Tough Questions CEOs are Asking About AI SEO (and How To Answer Them) w/ Gaetano DiNardi

They are asking if traffic is going down, how do we know that this is working? How do we know that SEO is

working? How do we know that SEO is working because traffic is going down?

And >> what's your answer?

>> I hit them with another question.

>> Oh, >> which is how would you know if it's working if traffic is going up?

[music] [music] >> Gano, I appreciate you coming back. It's

round two. Yeah, Ross, let's get it.

Excited to be back. Thank you for having me.

>> Yeah, the last time we chatted we were in we were in San Diego maybe. Oh, no.

We were out here. Yeah, we were just talking about that a little bit.

>> Yeah, but I did see you at AHF, so that was cool.

>> Yeah, that's true. That that triggered the thought process. I got to have this guy back out. [laughter]

>> Yeah, I saw the email come back on the pod. I was like, "All right, hell yeah.

pod. I was like, "All right, hell yeah.

Let's get it."

>> A few things have changed, a few things to talk about. Fair to say? Yeah.

>> Yeah. In fact, um the last time I was here, the topic was growing organic traffic from zero to like massive levels.

>> Yeah.

>> And sadly now, or maybe optimistically, however you look at it, >> that is irrelevant.

>> So today's uh topic is not going to be about organic traffic growth. It's going

to be about a lot of other things.

>> Yeah. You can't even say that you'd almost be like a charlatan with that title today, right?

>> Oh yeah. If you came out with that title now, people would just roll their eyes at you and not take you seriously. Yeah,

pipeline maybe. But

>> yeah, >> so we're in a wave of we were talking prior to the recording like everyone's getting questions from executives about AI LM visibility, where am I here? Where

am I there? Like there's so many variations of these tough questions people are getting and you've put out some great content on your Substack, on LinkedIn.

>> Thank you.

>> covering a lot of these that people should check out. We'll have it in the show notes. But

show notes. But >> yeah, starting there. I mean there there are examples of people who will put out there actual SEO programs that have undoubtedly just gotten nicks

completely.

>> Like what is that? Like what do you think is the cause of that most frequently? What's that? And maybe it is

frequently? What's that? And maybe it is a misconception, maybe it's not that's causing a CEO to actually absolutely nyx a program which is wild. [sighs]

Yeah.

>> Yeah. Well, the biggest factor is that traffic is going down. So the top question that I'm getting at the moment um now I will frame this before I get into that. I will say that all my

into that. I will say that all my clients are B2B SAS like around the mid-market level. So you got to uh keep

mid-market level. So you got to uh keep that in mind that I speak from a place of um mid-market B2B SAS context.

>> Okay. So maybe other segments see different kinds of things, patterns, trends, problems, situations, but we're talking B2B SAS

>> and um from from specifically the executive level, they are asking if traffic is going down, how do we know that this is working? How do we know

that SEO is working? Because traffic is going down and >> what's your answer?

>> I hit them with another question. Oh,

>> which is how would you know if it's working if traffic is going up?

>> Oh, wow. Okay.

>> Think about that.

>> Yeah. What's the the the line of questioning? Socratic method.

questioning? Socratic method.

>> Yeah. Yeah. Socratic method. Yeah. So,

so if traffic is going up, how would you know if it's working? Or let me frame it like this. If traffic was going up but

like this. If traffic was going up but revenue and leads and pipeline were going down, would you still be happy about the SEO program?

Whoa.

>> So they the answer out of should be they shouldn't be, right? [laughter]

But sometimes they are or they were is probably the better descriptor.

>> Let me put it to you like this. A lot of times they don't even know. A a lot of times, right, traffic is going up and everyone's clapping. Yay! Wow, traffic's

everyone's clapping. Yay! Wow, traffic's

going up. Everyone claps. There's

momentum. Someone posted our traffic stats in Slack. You know, there's a screenshot on LinkedIn about traffic going up and everyone's just like, "Wow, like this must be correlated to revenue

somehow." And um that's another thing I

somehow." And um that's another thing I wanted to bring up. I actually had a CFO say to me, you know, there used to be a very tight correlation between this

linear flow keyword rank position which leads to expected CTR which leads to traffic coming to the site. We can then

you know look at conversion rate of traffic going to the site to lead and then that would be correlated to revenue. we would be able to, you know,

revenue. we would be able to, you know, formula all that out. With the advent of AI overviews taking away a lot of that traffic and traffic not being the

primary thing that we can really use as a leading indicator, now that whole linear equation is breaking.

>> Yeah.

>> So, getting back to the question, how do you know if traffic going down? How do

you know if SEO is working if traffic's going down? It's the same way you'd know

going down? It's the same way you'd know if traffic was going up. demo requests

pipeline and the overall performance from the organic channel, >> right?

>> That's what you need to be looking at.

So, it's it's changing the mindset of traffic, traffic, traffic to let's look at holistic overall organic channel performance because that's the way things are going.

One thing you noted in, not to lead the the horse here, but I remember seeing in one of your posts that a lot of organic and LM may influence

conversion either direct or maybe through navigational search on Google, but it's hard to see that. So there's even situations

see that. So there's even situations hypothetically where organic conversion is flat, but if we can fairly communicate something,

the visibility is up, that could still be a positive, right? Yeah.

>> Yeah. That there's a lot of factors now that are different than before. So maybe

let's start with like traffic being flat.

>> Yeah.

>> As an example, traffic being flat, I view that as a good thing now.

>> Yeah, for sure. Um, if you're one of the few that's not getting decimated, that means you're doing something right. Now,

I have a lot of examples of this, but one thing that comes to mind is you have a lot of top offunnel decay.

>> Yeah.

>> So, that's where a lot of the decay is coming from from my experience. Um, a

lot of like the slash library content slash glossery content slash FAQ content really top offunnel consensusformational

stuff just getting decimated.

I've made the case to let's stop chasing our losses on traffic decline and let's just shift to bofu. Let's just shift to

bottom of funnel even though it's longer tail, less volume. Even the query fan out method I've been uh approaching content strategy with a bit of a different mindset. Okay.

different mindset. Okay.

>> Um lately um >> so what's happening is that traffic decline from top ofunnel is going down at an aggressive rate. Yeah,

>> we can't recoup that from the BFU gains that we're making because that's more longer tail and lower volume.

>> But if brand is helping us bridge that gap a little bit, then everything is maintaining at least at a baseline level or at a flat level to an extent.

>> So, we can't really make up the lost ground, but we can almost kind of stay at baseline uh flat levels. But if all that's happening while revenue is going up and traffic is flat, that's actually

great. Makes sense. Yeah, that

great. Makes sense. Yeah, that

decoupling you described and this is reminds me of another post you made on LM kind of rank tracking and visibility.

There's a decoupling in the middle of that, but I almost think about bottomfunnel rankings. There's even an

bottomfunnel rankings. There's even an element on organic search where if you you show up in those kind of like accordion I don't know the technical name like if you did a best CRM software

you often see the nine links or whatever volume that have the you open up the external links you may not even go to those external links but it's a high position

there's almost like a decent link maybe more than before between keyword ranking of on transactional and actual demo requests across both surfaces, LMN.

Would you agree with that?

>> Yeah. Yeah, I would definitely agree with that. In fact, um, with a lot of

with that. In fact, um, with a lot of the clients that I'm working with now, >> we want to do content that lends preference to LLM rankings. So, there is like a favorability factor with certain

content types like the X best list, um the comparison content, but I've also noticed that like a lot of LLMs are so good at specifically solving an intent

for a client or for a prospect that it can go to like more product content, um more feature content, help docs, uh integration pages, which through like

traditional search that doesn't really happen unless it's like branded intent.

