Episode 118 | The AI Hat Podcast
Hosted by Mike Allton
Scrunch's Kevin White reveals why 260% visibility gains start with measuring what you can't see
Your marketing team thinks they're winning because you rank on Page 1 of Google. Your SEO dashboard is green. Your content calendar is full. Your blog traffic might even be up.
And your prospects have no idea you exist.
Because they're not Googling anymore. They're asking ChatGPT which CRM to buy. They're asking Perplexity which marketing automation platform solves their specific problem. They're asking Gemini to compare your category - and AI is recommending your competitors, not you.
This isn't a "future of search" problem. This is a pipeline problem. Right now. If your brand doesn't exist in the AI answers your buyers are getting, you're invisible before the sales conversation even starts.
The Citation Economy No One's Measuring
Kevin White saw this coming early. As Head of Marketing at Scrunch - and previously leading growth at Twilio Segment, Retool, and Common Room - he noticed something strange in his analytics around January 2024. Traffic from ChatGPT and Perplexity was growing exponentially. Quarter over quarter. And that traffic was converting at rates that made his traditional SEO channels look anemic.
But he had no way to measure it. No way to understand what was working. No way to close the gaps.
"I saw that traffic and I was like, okay, there must be something here," Kevin explains. "But I didn't really have any way to understand what was going on, measure it, uncover gaps, build it into a strategy other than saying, 'Hey, we're getting traffic from these sources.'"
That curiosity led him to Scrunch - and eventually to join as their Head of Marketing. Now he's helping brands achieve 100% to 260% visibility gains in AI search in just 60 days. Their clients are tracking customers who convert specifically because they appeared in ChatGPT answers.
The difference? They measure what everyone else ignores.
Why Reddit Won't Save Your Pipeline
Everyone's chasing the same "expert" advice right now. Get mentioned on Reddit. Optimize for Wikipedia. Show up on YouTube. The logic sounds airtight - studies show these platforms dominate LLM citations, so if you appear there, you'll appear in AI answers.
Kevin's data tells a different story.
"Reddit's outsized influence is a composition effect," he explains. "It's a strong cross-industry generalist - similar to Wikipedia, YouTube, and Quora. But most marketers work in one industry, not all of them.
When you slice the data by specific industry categories, Reddit's dominance shrinks dramatically. Vertical specialists consistently punch above their weight. NerdWallet dominates financial services citations. Tripadvisor owns travel. Industry-specific blogs and review sites you've never heard of are getting cited for bottom-of-funnel prompts in your category.
And you're wasting time trying to game Reddit.
"The core thing to care about is just getting visibility into what those citations are," Kevin says. "You can't just blanket say, 'I'm gonna go after whatever Reddit subreddit' and expect to show up in LLM answers. You need to actually know the citations of authority as a first step."
The Three Metrics That Actually Matter
Kevin breaks AI search performance into three buckets - and none of them show up in your Google Analytics:
1. Brand Presence (Share of Voice)
When someone searches a prompt relevant to your solution category, does your brand appear in the response? This is the top-of-funnel metric. If you're invisible here, nothing else matters.
2. Citation Rate
Are LLMs referencing your domain as a source of authority? In the Google world, this was the "blue link." In the AI world, it's whether ChatGPT or Perplexity cites your content when answering questions about your category.
3. Competitive Position
Within the prompts you care about, where do you rank versus competitors? Are you always at the top of share of voice? At the bottom? Trending up or down?
"That's what the board and people want to see," Kevin notes. "Within the prompts that we care about, are we showing up? Are we showing up better and more often than our competitors?"
Referral traffic - the metric most teams obsess over - is the very last step in the journey. If you're only measuring clicks, you're missing the entire ballgame.
The Long, Long Tail of AI Search
Traditional SEO taught us to optimize for the "long tail" - those 5-7 word searches that signaled high intent. Kevin calls AI search the "long, long tail."
"In Google, the average is like five to seven words," he explains. "In an LLM - especially when people are talking into it - you're gonna get a paragraph. Twenty-plus words oftentimes."
This changes everything.
In the zero-sum game of Google's 10 blue links, you fought for generic, top-of-funnel terms because specificity meant obscurity. In AI search, specificity means discoverability. LLMs are exceptionally good at finding bespoke content that matches precise intent - and returning it even when the source isn't Wikipedia or Reddit.
"You can almost create content that is just talking about your specific differentiators in a way that maybe only you are an authoritative source over," Kevin says. "The LLMs are actually pretty good at finding that content and returning it back to the user when there's that intent on the other side."
Field of dreams marketing actually works now - but only if you architect the intent first.
The Workflow: From Invisible to Cited in 60 Days
Kevin's clients aren't getting 260% visibility gains by accident. They're following a systematic three-step process:
Step 1: Measure and Benchmark
Start with the prompts your prospects are actually asking. Use Google search keywords as a starting point and translate them into conversational queries. Load them into an answer engine optimization platform like Scrunch.
This gives you the baseline: How much is your brand being mentioned? Where are you being cited? What third-party sources dominate your category?
"That baseline measurement gives you a starting point and kind of stack ranks you versus competitors," Kevin explains.
Step 2: Identify Gaps and Opportunities
Analyze the citation landscape. Break it into three buckets:
- Owned citations: Your own domain and content
- Competitor citations: Competitors being cited as authorities
- Third-party citations: Publications, review sites, newsletters, Reddit, etc.
The third-party bucket is usually the largest - and the most actionable. These are the domains LLMs trust in your vertical. Prioritize them by citation frequency across your tracked prompts.
"That gives you this tops-down list to say, 'Okay, these are the sites that we either need to partner with and get mentions in, or maybe the content there is really light and that's informative for you to generate your own content,'" Kevin says.
