Profound with Josh Blyskal

In this conversation, Stacey Epstein and Josh Blyskal discuss the transformative impact of AI on marketing, particularly focusing on the concept of Answer Engine Optimization (AEO). They explore how traditional marketing strategies are being disrupted by AI technologies, leading to a need for high utility content that directly addresses consumer queries. Josh shares insights from his experience at Profound, emphasizing the importance of creating valuable content that can compete in an AI-driven landscape. The discussion also includes a case study of Ramp, showcasing how specialized content can significantly enhance visibility and engagement in AI search results. Finally, they speculate on the future of AI in marketing, highlighting the potential for more integrated and efficient systems.

Transcript

Stacey Epstein (00:01.762)
Welcome to the show! On today’s episode, I have Josh Blyskal from Profound. I’m really excited to have them on. You may have seen a recent LinkedIn post I put up—everyone is talking about what to do now that so much of our traffic seems to be disappearing.

Back in the early days at SuccessFactors, we were amazed by what Google paid search could do for lead gen and awareness. And for years, that's how we’ve done marketing. But with AI, the game is changing fast. I was talking to a potential client just last week who said their paid search leads are down 40%. We all know why—and I think that’s exactly why Profound exists. But Josh will explain it much better than I ever could and help educate us on something we all need to understand.

Let’s start with you, Josh. Tell us a bit about yourself.

Josh Blyskal (01:13.971)
Thanks for having me! I’ve been with Profound since almost the beginning—I was the second hire. I joined in September, and we came out of stealth in August. The mission has always been to help people understand how answer engines—like ChatGPT, Perplexity, or Google AI Overviews—perceive, recommend, or mention their brand, products, or services. We help companies take action and build a defensive moat in AI search.

My role is AI Strategist. I meet with customers, advise them on how to win in this new world, and stay on top of the latest trends. I’m deep in the trenches with what’s happening at the cutting edge.

Stacey Epstein (02:14.958)
That sounds like such a fun job. This is, in my opinion, the most fundamental change marketers are facing. So how did you end up here?

Josh Blyskal (02:33.617)
I actually started in SEO—though at first, I wasn’t sold on it. During my internship at HubSpot, I was assigned to SEO and honestly thought, “Really? Do we need to optimize for people clicking things? Don’t we just write content?” I quickly realized how deep the space is.

After the internship, I was set to return full time—and right before I did, ChatGPT launched. On day one, they asked me, “Want to research AI a bit? Maybe it’s going to matter.” Well, it did. It defined the business in 2022 and just kept growing.

I moved from SEO into AI strategy, helped prototype HubSpot’s AI Search Creator, and realized how critical answer engines would become. When I saw Dylan and James, Profound’s co-founders, launching this company and getting covered in The New York Times, I knew I had to be part of it. I emailed them and said, “There are maybe 20 people thinking about this in the world, and I’m one of them.” The rest is history.

Stacey Epstein (04:28.96)
Amazing. And now there are a lot more than 20—including everyone listening. And shoutout to HubSpot—they really did write the book on SEO. They’re still so forward-thinking.

In that LinkedIn post, I shared a chart from Andreessen-Horowitz showing all the companies in this space. I’ve heard this work called Generative Engine Optimization, or even LLM Optimization. Do you have a name for the category?

Josh Blyskal (05:20.051)
We like “Answer Engine Optimization.” There are other terms—GEO, LLMO—but to me, it comes down to solving for the query. Everything is now zero-click. You ask a question and get an answer immediately. These models are cutting out the middleman.

If you want to win in engines that provide answers, you need to optimize for them. “Answer Engine Optimization” just makes sense—it mirrors SEO nicely. Plus, terms like GEO get confused with maps, and AIO sounds like you’re training the model itself. AEO feels like the right fit.

Stacey Epstein (06:39.912)
I’m sold. And as someone who’s been part of a few category creation plays, I love that. If anyone should define this space, it’s you.

