you know, net new content to write. I mean we just built two different topical maps purely based on Google search console data and looking in, working with the SMEs figuring out okay, where can we get the best value where our customers are looking for but also where we already doing decently, where there's not a huge lift. So I would say that to me that's pairing that expert insights with our own first party data has been the biggest lift.
- [06:34]
B
And like from that though. So you have, you have those elements you have then targeting whether it's new content, optimization of existing content, there's you know, your internal linking kind of component from it from identifying those outside opportunities, placements. Like you mentioned YouTube. Maybe there's other social mentions that come into like play here. How are you looking at those off site, I don't know, locations so to speak. And like how would you be setting like priorities into which ones that you'd want to like prioritize and go after first?
- [07:11]
C
Yeah, I mean right now we do, I would say we do a lot with like webinars and things like that. So YouTube is big for us because we're an enterprise data management company. I mean look, I'm sure Instagram hasn't really been updated since like the late Obama era. So you know like that doesn't really matter as much. Even though. Yeah, I mean Instagram now is being pulled in search results but for an enterprise B2B company that's not really as important. But I would say YouTube, even some sentiment analysis on Reddit, we're starting to explore that as well. Even going into the data engineering subreddit for example, seeing what people are saying about us, seeing how we can meet those needs. Yeah, all those offsite things matter and then of course there's the digital PR component. Obviously as an enterprise company, that's something we always actively do. We have our CEO goes on Bloomberg, writes for Fortune magazine, et cetera. So like that element is already happening to some degree. So we don't have to do too much. I mean we do some link building, but we don't have to do too much on that level, at least to.
- [08:15]
B
Me there's like a separation in there that's more unique to the enterprise level, where I feel like maybe at the mid or smaller businesses you're treating this where it's like a target list, where it's just like, okay, I need to prioritize this task. And you're almost like working in a vacuum or a silo, where at the enterprise level, and then kind of like what you just shared is you have these other activities that may be already going on and then you have that piece of like, okay, how do I maximize these things that are already in motion? And it's almost like I feel like you're kind of like slightly course correcting or steering or like, hey, we can tweak this thing. Like you said, CEO going on Bloomberg, okay, hey, we can repurpose or utilize that in this other way. And so I feel like there's that almost like conductor element because you have all these different functions that you may not have control over, but it's like a low hanging fruit that you might as well go after something that's already in motion rather than trying to kick off something net new in the organization.
- [09:23]
C
Oh, absolutely. And that's where the cross functional component comes in. Because if you don't have those relationships with people, you're not going to know what's going on at every level. It's like I'm the only SEO at the company, right? So if I don't advocate for myself and my goals and make the relationships to make that happen, then nothing gets done. So that's the real challenge at the enterprise level for sure.
- [09:46]
B
And with the kind of evolution in where and how people are searching, is that shifting or changing how you're looking or thinking about topical authority or is it still intact in the same form that we're familiar with?
- [10:01]
C
Yeah, I mean, I would say the bones are still there, right? Like you want to take a topic, you want to cover it from a few different angles, a few different point of views, not just sure, obviously you're using keyword data to guide that, but ultimately you're trying to write about things that you think your audience is interested in. And you want to internally link those together. So that's the basic building blocks of topical authority. But then there's this element of you can't really track it in the same way that it used to be used to just be able to go into Search Console or we use Adobe analytics and get pretty precise performance metrics, but now not so much because of AI search. It's not even Google themselves doesn't provide information on AI overviews, AI mode, et cetera. So I almost feel like in a way, we're kind of going back to free digital marketing, where topical authority is like a billboard or something, where you're trying to put it out to as many relevant people as possible with topical targeting and content clusters and everything like that, but you're not going to be able to Measure, hey, well, 100 people clicked into this and 20 people gave their email or whatever. So it's much harder to track those performance metrics. And I think the whole digital economy has always been based on the idea of what gets measured, gets managed, and that's no longer the case.
- [11:28]
B
Maybe expand a little bit more on that last point because I think that's an interesting shift in how also we are working and that sense where it's like that last piece, so what gets measures get managed then I think becomes interesting because how we're measuring SEO in a lot of ways needs to evolve and go beyond kind of the direction that we've been leading the industry for so many years towards keywords. And so now it's like, okay, how do we revise that measurement to then reflect the strategies and the efforts that we want our teams to be making?