Step 3: Optimize and Scale
Two parallel paths:
Content optimization: Identify greenfield opportunities where no one's adequately addressing prompts you care about. Create net-new content or refresh existing pages. Technical optimizations that matter: FAQs, authorship, schema markup, content "chunkability."
Citation acquisition: Partner with authoritative third-party sources to get your brand mentioned. Kevin's team works with companies like Noble, which brokers these exchanges through a publisher network.
Results show up fast - often within one to two weeks, compared to the months-long lag of traditional SEO.
The Technical Layer: AI-Optimized Without Breaking the Human Experience
Here's the problem most websites face: they're built for human browsing, not AI retrieval.
"We build websites for humans with media, with JavaScripts, with all these different superfluous code elements that make the UX usable, friendly, immersive," Kevin explains. "That is not the way that AI likes to consume content. AI likes markdown and JSON."
Scrunch's solution? An Agent Experience Platform that sits at the CDN layer. When an AI bot visits your site, it gets a stripped-down, AI-optimized version - schema-rich, token-light, information-dense. When a human visits, they get the full immersive experience.
The impact: "We tend to see a lot more crawling of pages because it takes the token count down by about two orders of magnitude. Tokens are the economy for AI bots crawling your pages."
Think of it like Google's AMP for mobile - except instead of serving fast pages to mobile users, you're serving parseable content to LLM crawlers.
The Audit Every Revenue Leader Should Run Monday Morning
If you're wondering whether you have an AI visibility crisis, Kevin offers a meta-prompt approach:
"Tell the AI what you want and ask it to come up with a prompt for you," he suggests. "Give it your objective and say, 'Hey, I want a brutally honest, authentic way for you to go out there and see how my brand is being talked about or assessed in the market. Write me a mega prompt.'"
Include context: who you are, who you represent, what your business does, who you're trying to reach. Turn on extended thinking mode in Claude or ChatGPT. Let the AI architect the audit for you.
Then run it. And prepare to discover you're invisible in the channels that matter.
From Shadow AI to Sanctioned Workflows
Beyond Scrunch's core product, Kevin's deep into agentic workflows - using Claude Code to build public apps as marketing offers, automate spreadsheet enrichment, and analyze customer calls to inform product positioning.
We're not just writing with AI - we're automating entire marketing functions," he notes.
This is the agentic shift our revenue teams need to internalize. AI isn't a better spell-checker. It's an autonomous teammate that does work while you sleep.
But like everything in this space, measurement comes first. Kevin's team uses Scrunch to dogfood their own strategy - tracking competitive intelligence, diagnosing content gaps, and systematically closing citation opportunities.
"We've grown our brand presence in the last four weeks around the core prompts we're tracking by probably 30% to 40%," Kevin shares. "There's still a lot of work to be done, but we're walking the walk."
The Three Things to Fix This Quarter
If you can only prioritize three actions:
1. Get executive buy-in: This is a new channel and a board-level priority. Revenue leaders need to expose the gap - "I searched for this, we're not showing up, and this is where our buyers are asking questions.
2. Make it a go-to-market problem, not just a marketing problem: Sales, customer success, partnerships - everyone can help surface brand mentions, broker relationships, and amplify content. This isn't marketing's burden alone.
3. Run the diagnostic: You can't optimize what you don't measure. Start with a seven-day free trial of a tool like Scrunch. Run the brand audit. See where you're invisible. Then close the gaps systematically.
The Bottom Line
Your prospects stopped Googling. Your SEO wins are Pyrrhic victories. And every day you wait to measure your AI visibility is a day your competitors are closing the gap you don't even know exists.
Kevin White isn't a futurist. He's a practitioner who built the measurement infrastructure to prove what works. His clients are seeing 260% visibility gains because they stopped guessing and started architecting.
The question isn't whether AI search will matter to your pipeline. The question is whether you'll measure it before your board asks why you're not showing up.
Run the audit. Close the gaps. Win the prompts that matter.
Resources
- Start a free trial at Scrunch.com
- Connect with Kevin White on LinkedIn
- Download Mike's Executive Guide to Shadow AI at theaihat.com/shadow-ai
Ready to Build Your Team's AI System?
Mike delivers keynotes and hands-on workshops that turn abstract AI concepts into working solutions your team can implement Monday morning. No hype. No theory. Just the frameworks that reclaim your revenue team's time.
Learn more: theaihat.com/speaking
Subscribe to AI for Revenue Leaders: The AI Hat Podcast
Stop being a Pilot Purgatory statistic. Every week, we deconstruct how Revenue Leaders are building governance frameworks, digital crews, and operational systems that eliminate admin drag and reclaim their team’s capacity for high-value work.
Full Episode Transcript
(Lightly Edited)
Your Prospects Aren't Googling Anymore - Scrunch's Kevin White on the AI Visibility Crisis
[00:00:00] Kevin White: You hear things like add FAQs, like add authorship, keep things refreshed, make content chunkable. All these things are, I would say like LLM lore right now, like the LLMs themselves are telling you like this is actually what makes a difference. But those are the things that tend to get more pickup and you get pick up pretty fast in LLMs and, and I would say.
[00:00:21] Yes, like listen to what people are doing, but then try for yourself. Make sure that like test and measure for yourself. Otherwise, they're like flying blind and, and you shouldn't listen to experts like me or whoever else out there. You should actually look at the results and make sure that like they're in a controlled kind of experiment.
[00:00:36] Mike Allton: Mike's got The Hat, the AI kind. Unlocking power in business, you'll find. Cut through the hype, explore the tech. Every week, new ideas to check.
[00:00:59] Allton leads us on a digital quest. AI secrets, he'll always invest. Tune in tight, don't miss a beat. Knowledge, groove, keep you on your feet. Listen up! To the AI flow. Business growth, here we go. Cuttin' edge insights, that's the plan. With The AI Hat, you'll understand.