And so, um, we are doing a lot of comparative crash the party pages as well. So,

well. So, >> crash the party pages.

>> Yeah. Yeah. Crash the party SEO. It's a

cheeky technique. [laughter] Um, but I I I'd like to think that I've mastered it.

It's where you compare two brands and then you insert yourself into that brand comparison, but you don't necessarily bash the competitors. You just position

yourself um as an appealing option where you highlight your strengths rather than really like highlighting their weaknesses so much. you just really make yourself a more appealing choice.

>> So is a hypothetically you're a Microsoft Teams and Slack competitor. Is

that a Slack versus Microsoft Teams article?

>> Um yeah, that's that's the intent that you'd be >> on your website.

>> Yeah. Yeah. On your own website. Yeah.

Like let's just say you are campaign monitor.

>> Yeah.

>> Um and you want to do Active Campaign versus Mailchimp. You would do Active

versus Mailchimp. You would do Active Campaign versus Mailchimp versus Campaign Monitor, which is yourself.

>> Okay. So three-way. a three-way

comparison. Three-way comparison. Yeah.

I mean, the X versus Y is a no-brainer.

You should be doing that no matter what.

>> Um, but the three-way comparison is also very interesting.

>> Why I thought that was a unique angle, which there may be there's pros and cons, too. I agree that it's a

cons, too. I agree that it's a no-brainer to do you versus competitor.

Yeah.

>> The variant of >> obviously there's also people doing competitor versus competitor in isolation >> that maybe you're capturing with that search. which I maybe that's the idea.

search. which I maybe that's the idea.

>> Exactly. Exactly. Yeah. You're

piggybacking off of two other brand searches, >> right?

>> Yeah.

>> And I guess that's the very clearest transactional way to insert yourself.

There's pros and cons of being able to rank for that by diluting the topic, right? [laughter]

right? [laughter] >> Yeah. Yeah. Well, you can only rank so

>> Yeah. Yeah. Well, you can only rank so so highly for stuff like that because those other brands, if they're doing, let's say, ads, then those brands are also doing, you know, protective pages.

then you can maybe get to if we're talking in a classic sense position three or four because you got to imagine there might be affiliates as well. Yeah.

>> But I found that the affiliates are doing programmatic content >> and so if you're doing more of this handcrafted you know tailored content that's like way more comprehensive and it has a lot

more love going into it. I've seen those outranking the G2s and the Gartners because they're just putting programmatic data together and it's all slapped together and it just it works on the ultra longtail stuff because people

don't do the handcrafted content.

>> Yeah. Yeah.

>> But for the bigger brands, you are going to see a lot of the handcrafted content.

>> Yeah. It's interesting. I mean, I agree.

Even if you can't rank, >> there's going to be some consumer pessimism of like, is Slack going to give me a fair assessment of me versus team them versus teams?

>> Yeah. that you may actually have above average click-through rate curve for a lower position.

>> Yes. Yes, you might. And I think you have to be truthful and not so um exaggerative in a lot of the claims that you're making on these pages. Um I think the legal terminology is puffery.

>> Okay. [laughter]

>> You have to avoid puffery. But um one technique that I like to employ on these pages is citing everything that you're saying backed up by a third party reference or claim.

>> Okay? So like you know you can find what people are saying on Reddit about a certain tools features or aspect that they like or they don't like uh things that they don't like about the pricing model pricing changes. Um you can find

anything on you know any third party site and use that. Now you don't want to do it in a way that seems cherry-picking like let's just say things people hate about Zenesk and you do like 20 negative things in a row and you just cherrypick

all the complaints. Like you don't want to be doing stuff like that cuz it seems too one-sided and not believable. But if

you find the nice balance, I think it can really work well.

>> Yeah, I I agree. It's kind of just I we'll often frame it to our clients like it's more about finding the people you are absolutely the best for than just like disparaging the other person.

>> That's right.

>> That's right. Yeah. And a way to like not go down the rabbit hole of should we do every possible competitor out there who even overlaps with us a little bit, you know, cuz that could end up being,

you know, too many pages that are not going to attract the right people. You

can rely on Google Ads data.

>> Yeah.

>> And auction insights because that will really, you know, narrow it down to the relevance of the brands that come up maybe during competitive deals or who are bidding against your brand name and stuff like that.

>> Okay. No, that's a good tip.

>> Yeah.

>> So, you have encountered a unique situation. I mean, everyone's getting

situation. I mean, everyone's getting asked about LM visibility >> tracking on profound peak scrunch, your tool of choice.

>> Yeah.

>> And there's these dashboards that you create based on the prompts you input.

>> Mh. You've seen a few executives get misled based on what they're seeing there. Could you explain that situation?

there. Could you explain that situation?

>> It can be very misleading. Um, and it reminds me a lot of how organic traffic used to be blended.

>> I don't know. One executive calling you.

>> That was an exe. That was actually a client calling me like >> Oh, really?

>> Yeah. Yeah. Uh, that's rare that they would call me out of the blue like that.

>> We should have taken it live.

>> Should have should [laughter] have taken a hive. Hey, what the hell's going on

a hive. Hey, what the hell's going on with our rankings? Um, we should we keep that in there. We can keep it in. I

don't care. That actually was a client.

Um, that's hilarious. [laughter]

Um, >> talk about LM visibility.

>> LM visibility. So, um, one one more important point that I wanted to make actually on the comparison content was this is stuff that really shows up a lot

in LLMs because of the grounding factor.

So, I'm not a fan and this leads into the prompt tracking. So, this is a nice segue.

>> Um, I'm not a fan of tracking prompts that do not recommend a tool and do not

have website citations in the answers.

So, for example, if it's a very basic generic question about like how to, you know, do ABM for B2B marketing, it's just going to spit out the answer from its training data. Step one, do this,

step two do that, step three do this, right? And there's going to be no brand

right? And there's going to be no brand recommendation. It's not going to say

recommendation. It's not going to say try this tool. It's not going to say, you know, check out these websites for more information because the goal of the LLM is not to send traffic. It's to give you an answer.

>> Yeah.

>> And so if you're tracking a bunch of prompts where the answers basically just getting scraped out of the training data, there's no chance of you getting a

citation. There's no chance of you

citation. There's no chance of you getting a brand mention. If you have a bunch of that stuff in your prompt tracking set, you're going to make yourself look bad because the LLM

visibility dashboard is going to say, "Oh, here's all these prompts where you're not showing up."

>> Yeah. And you're paying for prompt tracking you don't need.

>> And you're paying for prompt tracking you don't need.

>> Yeah. Yeah.

>> And it causes unnecessary questioning.

So, it's like, why don't why do we have such low visibility in ABM category prompts? And it's like if you click into

prompts? And it's like if you click into it, it's all prompts that nobody has visibility in.

>> There's no visibility for anyone.

>> Yeah.

>> And no citations for anyone.

>> So that's one big thing for people to keep in mind. The next point is um on the tracking. So this reminds me a lot

the tracking. So this reminds me a lot of organic traffic blended.

You used to be able to fool an executive. You used to be able to show

executive. You used to be able to show overall organic traffic going up and them not knowing that 70% of it was brand. You just did an event. You're

brand. You just did an event. You're

running, you know, um, airport ads in Austin. Your CEO was just on a very

Austin. Your CEO was just on a very famous podcast. You just maybe got a

famous podcast. You just maybe got a round of funding. You know, whatever it is that caused that brand search to go up over time. Um, SEO teams were taking

credit for that falsely.

uh it wasn't non-branded growth that was attributing that traffic growth. It was

the overall picture. Now, the same thing can be happening with LLM prompt tracking. If you just show an executive

tracking. If you just show an executive a dashboard of your LLM visibility versus competitors, but like 50% of the prompts you're tracking are brand

questions.