- [12:04]
C
Yeah, I think the big shift we've had to make is definitely looking at it more from a visibility perspective. There's been this kind of like, I think Ahrefs had an article about this a while ago, like the great decoupling between clicks and impressions, where, I mean, we've seen this consistently in our data as well as clicks are down and impressions are up. So what does that really mean? Right. Does that mean that you're actually showing up so many more times, or does this mean that like brand search is up? Is it a combination of those? You know, we've definitely dug into this and I would say it's a combination of that, that, you know, Google itself is being used as an increasingly navigational. So, yeah, like our impressions are up a lot, but how much of that is brand? I mean, definitely some. Not saying all of it by any means, but, you know, so we can Take that. You know, if there's increased interest in that brand, if people are finding out about our brand through other channels, then what can we do to leverage that instead of allowing. I'll give you a good example. One of our main product areas is data integration. So something like in the past you would write, say a what is data integration article, have a few derivatives of that, like what's a data integration strategy, Data integration metrics, et cetera. Now we're like, okay, well we actually get a lot more searches for informatica data integration as opposed to just like the generic head term. So how can we leverage that? Right? Like what kind of content can we put out when it's a branded serp? So like we can control what's shown to some extent. So I think that's the real opportunity where people are doing their research. They're not going to Google for the research stage, they're going to Google when they already have some level of decision making.
- [13:50]
B
In a lot of ways it feels like we're kind of swinging back to more traditional marketing in the sense where, you know, whether it was TV ads, billboards, it's like essentially they were reporting just on impressions where you're not having that same direct line of attribution. But I think like that becomes more and more of a play as more and more experience is happening within the search engine rather than on the website. But then I think it is clever too and I like pushing people in that like yeah, be maybe monitoring and reporting. Maybe it's not within the organization as a whole off the start, but if you're also knowing on how you're trending from a branded traffic to the site versus like demand of the brand as a whole. Because I could see if you're succeeding in those other placements, you're going to see that direct traffic or the branded traffic increasing, in which case that can be kind of a proxy or an indicator into how you're also competing in these other spaces.
- [14:54]
C
Oh absolutely. And I think as you know, as SEO is like, our job is to evolve to that changing user behavior where people are not going to Google for the research stage. And if they are doing any level of research, an AI overview is likely answering that for them. So there's just no, you're not giving the incentive for these large head terms that in the past you would pick, oh, well, this has good volume, we can rank for this, write some derivative content, et cetera. But now nobody. I'll give you a great example. We have an article on what is Master data management, that's another big thing, product area for us. I think we're still ranking definitely in the top three for that. That's been really consistent. Yet clicks are down year over year to that, about, I don't know, half maybe to that article because it's, and you know, we're featured number one in the AI overview. Like there's nothing else you could do from an SEO perspective, but yet, you know, you see your metrics tumbling. So, you know, what do you do? Well, okay, I mean, people don't need to click on a website now to define master data management, right? But they might be looking for like, you know, what are the use cases for master data management in generative AI? Like that's an example of an article we created and like, that's more specific, you know, that that might require more research than like an AI generated summary. Right? So like, you know, we've seen pretty good click through rates to an article like that. So that's just like one example of, you know, it's our job, I think, not just to, you know, educate on the super high level. Like, you know, it's like every, every website is almost like a publication or like an encyclopedia. Even if it was like, not necessary, you know, like the glossary trend and all that. But now it's like we're there to influence the purchase decision. And to your point too, the more you come up, like a TV ad, let's say you need a new car, you'll need it eventually, but your car is fine, but you'll need one in a few years. But if you saw 100 ads for Subaru in the last two years about, oh, this is the safest car and oh, I know I'm going to have a kid and I need a safe car. And then that's top of mind for you when you make the purchase decision. Even though, you know, you can't attribute that as like, oh, somebody clicked on the thing and then they gave their email and then they were contacted by a salesperson and you know, the classic kind of like software model, you know.
- [17:20]
B
With it though, it made me think too. It's like, I feel like, especially on the kind of like enterprise brand side, there's probably a lot more room for kind of this like innovation into like thinking almost a few steps ahead in that buyer's journey as you described. Because I'm thinking like, you made the Subaru piece would totally make sense for someone in Subaru to also be thinking of like, what are the activities then of our target Demographic. And so it's like, okay, Subaru owners love camping. So then what are the type of queries that people are searching for? Whether it's, you know, best up and coming camp spots of the year. It's probably a bad example, but maybe it's something that's more, you know, towards cars, like car camping. And then there's something about roof tents and Subarus. I don't know, kind of on the spot, weird topics. Right. But like you could be not just looking at that lagging piece but also like how are we crafting the next 12 months, 24 months of a consumer demand aspect within this touch point. And I think more and more of the practice to me is like falling towards this consumer brand. Like how do actual users respond to these emerging inputs? But to me that's an interesting kind of add on into our strategies with the changing environment.