[00:01:45] Greetings program. Welcome to The AI Hat Podcast, the resource for revenue teams who are ready to stop being the bottleneck in their own businesses. Each week we architect some really simple AI operational systems that eliminate admin drag. And reclaim your time, freeing you and your team to focus on the high value work that's gonna truly scale your revenue.
[00:02:05] I'm your host, Mike Allton. Subscribe and let's build your digital team together. Here's the problem nobody wants to talk about. Your marketing team thinks they're winning 'cause you rank on page one of Google for something. Your SEO dashboard's green. Your content calendar's full. Your blog traffic might actually be going up, but your prospects have no idea you exist.
[00:02:25] 'cause they're not Googling anymore. They're asking Chat GPT which CRM to buy. They're asking Perplexity, which marketing automation platform solves their specific problem. They're asking Gemini to compare your category and AI's recommending your competitors, not you. This isn't a future of search problem.
[00:02:43] This is a pipeline problem right now. If your brand doesn't exist in the AI answers your buyers are getting, you are invisible before the sales conversation even starts. Now, Kevin White saw this coming. As head of marketing and Scrunch and previously leading growth at Twilio Segment, Retool, and Common Room, he's built marketing operations in the age of AI search, and he didn't just theorize about it. He built the measurement infrastructure to prove what actually works. Scrunch helps brands track their visibility across ChatGPT, Perplexity, Gemini, every other AI platform your prospects are using. They diagnose citation gaps, optimize content for AI consumption, and deliver measurable gains in brand visibility.
[00:03:26] Their clients are seeing a hundred percent to 260% increases in just 60 days, but more importantly, Kevin's a practitioner. He's using AI agents to build public apps as marketing offers, automate spreadsheet enrichment, and analyze customer calls to inform product positioning. He's not talking about the future he's operating in it. Hey, Kevin, welcome to the show.
[00:03:48] Kevin White: Hey, Mike, awesome intro. Thanks for having me. Really appreciate you including including me and happy to get into things and, and help out the audience here from what you teed up. So let's do it.
[00:03:57] Mike Allton: Absolutely. As a longtime content marketer and blogger myself. This is something that's near and dear to my heart because honestly, and I've been in AI now for a couple years, I still don't really have a good grasp for what I should be doing and certainly what my clients should be doing. So I'm looking forward to, frankly, learning from you and picking your brain.
[00:04:14] But I wanted you to start by just walking us through the moment that you realized traditional SEO metrics we're probably lying to marketing leaders. What was like the wake up call that made you realize prospects weren't Googling anymore?
[00:04:26] Kevin White: Yeah, I, well, yeah, it's interesting. It's what brought me to Scrunch in the first place and, and got curious when I was at my previous company that company's called Common Room we were noticing that we were getting quite a lot of maybe not a lot, but at first, but we were getting a decent amount of traffic from LLMs from Chat GBT Perplexity. Those were the two that were like top that we, we saw traffic from. And we noticed that that traffic was converting at a pretty high rate requesting a demo starting at trial. And we noticed that that traffic was growing pretty exponentially, quarter over quarter.
[00:04:57] And this, to date things was of started noticing like around January, 2025 last year. So, I saw that and I was like, okay, there must be something here. But, didn't really have any kind of way to understand what was going on, measure it, uncover gaps, like, build it into a strategy other than saying like, Hey, we're getting traffic from these sources, but like, what are we doing to influence our brand showing up on these things aside from, we're already doing with LinkedIn and branded content and stuff like that.
[00:05:23] And so, that kind of got me really interested in uncovering this space or like cracking, like how to, how to use this as a new growth channel. I noticed, I, as a user of AI myself and like lots of people in our space I'm sure are, if you're listening to this podcast you know that like, it's it's gonna be a massive channel and lots of people are gonna adopt it. This is already like a billion users, I think, on Chat GBT. And so, this fast growing new channel for any marketer is a new dream. It's like, this is greenfield opportunity. How do I, learn this practice? How do I like jump in and make this a new growth channel because there's usually lots of low hanging fruit with any new channel and it doesn't come by very often. And so that's what kind of got me started on, on everything. And then came across Scrunch, got recommended from a former CEO of mine who's now an investor and told me to look into scrunch and like talk to, talk with the, the team here.
[00:06:16] And that really just sold me on, Hey, I gotta, I gotta join a company like this. I think, I think one kind of fun heuristic for why I joined is that and I've had success with this in the past, or using this in the past, is that, I, I hear about a pitch or like I talk to a, with a CEO and if over the weekend, overnight, I can't stop thinking about how to do marketing for that company or like my curiosity, just like wheels are spinning. I feel like that's a really good sign and like I definitely felt that with Scrunch over the weekend after, after talking with Chris, our CEO on a Friday. And so, for anyone in the audience listening, like if, if that's the case, if you're interviewing or considering a job and you're just can't stop thinking about it. I think that's a really good sign to, to make the leap in though. So that's what I did. And, and I've been very, very happy with my decision. Started, six months ago and things have just been taking off since. So, really great gr great to be in the space.
[00:07:03] I'm, I'm very excited about it and can share a lot more.
[00:07:06] Mike Allton: I think that's absolutely spot on. If you can't stop thinking about a new opportunity, whether that's a client or a new gig yeah, grab that bull by the horns and ride it. But let's get technical when a revenue leader, sales manager, marketing ops, whatever they, there's, they're asking, are we visible in AI search?
[00:07:25] What's like the first diagnostic that you run? What are you actually measuring at that point?
