Qvery Podcast Episode #1: Founders Chat

Hey, Vlad here. So we did it. We recorded our first podcast episode. Just me and Piotr, the two co-founders of Qvery, sitting down and talking about our journey, the product, and honestly, just having fun.

I want to use this blog post to walk you through what we discussed, add some extra color, and give you the narrated version for those of you who prefer reading over listening. Let's get into it.

How We Met (12 Years Ago, No Less)

I kicked things off with a quick intro. I've been running marketing for SaaS brands for over a decade, started multiple SaaS companies, and I've always been obsessed with making brands discoverable on the web. Search engine optimization, content marketing, the whole works.

Piotr shared that he's been in IT for 14+ years, going through the full arc of software development: developer, team lead, architect. He's worked on mobile apps, web apps, backend, infrastructure, AI. The man has basically done everything except take a vacation.

We actually met about 12 years ago and kept in touch since then. We even did a project together during a StartUp Weekend once. So when I called him in early 2025 with this wild idea, it wasn't exactly a cold call.

Why We Started Qvery

For the past six years before Qvery, I was running a consultancy focused on making SaaS brands discoverable on the web. And somewhere towards the end of 2024, I realized something was fundamentally changing. The ROI from SEO and content marketing was going down. It was getting increasingly harder for brands to generate leads and customers from traditional search.

At the same time, we kept hearing from clients that people were discovering software brands on ChatGPT. We started getting requests like "how do you optimize for ChatGPT?" And honestly, in mid-2024, we had no idea.

We were trying to figure out how to see which brands are visible in ChatGPT and which aren't. And there was nothing back then. Not a single software tool doing that.

So around early 2025, I reached out to Piotr. "Hey man, I have this idea. I want to build software to measure and grow brand visibility on AI search engines. Do you want to do this with me?"

Piotr said "hell yeah." His words, not mine.

When I asked him what got him excited about the idea, Piotr explained that approaching his 40s felt like the right time to start something from scratch. He has a family, a stable career, but he's the kind of person who likes risk and exploring new stuff. He mentioned that we both have this spark to build something new, something useful. He said the idea was "very cool" because everyone will be asking ChatGPT and Google about products and companies in a conversational way, and there were no tools to measure or grow that visibility. Hard to argue with that logic.

Nine Months In and Everything Has Changed

Fast forward to February 2026. Qvery is nine months old. We started building officially in June 2025, and it has been quite a ride. In the AI space, June 2025 feels like ages ago.

What has been truly stunning is the speed at which the entire industry shifted from "maybe AI search is something important" to "this is definitely a priority." I haven't seen this kind of adoption speed in the marketing space. Ever.

There were always two camps among SEO influencers and marketers: the people evangelizing generative engine optimization and the people saying nothing is actually changing. Well, nine months later, everyone agrees this is the next big thing.

ChatGPT keeps growing. Google has rolled out AI mode in more than 200 countries. The agentic web is not just the future. It's present tense.

We put a stat on the Qvery website that 50% of search is going to be AI-driven by 2027, according to Gartner. I now think that's a gross understatement. By February 2026, we all know that 100% of search is going to be AI-driven by 2027. Whatever people search for, whatever recommendations they want, all of it will be AI-driven. And this is ultimately our mission at Qvery: building tools to help brands measure and grow their visibility in AI search.

Piotr chimed in here with a characteristically pragmatic take. He said it doesn't really matter whether we call it SEO, GEO, or AEO. The name doesn't matter. What matters is answering the question: how do you become visible in AI search engines? And how you achieve it is more important than what you call it. I agree completely.

What Makes Qvery Different

This is probably the most important part of our conversation. There are now 20 or 30 companies building in our space, and I want to say upfront: we hugely respect them. We're all trying to solve the same problem. But here's the thing.

Many LLM tracking tools are literally just dashboards that show you metrics. They might help you measure visibility in a certain context, but they don't actually help you grow it.

Many of these tools still operate on the concept of keywords. You need to upload your own keywords, which is really a concept passed over from traditional SEO tools. In AI search engines, you don't have keywords. You have queries. And people query very differently from how they used to search with keywords.

At Qvery, we do the whole thing automatically. You type your brand name and website, and Qvery takes it from there — generating everything you need to get started. No manual keyword uploads. No complicated setup wizards. No fighting with your boss to figure out the tool before you can actually use it.

And then we take it further by making your data usable and actionable through AI agents, not just pretty charts that sit in a dashboard collecting dust.

Piotr added an important point here. He mentioned that Qvery doesn't just look at your own website. It explores everything about your brand from across the internet, from competitors, and even from your CRM data if you have it. Your ICP information, for example, can be used to generate more relevant queries for exploration. The more context Qvery has, the smarter it gets.

The Engineering Perspective

I asked Piotr about what it takes to build AI-native software like Qvery. His answer was characteristically honest.

He pointed out something that everyone in marketing knows but rarely admits: most people don't know how to use their existing tools correctly. Google Search Console, Google Analytics, Ahrefs. Marketers struggle with setup and configuration. They would rather have someone, or something, do the work for them. And now, with AI agents, that's actually possible.

We are moving away from dashboards to something that will be done by AI agents.

From the engineering side, Piotr explained that AI coding agents are the key leverage behind Qvery. A small team can move fast and build at scale — no need for large engineering departments.

He made a fascinating observation about the future of interfaces. Right now we have browser agents that take screenshots and decide what to do in a loop. But Piotr thinks we're moving to a stage where agents communicate through APIs and protocols like MCP, not through visual interfaces. And Qvery is built API-first from the beginning, ready to expose its capabilities as MCP tools for external agents like Claude or ChatGPT.