Like if I'm Zenesk, what's Zenesk pricing model? Of course, you're going

pricing model? Of course, you're going to be the top recommended citation. Of

course, you know, your brand is going to be all over that uh answer. So, if you have all those branded prompts in your tracking set and you are not segmenting

non-branded versus branded, you will be misleading the executives because you'll be showing them a faulty dashboard that has branded visibility and you can

basically fudge the numbers in your favor to make it seem like you're getting more visibility than you're not.

And most executives don't know how to ask the right questions to to understand that.

>> Yeah, it's a great point. It makes me some tools. I know profound does this

some tools. I know profound does this probably intelligently.

There's probably still value in tracking those depending on the company for sentiment. So sentiment neutral

sentiment. So sentiment neutral negative.

>> Oh yeah.

>> But if they [laughter] a Yeah. aggregate

them with the non-branded terms and 100% they're going to feel not so great about that. That's right. For branded prompts,

that. That's right. For branded prompts, my view is that you should be focusing on sentiment.

>> Yeah.

>> Um and looking for any negative things that might come up.

>> Yeah.

>> Um and finding a way to do online reputation in the event that negative stuff comes up.

>> Yeah.

>> But for but for non-branded, I think this is another important point. There

should not be tons and tons and tons of prompts that you're tracking. Like you

shouldn't be tracking 500 prompts.

Probably not even a hundred. Like for

most of my clients, just like I had money keyword lists for classic SEO, I have money prompts. [laughter]

I have money prompts.

>> We need another name. Maybe that's it.

>> It doesn't sound great, honestly.

>> Yeah, there's something. We can think of something.

>> I got to think of something better. I

don't know. But yeah, like basically it's the prompts that are going to recommend a tool >> or uh recommend a solution um for any you know what is the best software that

does this, what kind of tool does this, what are the best alternatives to this, anything that would lead to um a purchasing decision essentially. Yeah, I

like starting on just creating a map of their features and sol features solutions maybe industries and like what is the best and then add that different

thing and that's like a pretty sizable list. Sometimes there's long tail, not

list. Sometimes there's long tail, not on that, and you don't get competitors in there. But

in there. But >> yeah, I think you could actually go kind of crazy with this. Like depending on, you know, how many personas you sell into, like for example, um I think IT teams,

>> they might be looking for specific information on like integration documentation.

>> Yeah.

>> Whereas in classic search, you know, let's just say a big deal is like Snowflake integrations. They might just

Snowflake integrations. They might just search like, you know, what tools integrate with Snowflake that do this.

But for LLM prompts, they're going to get way more specific with that. They're

going to, you know, probably look for, you know, ETL documentation for data pipeline between these two things, >> right?

>> You know, and and so um depending on how big of a factor that is for your for your purchasing journey, um you you got to cover your ground on. Yeah, that's a good call. Yeah. So, another post you

good call. Yeah. So, another post you just recently published was comparing Chat GBT conversion rates and Google conversion rates and maybe some misconceptions there. Could you dig into

misconceptions there. Could you dig into that a little bit?

>> Oh, that's a big one. Um, so first I will say that there is a study out there that corroborates whatever narrative you want.

>> I've seen, you know, a different one from AMS. I've seen a different one from Seir Interactive, >> Siege Media.

>> Siege Media's got one. [laughter] Yeah.

Um, who else got one? Search Engine Land put out one on like e-commerce.

So, so the point is like you can find a study >> anywhere you want because it's out there that will corroborate the narrative that you want.

>> I got to start with that because this is a B2B SAS midmarket example. And so this may only be relevant to B2B SAS midmarket, >> but it's still a finding that is

consistent. it's a pattern that I'm

consistent. it's a pattern that I'm seeing that's consistent as I look at more B2B SAS midm market because that's really all I work with.

>> So if that pattern is consistent across like every client, then I've got to believe there's something there.

>> Yeah.

>> So I'll start by saying that um you know I didn't want to lead by framing it that this is the absolute answer that is you know blanket across sweeping you know for everybody

>> and most of your clients are sales assisted right not self-service. They're

pretty much all sales assisted. Um, you

know, some of them have the PLG baked in with sales assisted, >> but the majority of them are sales salesled content. Uh,

salesled content. Uh, >> salesled GTM.

>> Yeah. Yeah. Yeah.

>> Salesled content. Yeah. [laughter]

>> Um, all right. So, where I was getting uh with that was um conversion rates on Google versus chat GPT. what I'm seeing

first of all the um number of conversions from uh chat GPT compared to Google in like a given period might be around the same but the amount of

traffic that is going that is being referred to the websites from chat GBT is so small it's like a tiny tiny fraction of what Google search is sending

um so by default chat GBT is going to have a higher conversion Right? If you

think about it, >> there's less traffic that's so much more targeted, so much more specific.

>> That conversation between the chatbot and the the user is just going to go back and forth and back and forth, narrow and narrow and narrow until they get what they want. And then if they end up clicking, which that's the only way

this attribution works, they have to click. The conversion rate is likely to

click. The conversion rate is likely to be high.

>> Yep. But there's also like a broken darkfunnel SEO thing going on where a lot of people are not clicking from LLM's and going to the website. Maybe

they're doing more research or they have to corroborate through Google to, you know, validate what the LLM is telling them. So, they're going to end or pause

them. So, they're going to end or pause their journey through LLM and they're going to go to Google, probably search the brand directly, go right to the homepage, poke around. Maybe they'll

convert on that session. Maybe they got to do even more stuff. Like the point I'm making to this level is everything is so fragmented that you're not going to be able to have that clean attribution that you want.

>> But getting back to it on the conversion rates. So Google sends so much tofu

rates. So Google sends so much tofu traffic. The amount of top of funnel

traffic. The amount of top of funnel coming from Google is automatically going to dilute the conversion rate from Google search. So, if you just look at

Google search. So, if you just look at conversion rates, sure, pretty much every study is going to say ChatGBT has way higher conversion rates than Google

search. That just makes sense by default

search. That just makes sense by default because of the way the systems work.

Chat GBT sends zero top of funnel.

>> Yeah. Zero.

>> Yeah. Zero. Nothing.

>> Zero.

>> Imagine if Google sent zero trap top of funnel.

>> We may get there eventually.

>> They may get there eventually. Yeah.

With AI mode, that's probably the way it's going. Um, but if Google gets to a

it's going. Um, but if Google gets to a point where there's zero top of funnel, then the conversion rates might end up being on par eventually. Yeah. But but

for now, because of the the nature of how targeted chat GPT and Perplexity and all them are, um, the conversion rates for LLMs are going to be bigger.

>> Yeah.

>> Just by default.

>> Yeah.

>> However, when you look into the deal sizes and the numbers, the difference that I'm seeing is staggering.

>> It's staggering. Like in this one client example, the one that I published on my Substack, um it was like average deal size around 200K from Google search and then average deal size from LLM 5K.

>> Wow.

>> So that could lead us to believe that, you know, there's more tire kickers using these uh LLMs, more tire kickers going through the LLM funnel, going to

the site and converting. Um there could be some other explanations too like maybe the conversions from Google search are better informed because they have

seen you at an event or because they've uh seen your ads more or because they um you know came from an affiliate source that went through Google who right like that that could also be it

>> um but generally the data is showing for me on the client side for B2B SAS midmarket that Google leads are way better in terms of deal sizes and and you know, maybe we need to look at

self-reported attribution to maybe close that gap a little more because maybe big deals are coming in, but we're not getting that qualitative level feedback like on the form. How did you hear about

us? Oh, uh, you know, I was using chat

us? Oh, uh, you know, I was using chat GPT and, uh, I narrowed you down and then went to the website.