[00:07:30] Kevin White: Yeah. It's a question that I feel like. If you're not a marketer and asking it right now, you're gonna be asked it from, whether that's coming from your board or a CEO or some other person in the org.
[00:07:41] So, I can guarantee that that's the case for a lot of people where, this, this happens in SEO all the time too, where like, your CEO. Google is your brand or your category or something like that. You're like, Hey, we don't show up here. The same thing is happening on these LLM platforms, whether it's your board or your CEO or something.
[00:07:56] And if you're not showing up, you're already falling behind on the marketing side. So I think that there's, in that space, there's a lot of urgency around, figuring out how to just even report and show per show our performance metrics as a starting point. And so. To get at like what is, what that starting point looks like.
[00:08:13] I feel like a lot of people are starting out with just like referral traffic. How much referral traffic from Chat GBT or Perplexity or Gemini am I getting that is actually, in my opinion, is just like the very, very tip of the iceberg. If you think about the user journey and users clicking through the little pill referral link on one of the LLM platforms, that's like the last step in their journey.
[00:08:35] And so the preceding steps to get to that point are where I would start with start measuring and that there's a, there's a few, there's a few metrics that we we get Scrunch to help people get visibility into, and then also track over time. And so the, the core one I would say is brand presence or share voice.
[00:08:54] So essentially like answering the question, if someone's doing a search that's relevant on an LLM. Is my brand being shown or not? That is like the, the first thing that you wanna make sure you're showing up for. That's not gonna show in a, in a referral report, right? It's, that's like the referral traffic is very far downstream from that.
[00:09:14] The other thing that we, the other metric is citations or, are the LLMs referencing my domain to grab information and return it back to the user. So that is, in the, in the Google world, it's similar to a blue link or, or is your domain being cited as a point of authority?
[00:09:31] If it is, then that gives you some sort of control over your destiny of that core metric brand presence. And so th there, those are like the two kind of like core metrics to start on the visibility front. And then the other thing that the board and CEOs and people are gonna ask is, okay, we're tracking these things.
[00:09:48] We're tracking how people are searching in these LLM platforms and how our brand showing up if we're being cited or not, but how do we compare with our competitors? And so that's the other big thing that, that people are looking at is like, okay, we have this set of prompts that we're tracking. We're looking at the trends of are we showing up or showing or not showing up?
[00:10:05] Are we being cited or not being cited? But then also within our competitive set. Of people who would, of other tools who we compete with. Where are we in that hierarchy? Are we like always at the, the top of share voice? Are we at the bottom? Like, how can we improve that over time? Because that's what, that's what the board and people wanna see is like, within the prompts that we care about.
[00:10:24] Are we showing up? Are we showing up better and more often than our competitors? So that's what I would say is a starting point. And then the, there, there's lots more places we can go from there.
[00:10:32] Mike Allton: Yeah, for sure. Well, it's interesting though because it's somewhat contrary to a lot of the advice that I've seen.
[00:10:39] Everyone that I've read about, they're obsessing over things like getting mentioned on Reddit, and I think your data is showing that that's just not the silver bullet that marketers think it is. What is some of your research reveal about these citation sources that you talked about a second ago across like different industries?
[00:10:54] Kevin White: Yeah, a lot of the like, quote unquote experts out there. And I put expert in quotes because, it's a very new space. It hasn't been around for more than, two, maybe three years. And so I don't know if they're gonna be an expert in a category or like a new platform when there hasn't been the 10,000 hours to be able to put in.
[00:11:13] But a lot of people out there will say. You gotta show up in Wikipedia, YouTube, Reddit. Like, there's a lot of go-to citation sources that it's like, okay, we're showing up here, then we're gonna show up. Well, in the lms. Granted like that there, that is where a lot of the volume is at, like the, a top of funnel volume.
[00:11:30] You're gonna be, you're gonna be seeing citations across like Wikipedia and Reddit and substack and these kind of like go-to general knowledge type of sources. But what we find is that the more niche you get and the more bottom of funnel prompt or queries that you see, typically those channels.
[00:11:49] Like, Reddit or Wikipedia, like actually don't show up as a citation of authority as much as, a niche blog or a competitor's content or something that's a little bit more specific to that particular cohort. Intent that someone is searching for. And so what I mean by that is if someone's searching for, tell me the best, like CRM software like Reddit and Wikipedia and stuff like that, for something like some, a query like that is probably going to show up as an authoritative source and a citation.
[00:12:21] But if someone drills down even further and asks, tell me the best CRMI live in this region. I have, I'm in this industry, I have. X, Y, z security constraints. If you, the more particular you get and the more chats you have with the LLM, typically they will find that content that is written specifically for that query that the LMS are really good at, sussing that out and they will return it back to the user.
[00:12:45] And it's typically those citations are oftentimes surprising. They're not always Reddit, they're not always Wikipedia, they're not always YouTube. They're probably some blog or something that you've never heard of and someone who's covering this quarter of of your space or your market or category, writing about something that is pretty bespoke.
[00:13:02] But the LLM found and they do a really good job of that, of that. And then returning that back to the user. This is gives you an advantage where everyone's searching for or trying to get shown on Reddit or like Wikipedia is really hard to get any kind of mention in. They're very protective over that type of stuff.
[00:13:18] So, I think the core thing actually to care about is just like getting visibility into what those citations are. Like you're, you can't just blanket. Say, I'm gonna go after, whatever red subreddit because that, that's in my category. People are talking about it and expect to show up in LLM answers.
[00:13:35] You need to actually know the citations of authority as a first step, and then you can like use that for your citation and mention strategy to partner with these third party domains or whatever who are authoritative in your space. And so like just. Looking and getting visibility is the first step.
[00:13:50] And then oftentimes I feel like it's surprising that it's not always Reddit. So, and hopefully that answers the question, but I can go deeper if we need.