It also perfectly describes the shift: we’ve lost control. With paid search or SEO, we could build our own landing pages, control the message, and own the user experience. But now, prospects are getting an answer—not a link to our site. Marketers, if you’re wondering what that answer is, just go use ChatGPT or Perplexity and ask the questions your prospects ask. That’s your reality now.

So tell us—how do we influence what gets returned when someone searches for a solution in our space?

Josh Blyskal (08:28.499)
Great question. My three-word answer? High-utility content.

These engines don’t have time to crawl every product page and review. They go to listicles. If someone asks “What’s the best men’s running shoe?” they’re not crawling Nike, Adidas, and Puma. They’re pulling from a top-10 article and maybe a Reddit thread.

The winning pages are ones that offer direct comparisons—price, features, benefits. That’s where brands need to play. And yes, it can feel a little uncomfortable. But would you rather your competitor define the comparison?

Stacey Epstein (10:34.252)
That’s so interesting. But what if everyone just spams the internet with their own biased top-10 lists?

Josh Blyskal (11:03.475)
Totally valid concern. But high-E-E-A-T content still wins. That’s experience, expertise, authoritativeness, and trust—Google’s qualitative framework.

For example, to get found in ChatGPT, your content needs to be indexed well in Google. So spam doesn’t work. At Profound, we help brands figure out exactly how competitors are framing comparisons—how many products, what dimensions—and we show you how to create something more useful, even if it highlights your brand.

If ChatGPT sees your page as valuable—even if it’s biased—it’ll still use it. It wants comprehensive views of the industry. That’s your opportunity.

In the long run, I think we’ll see a two-tiered web: one that’s beautifully designed and human-centric, and another layer that’s structured and machine-readable. The latter will feed the models. Marketers can finally focus on design and storytelling again, while LLMs handle the functional data underneath.

Stacey Epstein (14:24.438)
I love that vision. So how does Profound help someone like me make that shift—away from paid search and into this new world?

Josh Blyskal (14:58.387)
One of my favorite stories is our work with Ramp, a fintech in New York. They wanted better visibility for their Accounts Payable product. They noticed a gap: most of the top-cited pages were general.

So they created two focused listicles—one for small businesses, one for enterprises. Very specific, very high-utility content. As a result, they 7X’d both their visibility and their citation share. It’s about spotting the gaps and filling them with high-quality, structured content.

Stacey Epstein (21:21.642)
So is Profound a software platform, or more of a services offering?

Josh Blyskal (21:31.175)
It’s software. I just happen to be the strategist embedded in the product.

Stacey Epstein (21:33.954)
Got it. So I license the platform and get to work optimizing for AI answer engines—and maybe get a consult with Josh. But not for long—you’re going to need an AI version of yourself soon.

Josh Blyskal (21:58.001)
Exactly. And we’re building it! But we’re also hiring—so if anyone’s interested…

Stacey Epstein (22:04.266)
Love it. I feel like we just scratched the surface. We’ll have to do a follow-up soon. But before we wrap—where do you see this all going? What’s the future?

Josh Blyskal (22:23.115)
So much to say. Short term: I recommend adding an llms.txt file to your site. It’s like a map for the models—showing what’s on your site, where to find it, and why it matters. Think of it as a richer version of a sitemap, written for LLMs.

Longer term: I think we’ll have agents that take action for us. Like, “Hey Siri, plan my kid’s birthday party. Make it nut-free, order a cake, invite the class.” All done through answer engines.

LLMs might become simpler, just serving as language interfaces, while more of the intelligence and task execution lives in external systems. It’ll all be interconnected, and more accessible than ever.

Stacey Epstein (25:49.494)
It’s all fun and fascinating. I’m so glad to be part of it—and even more glad you joined us today. Josh, thank you. I learned a ton, and I know our listeners did too. How can people find Profound?

Josh Blyskal (26:07.419)
Visit tryprofound.com. We just launched self-service—book a demo, explore the platform, reach out. Thank you!

Stacey Epstein (26:14.018)
Awesome. Thanks for joining us!