That's something that's kind of an unknown. So, just like dark social, you

unknown. So, just like dark social, you know, the Chris Walker term that was so famous, I I think we're seeing that now with um

>> do I dare say darkfunnel SEO. [laughter]

>> Yeah. The we have that open attribution on our own website. Um

>> yeah, >> and I guess I would capture some of it, but I mean the the we're such a unique dynamic though because people are want

to find people that are doing geo through GEO. That's kind of

through GEO. That's kind of hypothetically. Yeah.

hypothetically. Yeah.

>> But it's very similar to the previous note about SEO services through Google.

>> Tire kickers.

>> Tire kickers >> getting a lot of volume undoubtedly that found us on chatbt. But there and there's quality leads. We've gotten

Fortune 500 level range through those channels. There's but

they're highly comparative.

Um still early, but I mean that doesn't shock me what you said. Yeah,

>> I would say that another finding is like the biggest deal like there's there's going to be whales that come in that throw the averages off.

>> So like for example, this client with Google search, there was a couple whales on the high six figures, right? Um but

even the the biggest whales on Google are much much bigger than the biggest whales on AI LLM leads.

>> Yes. So like the biggest whale we saw for this particular client through like Chad GBT was like a 30K.

>> Okay.

>> And and that's considered a whale because everything else is so small from from AI.

>> Yeah. And that's actually fair. The

companies that were big that came through were still small deal size.

>> That's right.

>> That's right. So I mean unless there's a way to close that gap on the dark SEO funnel, I guess we'll truly never know.

Um, but yeah, I mean it's just leading me to this outcome of SEO attribution is cooked >> in the AI era.

>> So what are you saying to executives are like that's the case?

How do I even know to invest in this channel? Like what's your kind of

channel? Like what's your kind of reaction to that question? Is that a question you >> Well, it's still a channel.

>> Yeah, it's still it's still a channel.

>> It's small. Uh potentially it has more growth upside in the future. Yeah,

>> you know, it's still emerging as the days go by. You know, maybe adoption gets stronger. Uh maybe things become

gets stronger. Uh maybe things become more agentic. Um another way I've been

more agentic. Um another way I've been positioning it is you can probably start thinking about AI LLM the same way you do as social platforms in terms of the

zeroclick stuff, right? It's it kind of functions in the same way. They don't

social platforms don't send traffic.

It's zero click. There's not much attribution that happens. a lot of people.

>> It's Yeah, it's eerily similar.

>> And there's a lot of underinvestment recently in those channels.

>> A lot of underinvestment in social. I

mean, would you stop investing in social? You wouldn't.

social? You wouldn't.

>> Yeah. People who get it. I mean, I know I mean know I know that dark social model is a little challenged when times get tough.

>> Oh. Oh, yeah. It's the first thing that gets cut. Yeah. Cuz the attribution

gets cut. Yeah. Cuz the attribution isn't there.

>> Yeah.

>> Yeah.

>> So, that's challenging. Yeah.

>> It's not ever hopefully ever good. I

mean, there's still a bottom funnel aspect we can always argue for.

>> Yeah, absolutely.

>> But h how are you do you even I mean we're so sprinting towards this bottom funnel idea.

>> How are you advising clients on the top?

Like is there still a top? Where are

they doing that?

>> Well, um it's getting them back to the overall organic channel performance mindset.

>> Okay.

>> You know, so um I just did this review, year-over-year review with a client where we busted our asses. like we

produced a lot of highquality content and traffic stayed flat the whole year.

>> So, back to that thing I said about top ofunnel decay >> and the gains that we're making on the BFU stuff, it's not even enough to create gains on the traffic front. So,

that question came up. How do we know this is working if traffic is going down?

>> Well, we have to start looking at um organic demo requests quarter over quarter, year-over-year. How's that

quarter, year-over-year. How's that trend? For this client, it was extremely

trend? For this client, it was extremely positive. It was like triple

positive. It was like triple >> for qualified pipeline. How's that trend looking >> from the organic channel? Um quarter

over quarter, year-over-year that was like double.

>> Yeah.

>> So, so far the revenue numbers are checking out. At this point, technically

checking out. At this point, technically the conversation could stop.

Technically, the conversation could be, look, revenue is up, so who cares?

That's the AHF's mindset. If revenue is going up, why do we need to track everything and care about everything so much in terms of attribution and this and that? But, you know, I don't do

and that? But, you know, I don't do that, right? I give them the leading

that, right? I give them the leading indicators because they want a story.

>> They want the narrative. They're like,

"Okay, what are we doing that's leading to these outcomes?"

>> So, then you start reverse engineering it into the inputs and the leading indicators. The inputs are simple. It's

indicators. The inputs are simple. It's

content created, net new, content refreshed, um links being built, um partnerships being established with other brands where you can get back links,

>> uh internal linking, right? um UX

improvements, template improvements, >> okay?

>> Uh ways to get more on page conversions by maybe putting a form at the top, any little testing you can do to squeeze more out of the traffic that you're getting, stuff like that. Um, and so a few leading indicators that I've leaned

on which have worked the positions in the top 10 classic if that is maintaining or even growing

or stable at least but traffic is going down that leads you to the AI overview topic.

>> Yeah.

>> Because you have a clear defensibility case here. You say, "Look, our top 10

case here. You say, "Look, our top 10 positions are maintained, yet we're getting less CTR."

Obviously, that's due to AIO.

>> Yeah. Yeah.

>> Yeah. So, that's defensible. Another

leading indicator that I've leaned on, which has been useful, especially for the Bofu argument is traffic value from hrefs.

>> Yeah.

>> Believe it or not, I have a few clients that are like on board with this as like OKR. M

OKR. M >> like I got to produce OKRs on AHF's traffic value. [laughter] How crazy is

traffic value. [laughter] How crazy is that? I got a client every two weeks

that? I got a client every two weeks that's like, "Yo, I need an update on what's the traffic value from HFS." I'm

like, "All right, here it is." Well,

we're doing a lot of bottom of funnel content, so it looks great.

>> Yeah. Okay, there you go.

>> Yeah. So, I mean, again, these things are hackable, right? So, if if traffic value is a OKR for a client, that can very easily be hacked. All you have to do is go for the highest CPC terms and

rank.

>> Yeah. I mean, it sounds easier. It's

easier said than done. Yeah,

>> you still got to rank and execute and create great content, but um that can be hacked to an extent.

>> Yeah, I agree. I mean, you can report on we're we're leading their conversions for sure. We had a history of like

for sure. We had a history of like putting traffic first in our reports.

>> Not intelligently, but uh >> yeah, >> people wanted that. Um and it's not our team's fault, but >> people Yeah, they they want that. So,

there's another thing you can do on the leading indicator side to put clients at ease or your boss at ease. It's only

tracking what I call the money pages or the money keywords.

>> Yeah.

>> Because these are far less volatile.

So, if you're rank tracking a bunch of stuff, again, back to the prompt tracking being irrelevant, you don't need to be rank tracking a ton of things. Just focus on the money keyword

things. Just focus on the money keyword stuff. Even you can create like these

stuff. Even you can create like these reaxes in search console that have like software solution vendor best, right?

This is going to isolate the money keywords automatically, >> right?

>> And so you can create these filters to show look traffic and impressions here are relatively stable.

So the traffic that matters is stabilized.

And if you're doing BFU content, it should even be growing a little bit.

>> Yeah. Um, and so that saves you on the argument of traffic going down because you can reframe traffic going down as tofu irrelevance.