[00:13:57] Mike Allton: Yeah. In fact, you've got me thinking about how, I've always thought about SEO and specifically search traffic because it feels like we're in this crucible moment where we're gonna boil off all the fluff.
[00:14:11] Like if I was selling a CRM, me ranking for what is a CRM isn't particularly helpful really to the brand. It might not have been something that my content marketers and creators wanted to hear, but people seeing that and maybe getting onto an FAQ or something like that on my website, that probably was not going to drive a lot of conversions.
[00:14:32] But to your point, more that middle to bottom of funnel questions that already informed and engaged prospects are asking. That's the one that we really want our brand mentions for. Does that kind of resonate with what you're saying?
[00:14:44] Kevin White: Yeah. Yeah. Actually I, I, I, I call this the long, long tail. Like SEO had the long tail of search.
[00:14:49] I call this the long, long tail of search. Where if you think about the. Length of, of a prompt that someone will plug into an LLM versus what they would, what someone would search in Google. Maybe the in, in Google, I think it's like the average is like five to seven words. In an LLM especially people are talking into it.
[00:15:06] You're gonna get, a paragraph, like 20 plus words oftentimes. And so, in that world, because Google is more like zero, like traditional search, organic search is pretty zero sum. Like there's 10 blue links and the more specific you get with the search, oftentimes it's just gonna give you like the most authoritative, like top of funnel type of results.
[00:15:24] And so with lms, what they're really good at doing is finding that bespoke. Bespoke content around what that user is searching for and then returning it back in a way that is answering their actual question. And that's why people are, I think that's why I said they're getting so much adoption.
[00:15:40] It's like a really good answer and they're getting directly to it without having to like click through a bunch of blue links that maybe are not answering the questions so well. And so. W in this new world, you can almost create content that is just like talking about your specific differentiators in a way that that like maybe only you ha are an authoritative source over.
[00:16:01] And the LLMs are actually pretty good at finding that content in returning it back to the user when there's that intent on the other side of things. And so, it's almost like, like field the dreams, like you build it and they, they do come because, you don't have to rely on the 10 blue links anymore.
[00:16:16] Mike Allton: No, I've been so good conditioned to say that that doesn't work for 20 years as a marketer. But yeah, I couldn't agree more. I rarely use Google anymore. I use Perplexity almost a hundred percent, and I use Wispr Flow, so I'm always speaking into. These AI systems and trying to give it great context.
[00:16:33] 'cause I don't wanna just a list of random in this case, we'll just go with that CRM example. I want to tell it what kind of business I am and, and what I'm trying to accomplish and help it understand the exact kind of CRMI might be looking for. But help us, those of us listening and myself.
[00:16:47] Understand when we're talking about citations the, the, the, the different sources, right? So let's say, on the one hand you've got like a B2B SaaS company, maybe they're selling marketing automation. On the other hand, you might have, say, a FinTech platform. What are the different citation sources that are actually driving AI recommendations in different categories?
[00:17:07] Kevin White: Yeah.
[00:17:07] Mike Allton: Does that matter?
[00:17:08] Kevin White: The way that I, I don't know how much it matters. What matters is what the authoritative source. Is and then or grouping of sources are, and how you approach, getting your brand mentioned or not in that citation or like finding greenfield opportunities where there's like light amount of content there and like you can be the authoritative source.
[00:17:24] The way that I like to think about grouping it to inform that strategy is there's three different buckets. There's the citations that you own, like your own branded domain or set of domains are ranking, are are being cited for the results. Those are competitors who are probably harder to work with.
[00:17:40] So you have this competitive set of competitors who are doing the same thing that you're doing with their content strategy and, maybe they're showing up as a cited source for the prompts that you care about. And then there's the other bucket of third party citations, which is newsletters, Reddits, like Wikipedia. All the stuff we were talking about is like this other bucket, which is. Usually, it's not always the case, but usually like that's like the biggest bucket. And so when that's the case it gives you an informative, prioritized list of like, okay, these are the domains.
[00:18:14] That we, that are authoritative for the prompts that we care about. Like, let's partner with them. Let's pay to get our content placed in these different different publications or partner sites or review sites or whatever. So again, the first thing is understanding what those are by measuring it with a tool like Scrunch and then, you can prioritize based on like how many times that domain is being cited across all the prompts that you care about. And that gives you like this, like tops down list to say like, okay, these, these are the sites that we either need to partner with and get mentions in, or, maybe there's there, the content there is really light and that's informative for you to generate your own content or optimize your existing pages so that you become the source of the cited authoritative source and surpass the other citations that are, light on content that aren't actually answering the query at hand.
[00:19:07] Mike Allton: Got it. Got it. So let's talk about results. Because I mentioned at the outset you've helped some clients achieve some really amazing things. Talking about like 260% visibility gain in just a couple months. Walk us through what that actually looked like in terms of a workflow. What are the actual steps to go from?
[00:19:23] We're totally invisible. If I'm asking somebody in perplexity about my brand, I'm not coming up to, we're getting actually recommended by these tools like ChatGPT.
[00:19:31] Kevin White: Yeah. There, there's like a three step process there. The first thing would be just measuring benchmarking, where you sit.
[00:19:38] The way to do that is similar to keywords. You can even use Google search keywords to translate them into prompts or what someone would type into an LLM. You load all those prompts into an answer engine optimization products like Scrunch. There's also many others out there and. That gives you a starting point of like, okay, here's how much, how much your brand is being mentioned.
[00:19:59] Here's where your brand's being cited, here's other third party citations, all this kind of stuff. So that, that like baseline measurement and then gives you a kind of a starting point and kind of stack ranks you versus competitors versus where your, your brand presence is showing up or not in that set of prompts.