- [18:46]
C
Definitely. And to your point, search is inherently, it's already becoming more conversational and I think it's only going to continue as AI mode gets rolled out more broadly into Google. People are Starting to use ChatGPT Search now. We've seen in our performance metrics for the first time getting a significant amount of referral traffic from ChatGPT this past quarter. We weren't seeing that before and these are really high intent people that convert really well because they've already talked to an AI system, they've already done their research and they're kind of ready to purchase at this point. So the example of Subaru. Well, yeah, if somebody was looking for, to say, hey, I need a new car and I love to go on road trips and camping, what are some good models, you know, blah, blah, blah. And then they keep coming across Subaru. Well yeah, I mean they're pretty prime to go right. By the time they. And then they're going to go to Google and just type in, you know, best Subaru for car camping or whatever. And that's a branded search. Right? They already know they want a Subaru at that point.
- [19:49]
B
Yeah, it's like am I getting the Forester or the Outback? Like where am I landing in the, in the family?
- [19:53]
C
Yeah, exactly.
- [19:55]
B
Well and it's like I think too because it's such a new space that it feels like there's almost like a guerrilla marketing type opportunity where it's like I remember back in college, you know, I liked a lot of the stuff like Dax Shepard was doing with like Obey and had that guerrilla marketing where you know, it was just getting the name up, having the placements, having like Leaning into the street art kind of component to it, which then generates into this longer term demand for the brand and stuff. And so I think especially as a lot of the bigger companies are still figuring out how to address and how they should be competing in these new spaces, it kind of feels like there's a little breath of fresh air for the smaller, more nimble players to kind of, you know, latch onto those placements and that guerrilla marketing type placements online as well.
- [20:48]
C
Oh, absolutely. I mean, to your point of the guerrilla marketing, it's probably the biggest opportunity for especially smaller brands since TikTok, I would say, where like, you know, if you're a small brand, where like, maybe you wouldn't be able to compete super well in, you know, Google search, in traditional Google search because of, you know, okay, brand is like one of the biggest elements and you know, who competes especially on like larger keywords. Well, now you can show up for all these like more niche conversational queries where like, you know that that wasn't an opportunity for you before. Just like in TikTok where there's like the, for you algorithm where like, if you're a tiny creator, you know, you can get a, you can get traction just by posting things that are really relevant to what, you know, people have.
- [21:32]
B
Already searched for and the, the diversity of results. I, I feel like we're going to see a resurgence of having, you know, a bit more variety in that sense where it feels like it's been a consolidation to the larger brands, but with having like longer prompt or longer inputs to it, it's going to inherently produce like more varying outputs and response. And so more and more that like niche opportunity and focus where before everyone's being like, hey, we're looking for, you know, running shoes. And the query might just be running shoes or best running shoes or like a handful of queries. Now if you take that same topic and put it into 100 people, put it into your own prompt, you're gonna get more variation by nature, which I feel like we're gonna see also a result in who shows up and who can win in each of those placements.
- [22:30]
C
Oh, definitely. And I mean, you know, with running shoes, for example, there's a huge difference between like running shoes for marathons or running shoes for someone who takes a one mile run each morning. You know, those are, and like, you know, you could sit there and describe in a whole paragraph of like what you're looking for and you know, you might get recommended a shoe you've never even heard of. You Know, there's not. There's more than Nike and Adidas, you.
- [22:52]
B
Know, So, I mean, we've been spending a fair amount of time in this. Like, how are we thinking about, like, where we're competing and how it relates into the kind of like these efforts? What areas are you feeling like you're using some of the, maybe like new workflows or new kind of processes for some of these areas, whether it's content creation, link building, like how. How are you evolving the tactics and execution with additional tools and opportunities?
- [23:24]
C
I would say in terms of content creation, we still do human created content. And I'll tell you, I think for some brands it can work to do a little bit more of a hybrid. And I'm not saying you could definitely use it for optimizations and things like that, but because our audience is so technical, we need to make sure that everything is really reviewed by experts. Right. So this is not something about the best places to hike in Denver, where there's only so many places to hike, and it's going to be similar so that we're still doing human elements. We've definitely explored using AI, but it's not, I think, still not quite there for us, especially as a more technical audience. That being said, AI workflows really come into play for things like topical. Actually, just before this call was working on a topical map with uploading our Google search console data, having it go through refined patterns where we're showing up really well, where we can. And that can really guide our outlining process because we can actually use all that GSC data to say like, hey, we're already showing up well for this, that and that. How can we leverage that and put it into our outlines and then we can pass it off to our product teams who have their own campaign writers and things like that, and they can actually write the articles.