>> Well, on that note, I mean, this is kind of one of my curiosities is you're describing SEO as purely and maybe this is just true, evolving into is [laughter] performance channel the

right word where it's kind of like dollar in dollar out. It's not

>> it. So Mike King and a lot of the pros say it's not a performance channel.

>> It is a brand discovery channel. Um

executives just don't like that.

>> Yeah.

>> Right. Even though that's kind of what it's becoming, they don't >> I mean he has more consumer brands.

That's good kind of like friendly.

>> He's got more consumer brands. Yeah.

Yeah. He definitely works with like bigger like D2C brands and stuff, but like for B2B SAS like they're kind of old school. Like they still view it as

old school. Like they still view it as performance. Like if you don't know how

performance. Like if you don't know how to piece together the storyline, >> you're cooked, >> right?

>> You just are because they're going to eat you alive on the traffic going down thing. You're going to be like uh uh uh

thing. You're going to be like uh uh uh uh and they're going to eat you alive.

Um >> is there a cost to this? Like uh in a world where only the bottom funnel exists?

>> Yeah.

>> Through search anyways.

>> How does that change? There's still we still would like to get in front of people on awareness where on transactional suddenly they'd be like >> I'm I'm going to that person because I'm

aware but how do we accomplish that or do we not worry ourselves with that activity and let the brand do that now?

>> Well, I have a I have an interesting thought on this. Uh maybe not everybody agrees, but like all right, getting back to the defensibility >> uh point. Um,

so one aspect that I didn't bring up was like the LLM visibility tracking. You

can show clients or your boss that >> you know LLM citations are another important indicator you need to be referencing and and showcasing and highlighting. So if the SEO content

highlighting. So if the SEO content you're producing I hate that term SEO content by the way.

>> Yeah, we've actually reframe internally organic content.

>> I like that.

>> Yeah. Feels it feels better. It's just a better framing. But let's just say um

better framing. But let's just say um the content that you are creating as a organic channel team.

>> Yeah.

>> Um [laughter] that's hilarious.

>> That's us.

>> Yeah. Let's go. Um you sort the highest amount of LLM citations in any of those tools, whether it's Scrunch. I'm I'm no plug in particular. I'm not associated with Scrunch AI, but I just happen to

use them because I like the founders.

Yeah.

>> Um, >> we like [clears throat] I mean we like Profound and we like Peak. We're

currently using Peak because it was the more affordable solution. Still very

great product quality.

>> Peak is great. I've seen the product.

Uh, I like the founder too. He's cool.

Um, both great company. All great

companies really. I have no dis about any of them. But um anyway, if you sort by highest amount of citations from your

site um and by page level, it's usually all the SEO content or the organic content. It's usually the

organic content. It's usually the comparative pages, the crash the parties, the X best lists. Of course,

there's going to be some new things in there like homepages.

>> You guys had a great um data point on that where homepages are getting cited.

Uh referral clicks to homepages from LLMs are going up. seeing that

consistently with every client. Um, but

more often than not, it's the XB best lists, the comparisons, um, the crash, the parties, the tearowns on, um, you

know, um, even competitor feature tearowns, pricing tearowns, stuff like that. Is X worth it? Stuff like that.

that. Is X worth it? Stuff like that.

Um, >> yeah, I've seen that. I It probably depends. I definitely on B2B, I see that

depends. I definitely on B2B, I see that a lot. I was just looking at and sharing

a lot. I was just looking at and sharing with the team today in B TOC. There was

like a a buying guide for a high ticket e-commerce product that was like X buying guide. There was product type

buying guide. There was product type versus product type, >> right?

>> So, not exactly brand versus brand, but basically >> style versus style.

>> Yeah. Am I defining this correctly?

>> Yes.

>> Showed up in their same list of like what is the best X e-commerce product?

>> Yes. But that's currently my way of thinking is like taking that transactional list you talked about.

>> Yes.

>> Aggregate who what URLs are getting sourced the most, see which ones are done by people like us that we can actually execute and that should help prioritize those types of topics like

the product versus product comparison, not a competitor comparison. If we hadn't

competitor comparison. If we hadn't prioritize that, but the client actually cares about it.

>> Yes. And it's in a space where like we have to be concerned that in e-commerce this it's 0.25%.

Ah but if CEO cares I don't show up for that.

>> Oh yeah Ross another thing was um if there's a topic that has a lot of fast changing ideas and it's constantly sort

of like evolving or there's like a new definition emerging.

Um, I find that LLMs because they don't have that consensus, they need the grounding factor. So, they need to go

grounding factor. So, they need to go and reference like a lot of websites all the time um to to come up with an answer that's the most current and relevant.

>> Okay. So something like that would be um for example like I work with a engineering management platform >> and one of the things that comes up all the time is can you actually track

developer productivity is there a way to do it so there's constantly like analysts coming out with new you know POVs on that like McKenzie says yeah you can kind of do it but

there's this and then the the SAS platforms come out and say yeah we have a new methodology for doing it. Um, even

another client with with the concept of, you know, GTM intelligence, market intelligence, GTM AI, you know, what is it? How does it work? How do you build

it? How does it work? How do you build AI into your go to market processes?

Because this is like an exploding topic.

>> Yeah.

>> You know, shout out to Brian De. Um,

>> anything that can be considered an exploding topic, even if it's not Bofu, >> it's going to get a lot of citations.

>> Yeah.

>> Because there's no consensus on it. Um,

in fact, I was talking about this with uh Bernard from Clear Scope.

>> Uh, we were talking about the um the necessity to inject information gain on top of consensus. Like you can't be

too information gainy without layering the consensus because it won't work.

>> Yeah. Okay. Interesting.

>> Yeah. Yeah. You won't you won't get surfaced because there's not enough consensus. So you need to have the

consensus. So you need to have the consensus answer but then you need to like have the uh information gain on top of that in a way in a way that's like

additive as like an extra layer. Okay,

>> but you don't diminish the consensus point of view too much.

>> So he is that relates to kind of this news breaking idea how or does does it?

>> Well, what he was saying was um he made a great point on this. The engagement is actually the thing that validates information gain.

>> So think think about this. If there if there was something that was very consensus like I don't know >> what is ABM.

>> Yeah. Then AI shows up.

>> Yeah. And an AI shows up and then your information game is your information gain is like so kind of out there that people don't agree with it >> and like engagement is like really bad

on that on that content. like the clicks end up getting back clicked a lot or pogo stick or time on page is real low >> that's going to falter >> yes

>> because there there's no engagement based validation on it >> and if you think of how social networks work like think about Chris Walker and his claim to fame how he just took

everything by storm that was all information gained by default he was spinning everything in a backwards way >> okay >> and because the engagement was so strong

every single time and there was no gaming of it, right? Like there was no engagement pods or you didn't see a thousand comments that said like great job or yes. Like he wasn't gaming it.

All that engagement was was natural.

>> Yeah.

>> So that information gain is therefore validated >> by the level of engagement that it's getting.

>> Yeah.

>> And so and so that also applies to SEO.

>> Yeah.

>> If he was doing SEO, he would have won.

like if he started putting out what is a go to market strategy with his new twist on it, he would have won.

>> It's an interesting observation because I think a lot of people are grounded in the idea that this is like some specific SEO like Google Northstar initiative.

>> Yeah.

>> But what I'm hearing from you is that's just adding value to content in a lot of ways. It's not some specific check box

ways. It's not some specific check box they put in the algo.

>> It has to make sense to the people reading it. They have to believe it.

reading it. They have to believe it.

They have to engage with it enough to where the signals are strong enough that okay, yes, this this deserves to be where it is.

>> It can't be something that's so wacky.