[00:20:14] You want to then look at how, where the wind is going or like how that's trending over time, because it's not deterministic, it's not 10 blue links. And so, one result one day might be a different result the other day. But if you measure and look at that performance over like a two week or one month window, that gives you a sense of like how you're performing overall.
[00:20:33] And so that's something that people kinda get hung up a a lot on with AI search. Is, the changing results. Whereas Google is a little bit more organic, search is a little more deterministic. But if you measure it over time, you get a good sense of like, if you're doing well or not.
[00:20:48] And then the part where our customers focus on is two different things. It's either, there's greenfield opportunity here for us to be an authoritative source across these prompts that we care about and no one's showing up well for 'em. And so we're gonna create net new content, or we're gonna optimize our existing content refresh it so that we get picked up and that we are addressing this, that prompt in the way that someone's asking an LLM about.
[00:21:13] There could be many things there. That to improve your existing content or create new content, which is like, you hear things like add FAQs, like add authorship, keep things refreshed, make content chunkable. All these things are, I would say like LLM lore right now. Like the LLMs themselves aren't telling you like this is actually what makes a difference.
[00:21:32] But those are the things that tend to get more pickup and you get pickup pretty fast in LLMs. And and I would say. Yes. Like, listen to what people are doing, but then try for yourself, make sure that like, test and measure for yourself. Otherwise, you're like flying blind and, and you shouldn't listen to experts like me or whoever else out there.
[00:21:51] You should actually, look at the results and make sure that like they're in a controlled kind of experiment. And then the other piece of the equation, aside from like content optimization and content generation would be citation or mention acquisition. And so when there are those third parties who are adequately addressing the prompts that you're tracking or what users are asking, it's doing things like reaching out or partnering with those authoritative sources to create new content or, get your brand or product placed into an article that's already being cited or ranking well.
[00:22:23] And making sure that you're showing up there. We actually work with this company called Noble, who is really good at, they, they, they have this like marketplace to broker that exchange, which is really great 'cause you can give Noble all the prompts that you care about. They will find this publisher network of third party sources and then do that kind of like in, in SEO or those like backlink building exchange.
[00:22:45] But they'll do that kind of brokering of like, Hey we'll give you this amount of dollars to show our brand here, or, we'll, we'll do a reciprocal mention. And like, they just kinda like tee that up for you. And so that's like a really great way to get mentions and mention acquisition and that is what gets you that like share voice or it gets your brand to be seen in the LLMs.
[00:23:05] And then. The last that's part getting into kinda like the action layer. The action layer is like actually doing it, executing that stuff. And like once you do that, then you'll start to see results of, within this prompt set that we are, we're tracking, this is how. Much we're how much share voice we're growing over time, or how much citation, how much more we're being cited.
[00:23:23] You can measure all that in a product like scrunch. And the interesting thing that our customers, and like anyone in this space actually is seeing really right now, is that by doing, by uncovering where there are opportunities and acting on it. Results actually really fast, especially compared to like the old SEO world, um mm, where, which takes like weeks to months.
[00:23:42] When you write a new article or get a new mention or something like that, you typically see the LLMs pick that up in like maybe a week or two's time, which is relatively pretty fast.
[00:23:53] Mike Allton: That's awesome. I particularly love your point, Kevin, when you say basically for folks who are, looking at a blank slate, they don't know where to start.
[00:24:00] Sure. Listen to someone like Kevin, but you have to test for yourselves. You have to experience for yourselves what's gonna move the needle for you folks. We're talking with Kevin White about how to make sure your brand and solutions actually show up. AI answers, and I've got seven more questions you're not going to want to miss.
[00:24:15] But first, a quick word about protecting your internal IP and guarding against Shadow ai.
[00:24:20] Right now, your best people are drowning. They're buried in data entry, meeting prep, and internal reporting. The admin drag that kills momentum. To keep their heads above water. They're turning to AI tools you haven't approved. They aren't trying to be reckless. They are trying to be efficient. They are trying to hit the goals you set for them, but this Shadow AI creates massive blind spots in your data security.
[00:24:49] You can't afford to ignore it, but you also can't afford to ban it. You need to architect it. I've created the executive guide to shadow AI to help you bridge this gap. It's a blueprint for turning a security risk into a competitive advantage through proper governance. Don't fight the shadow, sanction it.
[00:25:12] Download the free guide today theaihat.com/shadow-ai. That's theaihat.com/shadow-ai.
[00:25:25] So Kevin, let's, let's drill more into that technical side. 'cause you said a couple of things that, that I've heard of that other people have heard of are lore.
[00:25:32] What makes a website like actually AI readable versus just human readable. What are some specific technical optimizations that you're aware of that might actually move the needle?
[00:25:42] Kevin White: Yeah, so this is actually our premise or where we're headed with building what we're building at Scrunch, which is kind of a, a platform for almost like a CMS, if you will, for the AI bots that are crawled on your site.
[00:25:54] And if you think about the way things are headed with more users asking AI platforms for discovery, research comparison type of questions, there's less and less visitors probably coming to your site from those types of queries. And instead, you have this AI retrieval bot that is going in, ideally finding your site, getting information, and returning that back to the user.
[00:26:15] Now. The way that we build websites for humans with media, with JavaScripts, with all these different like superfluous code that is included on a web, on a typical webpage to make the human experience that the UX of it, like usable, friendly, immersive, et cetera. That is not the way that AI likes to consume content. Like AI likes markdown and JSON and like they want a part, it wants to go read and parse information very, very quickly. And so, there are ways to do that. Like, I, I mentioned some of those things before where, you can use schema or FAQs and it can like, find that type of content on your site. Like that's, that's one approach.