He also described his current role as more of an "orchestrator and architect," preparing requirements and letting AI agents do the heavy lifting, jumping into the conversation when they need judgment or decisions. But he's still very much in the loop, reviewing code, understanding what's being built. It's not blind delegation. The codebase is designed to be scalable, pragmatic, and ready to grow from 20 customers to 200 without breaking a sweat.

The Vision: From Dashboard to Autopilot

I asked Piotr about his vision for how the Qvery workflow should evolve over time. His answer was clear and honestly pretty exciting:

• Step one: An AI assistant that understands your brand, your metrics, and your data, and helps you prepare everything to grow visibility

• The end goal: Everything on autopilot. You go to sleep, wake up, and see that Qvery agents prepared actionable insights, found opportunities, and suggest specific actions you should take

He gave a great example: imagine the agent notices an interesting thread on Reddit where someone mentioned your brand, and it prompts you to jump into the discussion. Or it spots a visibility opportunity and prepares a strategy while you were sleeping. Piotr described it as "babysitting" at first, where you teach the agent and give feedback, and over time it learns enough to run on its own.

It's like babysitting. Once the agents learn from your data and your feedback, they start doing the work themselves without any human in the loop.

My own dream for Qvery is to create hundreds of reusable templates for marketing teams. Imagine accessing these from Qvery Assistant: "take my LLM visibility data, compare against competitors, and create a content strategy." Or "write me a blog post about this topic and make sure it's different from competitor content." Or even launching an outreach workflow with one click based on your visibility data.

I also shared a story from earlier that day where I was demoing the GEO audit in Qvery Assistant to someone in real-time. It generated a full audit of their website in about 10 seconds. Their reaction was just "wow." That kind of instant, actionable analysis is exactly the direction we're heading.

Will Google or OpenAI Build What Qvery Builds?

Piotr asked me a question that I know a lot of people are thinking about: will the big players like OpenAI and Google release their own LLM analytics tools?

The short answer: sort of, but not really what we're building.

Bing actually just released basic LLM analytics in their Bing Search Console, which is interesting. But Google? I have a theory about Google.

Google's business is built around advertising, and that's precisely why their free analytics tools are, to put it diplomatically, not great. Google Analytics is a disaster of a platform. Google Search Console looks outdated. And there's a massive disconnect between the two: Analytics shows organic sessions and conversions, Search Console shows queries, but neither tells you which queries actually drive revenue.

The one place where you get the full picture? Google Ads. Pay for ads and suddenly you get complete analytics. Funny how that works.

I expect the same dynamic for AI search. Google probably won't reveal how the query fan out works in AI mode, probably won't show how many people convert through it. They haven't even figured out how to monetize Google AI mode yet, which is a massive internal problem for a company whose stock price depends on ad revenue growth.

So no, I don't expect Google to build the tools Qvery is building. Google isn't in the marketing software business. Our vision goes well beyond dashboards and basic analytics.

Piotr agreed and added that even if the big players build basic LLM analytics, that's just one side of the coin. The real value is in how you process the data and what you give marketers to actually grow their visibility. That's where things get interesting, and that's where every company can solve it differently.

The Three Pillars of AI Visibility

Before wrapping up, Piotr asked me to break down what brands can actually do to grow their AI search visibility. I shared our framework at Qvery, which rests on three pillars:

Pillar 1: Your own website. This is your opportunity to control the narrative. But here's the mindset shift that most marketing teams struggle with. In the old days, SEO meant creating long articles and publishing in volume. In the AI search world, it's about creating unique content that doesn't exist anywhere else. AI search engines only cite your content if they don't already know the information. They're looking for new perspectives, case studies, original data, and real-life experiences. You're essentially exchanging unique information for citations.

Pillar 2: External websites. It used to be all about building links. Now it's about building mentions. An unlinked mention is super valuable depending on the context. In the old days, you could have a link saying "this software is terrible" and it would still help your rankings. Not anymore. The context and sentiment around the mention is what matters now, not the hyperlink itself.

Pillar 3: UGC platforms. Reddit is the single most important website on the internet right now, responsible for close to 20% of all citations on Google AI mode and ChatGPT. Both OpenAI and Google get data from Reddit through APIs directly, and AI search engines are smart enough to recognize that recommendations on Reddit come from real people. The challenge? Reddit is hard to engineer. You need a strong brand and genuinely creative marketing to earn authentic mentions on a heavily moderated platform.

AI Engine Optimization really takes you back to the basics. Have a great product. Focus on brand building. Play the long game. Do genuinely good marketing. You will succeed.

The Agentic Web Is Already Here

We wrapped up talking about the bigger picture. The agentic web isn't some distant future. It's happening right now.

Brands aren't just competing for people's attention anymore. They're competing for agents' attention. Google's universal commerce protocol means AI agents will search, compare products, find the best prices, and buy on your behalf. That's pretty meritocratic. And also a bit terrifying if you're not prepared for it.

There are fascinating developments happening, like Cloudflare's agent passport system where agents need a "visa" to access certain websites. The agentic web could actually end up being more organized and regulated than the current web, with agents communicating through protocols and earning access based on merit and trust.

Piotr made a thoughtful point about data privacy on our platform. Each brand's AI assistant learns from their own data, not from competitors' projects. Your unique strategies and insights stay in your own context. So as agents get smarter for different brands, they compete on the quality of orchestration and strategy, not on who has access to whose data.

And then Piotr closed with something that I think perfectly captures what we're about at Qvery. Yes, everyone is going with the "our AI agents will do the stuff for you" pitch. But sometimes you just need a person with a nice smile on the other end of the call. We're here to help, discuss, and yes, make a joke when the AI world gets a little too intense.

It's an exciting time to be alive. It's an exciting time to be building AI software tools. Thanks for reading, and if you have any feedback or ideas, feel free to reach out. You know where to find us.