Now, maybe you can try and hack this to an extent because if you have a lot of other third parties that you reach out to to include your data points to shape

a narrative because now there's corroboration. Now you have your

corroboration. Now you have your information gain on top of a consensus page that's ranking and you get other third parties to also corroborate your

angle. Now you've built this sort of

angle. Now you've built this sort of corroborative ecosystem on your point of view.

>> Okay?

>> Because other ranking pages are talking about it or even low ranking or not ranking. Um that's how LLMs are going to

ranking. Um that's how LLMs are going to start to um surface your point of view more because of the what's called co-mention.

>> Interesting. And it's I it reminds me of a period I mean very smart Ryan Law >> did an article on sort of a very he's so good at this generally like he had an

article on don't be a content engineer or something you might have seen >> uh it's on HS you should check it out we actually did a pod with >> uh Josh Spilker from Aerops which was

good but anyways >> he had the counternarrative down pat >> very it was very good I think one of the catches of that if people took that too far, it's like

you end up with like this spammy clickbait feeling blog where you're kind of like like you have authentic takes.

>> Okay. Yeah.

>> Of all these this is your natural persona.

>> Yes.

>> If someone tries to process that concept, >> okay, >> it can be difficult, right?

>> You can't be artificially contrarian and win. Yeah, that's yeah,

win. Yeah, that's yeah, >> it just doesn't really work. Like

something comes to mind where um when I was uh leading this um SEO project at an identity theft company, Aura.

>> Yeah.

>> We had a lot of contrarian takes that were honest and they did well.

>> So, here's an example, right? Um, is

insurance for identity theft protection worth it or do you need identity theft insurance?

All the top ranking pages like the Lifell locks, um, the Nortons, the McAfei, all those legacy brands were like making a strong case for it. Oh

yes, you need it if you don't get it.

And then fearmonger of what happens if you don't get it? You're screwed. You're

not covered. You can lose your life.

like etc etc. We were the only page at that time at least that said well you need it only if you can't do all these

things. If you can't check your credit

things. If you can't check your credit score manually and make sure there's no, you know, funny business going on. If

you don't have the time to manually log into all your credit cards and check for fraudulent activity.

>> If you don't have the time to, you know, etc. to go through all those like financial and hygienic things you need to do to make sure fraud isn't happening in your life.

>> Um, if you don't have the time or patience or know how to manually do all that stuff, then you need it.

>> Yeah.

>> But if you're comfortable doing all that manual work, then great, you probably don't need it.

>> And that works. We ranked. Now, maybe

there's other factors at play, you know, contributing to why we ranked such as, you know, momentum, authority, links, but I'd like to I'd like to think that the counternarrative there because it

was honest happened to work.

>> Yeah. Yeah, that's true. And that's a good framing generally is like it's counter narratives versus information gain. Like technically counter narrative

gain. Like technically counter narrative is information gain but it's like if you have an article on GTM something and you add AI to it just because that's so relevant now that's

not a counternarrative.

>> That's not >> that's just information gain.

>> Yeah.

>> And that may be a more palatable brand friendly thing to do more often where you have that ability to be like it's this plus this rather than it's the

complete opposite of this thing.

>> Yes. Yes. I'm with that. I'm with I I think I think it's a great topic. Um

Yeah. I think you just want to do it in a manner that's authentic and you don't want to be stirring the pot or doing anything that's like artificial.

>> Yeah.

>> Um just to spark engagement cuz I think you're going to do more harm than good.

>> Yeah. Makes sense.

>> So I mean we hit a lot of stuff. I feel

like I still don't have an answer in my head and maybe you don't either. We

don't have a full answer. This is like what happens to the top funnel?

>> Oh yeah. What happens to the good old top of funnel? Well,

>> yeah, >> I don't know if anyone will agree with this but

>> I feel that top offunnel for SEO is basically going to be dead. It's going

to be it's going to be essentially zero.

Um, let's make the the bare case and the bull case for it or the pessimist and the optimist case for it. The pessimist

bare case for it is you're never going to get any clicks. You're never going to recommend any brands in tofu queries.

Um AI overviews are just going to answer that, you know, right in Gemini when whenever Google becomes AI mode, maybe 10 blue links will be gone altogether

and there will be very very low if any clicks uh going to tofu. And that we already know that um chatbt and perplexity they just don't really site any links when it's clear tofu consensus.

>> Yeah.

>> So therefore you must generate tofu by doing other things such as events, social media marketing, YouTube,

um airport ads, car wrap advertising, advertise on podcasts, partnerships, whatever it is to generate that buzz, brand word of mouth thing. You got to go and do that.

>> Yeah. or you do what the siege medias of the world do, the AHFs of the world do, the SER interactives, so on. You produce

data studies where you come up with like new unique information that no one else has. You do stuff that like the drifts

has. You do stuff that like the drifts from back in the day did where they wrote a book on conversational marketing.

>> Yeah.

>> You know, you you do these things that um nobody else can replicate or do that essentially becomes the new tofu. Like

you can't out tofu stuff that has been tofued to death. Like answering

definitional queries that have long been answered by every SAS company out there.

What is enterprise SEO? I remember

searching enterprise SEO and seeing 10 blue links with the exact same meta title.

>> What is enterprise SEO? What is

enterprise? [laughter]

>> All the way down the line saw the same page just regurg.

>> It's actually small business SEO. That's

the counter.

>> That's the counter narative. Yeah. Yeah.

[laughter] Gotcha. Um, so that's my take on tofu that u, so that's the bear that's the bear pessimism take.

>> The optimism bull take is you do it for topical authority.

>> You do it for, you know, semantic embedding to sound all fancy and cool.

Um, you do it for internal linking purpose. You do it for, you know, you

purpose. You do it for, you know, you need those pillar hubs for the content clustering model. um you do it to have

clustering model. um you do it to have linkable assets.

You do it uh why else would you do it?

Because you have the um the tile ranking. You may still have the tile

ranking. You may still have the tile ranking um in the AI overview, the little tile carousel. It's unclear to me how many clicks, if any, those get. I

don't know if there's any studies out there that can figure that out. But um

you can make the argument that having your brand there at least >> it's pretty low unfortunately. I think

we looked at it once.

>> Yeah. But you can at least make the bear I mean the bull argument that yeah having the brand there as an impression thing on the tile is like valuable.

Maybe you'll get a click. Um I can't think of that many other optimism cases for it though. Can can you?

>> Yeah. I wouldn't say I mean it's not a ripping channel.

Uh, I think there's value. I mean, it's not really top funnel when I when I hear you describe it and actually I think about it again, but if it's findable, it's not it's not so deep in your site

architecture that users can find it easily like it's on the blog hub. Maybe

it's in the top nav. It's like core assets like trends reports, data studies, those those are great.

>> Those are valuable.

>> Yeah.

>> Yeah.

>> Uh, is there are they Yeah. Can you

repurpose it in some way like >> Yeah.

>> What? But I I generally agree the definitional topfunnel not a lot of value.

>> There's not a lot of value. Yeah. I

think Will Reynolds um man he had a great great take on this. It was

something to the effect of if you have URLs that you compare how much social traffic they get

versus direct traffic versus say SEO or organic traffic. And like there's no

organic traffic. And like there's no social traffic at all, but like it's getting and there's no direct traffic at all. That means it's not getting shared

all. That means it's not getting shared privately in WhatsApp groups and stuff and people are not buzzing about it.

It's just made for search.

>> Unfortunately, that's that's bad.