[00:26:56] On the more technical side of things though we have a product, we call it our agent experience platform. And what it does is it'll take the human version of your site. It'll render a AI optimized version of it by stripping out all of that superfluous code, adding maybe schema to it adding more information that matches the same intent of the page potentially and.
[00:27:20] Then AI can consume that content and in a better, more optimized way than it would if it was trying to struggle through going through the human experience of your site. And so the way that we do that is at the CDN layer. And so we identify bot traffic with CloudFlare or Akamai or Fastly or all these different integrations of content delivery networks.
[00:27:41] We will identify what type of bot is visiting the site, and then we will deliver content to that bot that strips away all the unnecessary elements that are needed for a human experience. And then if a human visits they get that same immersive experience that they're used to, or your site is optimized for.
[00:27:58] And so when you do this we tend to see lots of one more crawling of pages because it's, it, it takes the token count down by about two orders of magnitude. And tokens are the exchange of economy for the AI bots crawling your, your pages. And so if you can make that a lot lighter, they tend to to crawl more pages. And then on top of that, it's more information dense and so it can get the answers much faster and return those answers back to the user. And so, that is like the new world that we where we see things going. And then even on top of that, it could lead to agentic commerce where you just have standards and protocols of how to do a transaction or fill out a form or do whatever on the site.
[00:28:38] And the AI's just kinda like doing that all on behalf of the user and using your site or this ai ai delivery mechanism to, to do all the things that the user's trying to get to at, on the other end of the, of the table.
[00:28:51] Mike Allton: Got it. That makes sense. So it's like when, oh gosh, eight, 10 years ago Google released AMP for mobile experiences.
[00:29:00] Right? Right. So we had to, we had to have that amp version of every page. And if you had it set up right, it was all automated. But the AMP version was designed for Google to serve to mobile users as fast as possible. Right.
[00:29:12] Kevin White: Right. Yep. Yep. I remember.
[00:29:13] Mike Allton: Got it. Yeah. And now we don't have to do that anymore.
[00:29:16] So I'm sure these things will continue to evolve.
[00:29:18] Kevin White: Yeah. Google just released this is probably dating this, this episode, but they released web MCP, which is like schema markup for AI bots visiting your site. And so we don't see that we, I don't see these, these standards and, and things going away.
[00:29:32] And then there's also multiple different models that you need to support. And so that's like our opportunity is to, take that, take the complexity out of it for our customers and own all that process so they can just say like, Hey, this is my goal. And then we do all the like legwork and hard, hard part of managing how AI will re aite for you on the backend.
[00:29:52] Mike Allton: Love it. I, I really, I love it. I used to develop websites. I still maintain all my own websites. This sounds like a really fantastic approach, but I know you're also doing some other projects. You're not just running Scrunch, you're using it to market Scrunch. How do you use your own product to track competitive intelligence and optimize your content strategy?
[00:30:09] Kevin White: Yeah, so, just like our customers. We, we are our own customer. We're dogfooding our product. We have been a little bit behind in the game, like, and so one of the things that I had the, the joy of doing when I joined is like taking taking over instance, which is. Our, our own workspace and like, going through what our customer would go through to like, develop that empathy for the customer on the other end, identifying the set of prompts that we wanna track, loading them in, cleaning up the workspace.
[00:30:34] Like, it's funny when you work at a company, i've seen this multiple places and everyone has access to the tool. The tool like happens and the tool like fidelity and it gets, it becomes messy very fast. And you would think like, oh, this is the company doing it. Like they should have the most clean, best example.
[00:30:48] And so I came in and I was like, I really wanna make this happen this time. And so I I've been taking ownership of it and it's the same kind of stuff that we're the same way I'm, I'm growing Scrunch is the same thing that our customers would do with this approach. It's, looking at where there's content gaps, making sure we're really getting the right set of prompts that we care about that.
[00:31:07] If someone searches for something like that are, are prompts and LLM that. If we are the best tool for the job, we're gonna show up for that answer. And there is definitely some, some brand presence and market share, share of voice growth that we needed to do there. And so that, that helps us inform our, both our content strategy and our citation to mention acquisition strategy.
[00:31:28] And so it was essentially just like, drinking your own champagne or eating your own dog food. Like this is what we're doing and we've seen. By focusing on, on this, like we've created a whole repository of FAQ pages. We've taken our docs and we've made our docs more public and accessible.
[00:31:44] We use I think it's called Minify as a, as a product. And like that is a very good AI optimized docs product. And so by, by executing on these things and putting more content out there that is addressing the gaps that we're seeing on the measurement side of things. We've seen that similar type of growth.
[00:32:02] I think we've grown our brand presence in the last like four weeks. We've, around the core prompts that we're tracking have grown at probably 30% 30 to to to 40%. So there's still a lot of work there to, to be done, but we're walking the walk, I would say.
[00:32:17] Mike Allton: Love it. Love it.
[00:32:18] So. A lot of revenue leaders, revenue, revenue teams listening to this right now, they may be realizing that their pipeline's totally invisible in AI search. How do they like prioritize it? Like if they can only fix three things this quarter, what should those things be?
[00:32:34] Kevin White: Yeah, I think one thing is just recognizing and getting board level or CEO level buy-in that this is a new channel, this is how people are finding new products and finding new tools sharing that message or getting that message up to the company, like top level priority is is probably something that I would say is necessary, you need to have buy-in at that executive level to make this channel a priority.
[00:33:00] And then if you are, if you're a revenue leader and not, don't have control over this, if you're not on the marketing side of things, I think exposing not to call the marketers out. I hate when people do this to me, it's part of the job is like exposing like, Hey, I searched for this. We're not showing up here. We probably should. I think this is like where a lot of our buyers are, are asking questions and like, we need to have. Brand presence here and just kind of like highlighting that there's a gap. There is one way to make that happen.