- [24:50]
B
Interesting. Can you elaborate on that maybe a little bit more in the sense of. So you were taking the outputs from gse, modeling it into the topical authority. And then from that point, are you then ingesting performance numbers to see coverage in each of those areas or what would be the next step?
- [25:12]
C
Yeah, we're looking at impression data and showing like, hey, where are we showing up? Well, so if something's like, oh, we have five impressions or something. Well, that's not really. We're not seeing as super relevant for that. But if we have 1,000 impressions or something, well, okay, maybe it's just the case where we don't. We have five different articles that are on page eight. For this term. But we don't have a dedicated article. And okay, if we create that article and then we create opportunities to link to that article within our existing stuff and optimize a little more, then we can show up really well. So I would say it's really finding though I feel like where AI excels the most is pattern recognition. And if you're a human you might miss that. You might just go through the data of you might create a regex and go through and say, oh that's interesting but not really pull out the data and then find insights from there. So finding those patterns, that's where I would say the biggest lift for us with AI has been.
- [26:12]
B
And with that too, it's guiding that area of where to look, where to work. And so it's like I like that utilization in the sense of comb through all this, surface out the areas that I need to then put more human touch into rather than the like I'm going to entirely automate this one task and then never think about it again.
- [26:34]
C
Yeah.
- [26:34]
B
Or I think like that at the current state is a strong. And it's like a. Especially if maybe someone's like newer to utilizing things or utilizing these systems. Like to me that feels like a really comfortable starting point rather than being like I'm going to automate my entire content production. That'll be my first project with Gen AI.
- [26:54]
C
Yeah. And I think some companies have, we've spoken to other SEO teams like that and I've spoken to them personally and I think some companies are doing that, but I don't think that's a long term strategy, honestly, unless there's some sort of quality control. I know that there are some platforms that are helping with it now and maybe they're good. I haven't tried them to the degree, but I just know that this is working for us and we're going to continue it. But again, we have a technical audience and we need to make sure that expertise is leveraged at the organizational level. Otherwise, you know, because it's not just about, oh, getting the most possible clicks. Like we're not a publication, you know, where we're relying on affiliate income or ad revenue or something like that. You know, we want to make sure the clicks we're getting are actually relevant.
- [27:39]
B
Kind of going down that path just a little bit further. Like for someone that's maybe newer to the industry and maybe they haven't been utilizing their know in their head, they've been listening to maybe this show and they're like, okay, I have to get on this, I have to start utilizing, like, these systems. What advice would you have for, like, starting points or areas to focus on in kind of like early adoption days?
- [28:04]
C
Honestly, just, I mean, if you're just talking about using AI, I would say learn how to build any workflow. You know, go into, you know, take a piece of content and put it into Claude and say, like, hey, like, can you help, you know, look at my domain? Can you help me come up with 10 internal linking opportunities? You know, that's an easy one where you could say, okay, and then maybe there's not a place where your keyword for your article or your topic is directly referenced, but you could say, like, hey, these are relevant topics to this. Can you find me 10 URLs on the website that are similar? And then we can go through together and maybe you could use Claude to rewrite a sentence or something just to link it in. And I think that's a really easy first step. But internal linking is one of those things where I think it often gets overlooked because it's kind of tedious. And I think, like, people think, oh, well, people don't really click on those links. You know, it's not. Sure, it's definitely for humans, but it's not just for humans. You know, it's for, you know, search engines, AI crawlers as well, to create semantic relationships between content. So, yeah, if I just was to do one thing with AI to get started, that's probably what I would do.
- [29:16]
B
Love it. Well, with that, that's going to wrap up this episode of the Voice of Search podcast. Thanks again to Daniel Hurwitz from Informatica for joining us. If you'd like to contact Daniel, you can find a link to his LinkedIn profile in the show notes, or you can go on and visit his company's website@4matica.com if you haven't subscribed yet and you want a daily stream of SEO and content marketing knowledge in your podcast feed, hit that subscribe button in your podcast app or on YouTube and we'll be back in your feed in the following day. That's all for today. Thanks for stopping by and we'll see you in the next episode. That.