>> And that's what a lot of tofu is, especially in B2B. Like I I actually um took a jab at all the cyber attack companies cuz the first cyber tech company to do this uh glossery concept

of like every single possible cyber attack out there was PaloAlto Networks um back in the day and then like every company copied it and I mean every you

know Fortnet, Zcaler, uh Rapid 7, Cyberhaven, like they all have done it right. They they just all do it. It's just for some reason for

do it. It's just for some reason for cybertech, B2B, SAS, the go-to SEO playbook, Crowdstrike, they all have one. It's a glossery of terms with all

one. It's a glossery of terms with all the same terms and all the same pages.

And the only way that they outrank each other is who has more brand, who has more links.

>> Yeah, >> there is zero information gained. And

now I would guarantee it's like all AI slop. Majority AI slop. There's no

slop. Majority AI slop. There's no

analyst putting thought into any of those pages. Come on.

those pages. Come on.

>> Yeah. [laughter] I mean the only like bullcase for some what is is like there's simple what is and then there's complex thing like there's really a deep

question in that first layer like >> maybe some cyber security topics or >> development related things where you're not going to get the answer in one sentence >> right

>> it's maybe it's multimodal you need you need graphs and visualizations to represent the concept >> that >> that's at least holding up a little bit better and I think more argument. That's

the only >> Well, I I saw Ryan Law himself and I I love him and he came out with a great

post. It was a while back, but he said

post. It was a while back, but he said um the value of educational business blogging is zero.

>> Yeah. Like it was a while I got to find the post. We'll put in the show notes.

the post. We'll put in the show notes.

Um it was a great >> good counter narrative.

>> It's a great counter narrative. Yeah.

And uh uh um Peter Caputo, the guy who is uh running datab box, he came out with a similar post saying business blogging value is zero.

It's zero.

And I actually kind of agree to some extent, but only if you isolate to the tofu stuff. Because if you're going to

tofu stuff. Because if you're going to do tofu, >> that's sort of what they're saying.

Yeah, >> that's sort of what they're saying. Like

here's my view of what the new business blogging is. It's LinkedIn post. It's

blogging is. It's LinkedIn post. It's

your personal brand. It's YouTube. It's

Substacks. It's beehives.

And why is that? Because

most of what you read, especially on business, corporate, SAS blocks.

There's no real authorship to any of it.

>> Yeah, >> let's be real. It's all ghost written.

All of it. Full stop. Maybe 1% of it, you know.

Somebody from the company will review.

Let me let me say this almost none of it is fully written by somebody from the company.

>> Yeah.

>> SEO programs that are producing bofu. I

mean I'm doing this. We are using external writers but getting it validated by internal PMM. This is

getting reviewed. It's a collaborative process but then it's getting authored by somebody at the company. But I would argue that for so many of them, not even

that's happening. It's just an external

that's happening. It's just an external writer putting together content, you know, slapping somebody from the company's authorship on it, and they're calling that EAT. No way.

>> No, no way. No way. Right. And so why is Substack so much more valuable um from that standpoint? Because you are forced

that standpoint? Because you are forced to put your personal name on something and you are promoting that as yourself.

So, you better have written that or at least have done like most of the work that you can stand by. Um, because if you're if you're launching a Substack and you're outsourcing that, that's

crazy to me.

>> That's I mean, I I totally get that from a consultative approach, but >> yeah.

>> Is are you are you have clients doing that? Like mid-market SAS doing that?

that? Like mid-market SAS doing that?

>> Doing substacks? Yeah.

>> No way.

>> Oh, okay. Oh, I thought that's where you're >> Oh, no. No, no, no way. No way. I'm just

making I'm just making the point that um >> I mean some CEOs are doing it right.

Like some CEOs and startup founders are doing the whole founder marketing thing.

Yeah. They're running Substacks.

>> Yeah. And they're getting a lot of engagement building a following that way. I don't know of any corporate SAS

way. I don't know of any corporate SAS um brands doing a Substack. That would

be really interesting.

>> Um >> Yeah. I mean I would just

>> Yeah. I mean I would just >> that' probably fail though. It would

fail because when that person leaves the company the Substack implodes.

>> There's nobody to take it over. It's not

dead.

>> It's Yeah, it's the same reason why there's no loyalty with corporate SAS newsletters.

>> Like when I see yours come, Ross from Siege Media, I'm reading it.

>> Appreciate that.

>> You know, when I see I mean maybe to some extent, you know, there's some interchangeability there, but like even Hrefs, the same guy has been sending it.

Squan. Um

>> that is our team drafting that. I'll be

honest with you.

>> But I do approve what's there. I approve

the list. I suggest changes. But still,

I I know that you're real.

>> Yeah. Yeah.

>> When it's like >> Sally from Zenesk.

>> Yeah.

>> Okay.

>> Yeah.

>> And then it's just it just has this soulless dry approach to it. Like I

don't know. Corporate SAS newsletters are just like I don't know anybody who's excited about them.

>> Yeah. Yeah. Agreed. Yeah. It's tough

times for for tofu. Uh but yeah, I mean there's still I think that organic like is there value outside of it? But like

why not just transfer all tofu to YouTube and social?

>> Some percentage of customers won't ever watch a YouTube maybe.

>> Okay.

>> I've heard that before. Have you not?

Like some people will not watch videos.

They just not their >> I've never seen my wife is no longer in the she used to work for >> Okay.

>> Never seen her watch a video on YouTube.

>> Fair. I mean it's a [laughter] it's an interesting discussion, man. Everybody's

got different ways they want to consume content, you know.

>> Um some are podcast only. Like it it's hard to say what is going to be the new Tofu. I guess the whole point is like

Tofu. I guess the whole point is like what happens when AI mode becomes the new default cuz that's where all the evidence is sort of like hinting at at the moment.

>> Yeah.

>> I've got to believe it's coming. And so

when that comes that's going to be the final nail in the coffin because the little bit of tofu traffic that still exists today is primarily through the 10 blue link system.

But they got to retain the incentive system of like the open who's going to create that same content for that new topfunnel or is it simply because we can

have it temporarily. So my theory on this is because the training data has already been extracted for so long and that it doesn't really change on the majority of these consensus

informationational queries. Right back

informationational queries. Right back to the information gain. There's not a lot of room for information gain. And I

don't know how often these training data sets get refreshed, so to speak. But

there's no need. They have the answers.

It's never going to change. What is ABM is never going to change.

What is SEO that changed because AI?

>> Well, you say that. I mean, I think of two. I mean, there's

two. I mean, there's >> Yeah.

>> Postcoid, everyone had to modify so much consumer content. Then post AI

consumer content. Then post AI >> Yeah. if your articles didn't say I I

>> Yeah. if your articles didn't say I I mean AI would also understand that and likely at least tweak it even if it doesn't need access. I just think there's got to be some degragation over

time.

>> Let me run this by you Russ. [laughter]

So I have one client that's very AI forward thinking and very open-minded.

So we can do a lot of experimentation.

This is cool. So um we did what is the X best insert SAS for 2026

masterpiece page it ranks highly it does the job for the core head best software term >> okay >> it it's a list page it ranks it does very well >> then we had this theory of what about

like longtail variants of that like what is the X best software that as reporting capabilities, advanced reporting capabilities.

In classic SEO, we would never have built another page >> to to dial into the reporting aspect of the SAS. What we would have just done is

the SAS. What we would have just done is have that get soaked up by the the main head term page.

>> Right. Okay. Yeah.

>> Yeah. because we would be talking about reporting capabilities in the main mega page and because the volume would be non-existent essentially for the reporting or negligible.

>> Yeah, >> it wouldn't justify the effort of creating a new page and in fact it would likely even cannibalize.

>> Yes, that was where my brain went.

>> Yeah, it would likely even cannibalize, right? So, hear me out. We said, you

right? So, hear me out. We said, you know what, let's just roll with it.