[00:33:26] And then the, the other ways is just helping out the marketing partners that your coworkers out and when, so that would be things like trying to find brand mentions.
[00:33:36] Putting stuff on LinkedIn, like everyone doing their part to say like, Hey, this is an initiative. This is how our brand can get picked up in, in these LLMs. And making it a, a company wide goal, not just like a marketing goal is something that I feel like, goes a really long way when you can kinda like rally everyone around thinking about like, oh, I saw this, I saw this blog post here. We should, we, we were mentioned in it, our competitors were, but like, maybe I can help out my marketing team by surfacing this or brokering that relationship or it's all these kind of like things where, where the marketer can say like, this is how, this is our strategy. This is what we're doing.
[00:34:07] And then the other people enabling the other people across the org of how to. Go be, go out there, be helpful surface things is like how I approach it.
[00:34:16] So, yeah, I don't know if that answers the question or its three things but yeah, I think that's a, a good way to make it happen is like, get everyone on board.
[00:34:23] Okay.
[00:34:24] Mike Allton: Yeah, I love that perspective because I'm not trying to call out marketers either, but I think you'd agree this is a marketing problem, a marketing responsibility, but there's ways that other folks in the organization can help share the load. Is that fair to fair way to paraphrase what you just said?
[00:34:40] Kevin White: Yeah. It's a go to market problem. Like, whether it's a, I think the, the. As a marketer, if a sales person had a problem with like copy or messaging or positioning, like that is something that marketing can help out with, even though it's not quote unquote like marketing's ownership or responsibility.
[00:34:55] And then it goes both ways on the sales side of things too, if they, the sales, sales team or customer success team is like seeing something that, marketing could, could use some help out with, like chip in and, and we're all on the same team together, all driving towards the same goal. So I think that's a really good mentality to have.
[00:35:10] Regardless of what team you're on.
[00:35:12] Mike Allton: Yeah, so I think, A CRO or a CMO listening to this or a sales manager, they gotta realize that this is within their bailey wig to solve. But if you're not able to get the bandwidth, if you're not able to get this in your OKRs or get budget for it, well then, share this episode with the CEO and say, Hey look, here's an expert, Kevin White. He's saying these are the things that we have to do that we aren't doing up until now. We need to figure that out.
[00:35:33] But here's an audit question. Sure. Kevin, just, put on your marketer hat for a second. You're listening and you obviously the marketing team is Scrunch. What's the one prompt that you should go ask ChatGPT or Perplexity today to find out if you have?
[00:35:47] An AI visibly visibility problem, what would you type in?
[00:35:51] Kevin White: Yeah okay. So my cheat code here is to tell the AI what you want and ask it to come up with a prompt for you versus me giving you the prompts. But the way that I would go about doing that is just to give it your objective and say like, Hey, I want a brutally honest authentic way for you to, go out there and see what is, how my brand is being talked about or assessed in the market. Write me a mega prompt and look at all these different, different sources. Turn on extra thinking or extra, Claude Pro mode on, and then just have it, go write it, write a prompt for you one, and then run that prompt on that extra thinking mode that'll get you pretty better, like better than me saying like, prompt this a better answer and more objective answer, I think. So, so that would be the way I would approach it.
[00:36:42] Mike Allton: I actually really love that perspective if you guys listening, you wanna do this. Maybe you haven't been using AI a lot personally, your team members have been talking about it and so on. Definitely make sure that you include some context in that prompt who you are, who you're representing, what your business is, who you're trying to reach, that sort of thing. Give it to your website. Give it that kind of context so that it can create a great prompt for you and it will.
[00:37:03] That's the beauty and the real differentiator between AI and every other technology we've ever dealt with. You can ask the AI how to use it, how to help you give it a better prompt. And it will. It's fantastic. Kevin, you've been absolutely amazing. This has been really, I think, truly an important interview that everybody needs to listen to, but I know folks are gonna have more questions and they're gonna wanna reach out.
[00:37:26] If they do, where should they go?
[00:37:29] Kevin White: Well, you could always reach out our, go to scrunch.com. We have a seven day free trial. We have a way for you to run a brand audit for free and also talk to us if you wanna, get a demo and we can like, run your brand through our product and just cover, uncover the results, like the first benchmarking side of thing and show you how you, you compare to competitors.
[00:37:46] If you want to just connect with me personally LinkedIn is probably the best channel for that, where I'm most active. I'm posting about this stuff on a pretty regular basis and so, I'm sure just. Google search me like Kevin White LinkedIn or some Scrunch, Kevin White, Scrunch LinkedIn or something like that.
[00:38:00] You'll find me or LLM search, search me. You'll probably find me there too. I haven't done AI search for my own brand, personal brand yet, but hopefully I'm represented somewhat decently on there. Yeah, so those are the best ways to find me and hopefully we can throw some links in the show notes as well.
[00:38:15] Mike Allton: Fantastic. Yeah, absolutely. All the links that Kevin mentioned today, including his LinkedIn will be in the show notes below. Thank you, Kevin. Thank all of you for listening. That is, of course, all the time we've got for today, friends, but if you haven't yet talked to your team about how they're really using Ai, Shadow Ai, it's time.
[00:38:30] Learn what shadow AI is, why you should lean into it, and how to proceed for free at theaihat.com/shadow-ai Check it out and I'll see you in our next episode. Welcome to the grid.
[00:38:41] That's a wrap on another episode of the AI Hat podcast. I hope you've gained some valuable insights and actionable strategies to boost your business with AI. If you enjoyed this show, please subscribe, leave a review, and share it with your network. And don't forget to visit theaihat.com for more resources and to connect with me.
[00:39:00] Until next time, keep innovating and remember, the future of business is human and it's powered by AI.