Let's just let's just see what happens.

Let's go and build a new page about the XEST software that have advanced reporting capabilities.

>> Mhm.

>> And to be honest with you, it's just a regurgitation of the original page. It's

not a carbon copy, but it's a regurgitation, >> but but it's more reporting focused.

>> It's very much reporting focused. Um,

and when we search in Google that query for the longtail intent on the reporting, something incredible happened.

>> And this has stuck now for like a few weeks. So, I think it's going to stick

weeks. So, I think it's going to stick and this is the way it's going to be. In

the AI overview with the tiles, the answer from our page is extracted and at the top and our longtail page is featured as the tile, the number one

ranking tile.

When you go into the blue links, the other page is there.

>> I've seen double stacking double double stacking. Yeah.

stacking. Yeah.

>> So, there's like a double occupancy SEO game potentially for some of these long like like the what I'm saying is that the cannibalization effect of classic SEO

may be less of a threat now because of the AI factor. like it may like to me >> well because they want more infuration information gain by default.

>> Yeah. And also what what is the goal of AI to give you the most relevant answer?

>> Okay. Yeah.

>> Like this wouldn't have worked in classic search, >> right? Yeah.

>> right? Yeah.

>> Like the like the the longtail page would have been buried or not indexed because it would have been >> considered to be too similar or something like that. It would have been hard to rank that one or index it even

>> where now like AI overviews are the precursor to AI mode.

>> Yeah.

>> In an AI mode only world, guess what happens to those 10 blue links? They're

gone. Guess what happens to that chunky head page if you're searching for the reporting aspect? That that page is

reporting aspect? That that page is gone. It's not there, right?

gone. It's not there, right?

>> The AI overview is going to default to the most relevant longtail. Bang. But

for now, because the 10 blue links still exist, you've got the opportunity to double stack on the AI overview. Maybe someone

will click that, maybe they won't. Um,

the 10 blue links, maybe they'll click the chunky page, the the headterm page that has, you know, a lot more information in bulk. It's less focused

on reporting. But you can to an extent

on reporting. But you can to an extent get away with put that in air quotes because you know who know who knows what will happen.

>> But that's an observation I' I've been seeing is you can potentially double stack with longtail and AI overview and if you have a headterm page that gets the classic ranking that could still be there.

>> Yeah. Yeah, another just side effect thought you made me have is I wonder with chat GBT and LLMs and AI mode and Gemini if the natural length of the

average searchers query even on Google is elongating. So

are they doing longer searches on average on Google because of these other tools? I would ex could see that it

tools? I would ex could see that it probably is very sub subtle over time.

Um I think about my own behavior and yes um it is getting longer.

>> Yeah.

>> Um >> I would I I I still think in a AI mode world it's going to be much longer because of the automatic follow-ups.

Like the automatic follow-up prompts seem pretty long.

>> Oh yeah. Yeah.

>> You know, would you like me to look into the best reporting tools that also have pricing that you want under this?

Whatever, right? Like

>> Yes. Oh, and then that you can auto kind of select. Yeah. So those are longtail,

of select. Yeah. So those are longtail, but I have found myself even searching Google in a way that's like more natural language to be honest.

>> Yeah. So it's been great. We could go two hours. Uh props whoever's made it

two hours. Uh props whoever's made it this far in the conversation.

>> Yeah.

>> Yeah. Uh we appreciate you.

>> This is a banger, Ross. [laughter]

>> Let's go.

>> Exactly. If there was like one thing you'd present to these executives to kind of wrap it all up that would change the conversation or you suggest people do around GEO SEO like what what is that

thing you you would share? Yeah, I think it it gets back to the illumination of the dark SEO funnel. I know that sounds like whoa, but um the first step in that

is getting executives to be weaned off of the traffic.

That is a big step that needs to happen.

um that happens through education, patience, and then piecing the story line together strategically with the data that you have.

>> So you want to start with overall organic channel performance. Again, you

know, pipeline qualified demo requests that should be moving in the right direction.

>> Um you want to rely on some self-reported attribution if possible. I

think there's a lot of gold in there that's not being extracted.

>> Um and that can help, you know, close some of these gaps. Um, and then lean on the leading indicators, LLM citations from the organic content pages that you're creating. How are you maintaining

you're creating. How are you maintaining in the top 10 ranking positions? Because

if those are stable yet clicks are going down, you have a defensible argument.

>> Mhm.

>> And um I mean I think that's really a good one. And then traffic value. If you

good one. And then traffic value. If you

want to look at some other stats that can be interesting, you know, show that hey, we're focused on more commercial intent. Therefore, the traffic that is

intent. Therefore, the traffic that is coming in has higher CPC value. Um, that

that can be defensibility as well. And

then the input metrics too matter. Like

you want to show that like things are happening.

>> Yeah.

>> If you're not updating content, keeping content fresh on like a quarterly or at least four or five month, you know, revolving basis. Yeah, it's going to be

revolving basis. Yeah, it's going to be tough.

>> Um, you know, you want to be doing query fan out approach to your content strategy. I think like targeting big

strategy. I think like targeting big chunky head terms is kind of like a old approach that's not going to work well.

Um, yeah, these are just some things that come to mind. I don't know if you have anything to add.

>> Yeah, no, I agree. I mean, we spent a lot of energy on reporting this year.

Literally was in all hands talking about reporting structure today.

>> We'll continue to preach that to to the choir.

>> I like it. You know, last one for you, Ross, before we sign out. Um, branded

traffic.

>> Yeah.

>> Homepage traffic. Um, increases to the pricing page, right? Are these things in your mind that help create a more

defensible SEO storyline in the AI era?

>> Uh, on page six of the report, I think if you're linking that to the LM visibility growth, as you noted, I think you're absolutely correct that it may

misrepresent it if it's PR, if they're doing other things as well. But if

there's a very similar cadence to that graph of LM visibility compared to >> and also rankings I guess as we speak

through organic bottom funnel rankings >> if if those are increasing 1 to three 1 to 10 >> I would expect that I think it's a reasonable argument to make.

>> I think that's a great point. Um you

know you reminded me of something I did with a client. We just did a metatitle change on a homepage to rank for a non-branded term.

>> Shot up to number one, skyrocketed clicks.

>> That's value. Yeah,

>> that's SEO value. It's an easy win, but nevertheless a win. Um, and maybe last thing I'll toss at you. How about uh LLM referral traffic going to the homepage

or going to the pricing page?

>> I think that's value.

>> Yeah. Yeah, for sure. I wouldn't count that. I don't think there's like truly

that. I don't think there's like truly navigational search on Loom.

>> Yeah. I mean there's navigational. Yeah,

we talked about brand searches that have refinements that I guess could happen that way. Um, so which we we don't have

that way. Um, so which we we don't have clarity to, but I I expect that to be >> I mean I have seen in some cases like LLM prompts for X best queries where the

homepage has such a strong association with that that it just goes there because of the brand.

>> The citation for the homepage comes up ahead of like a product page even, >> right?

>> So that that could be massive value. You

just got to be able, [clears throat] you know, highlight that and tell the story appropriately.

>> Yeah. Well, I think a great way to sign off is to say this is TopFunnel right here. [laughter]

here. [laughter] This not every brand should do this, but >> Yeah. I mean, there's no sales intent

>> Yeah. I mean, there's no sales intent here. We can't pitch you anything. No

here. We can't pitch you anything. No

BFU.

>> Yeah. We have nothing to sell here.

>> We got nothing. All right.

>> There you go, Tano. That was amazing.

>> Thank you, Rob. Appreciate it. Peace.

[music] >> [music] [music]