An interview with Justin Corres
We’ve been tracking the intersection of artificial intelligence and human expertise across industries, including libraries reimagining their futures and organizations navigating the complexities of AI-driven workflows. Our latest conversation brings that lens to the world of digital marketing. Justin Corres, Co-Founder and Managing Director of Paid Media at STOCK, offers a grounded, practitioner’s perspective on what it actually means to build an AI-native agency without losing the human judgment that drives real results. In an era when many marketers are racing to automate everything in sight, Corres makes a compelling case for knowing exactly where the machine should stop, and the strategist should step in, a theme that resonates across virtually every domain we cover at Serious Insights.
Top 3 Takeaways
- AI-native doesn’t mean AI-only — human judgment is the non-negotiable differentiator. Corres draws a sharp line between tasks AI handles well (bid management, keyword harvesting, performance reporting, A/B testing) and the work that still demands human expertise (strategy, brand positioning, client relationships, creative direction). His sharpest warning: the real threat isn’t AI replacing marketers, it’s marketers outsourcing judgment to AI and calling it strategy.
- Authentic, specific content outperforms high-volume AI-generated output. Rather than flooding channels with AI-produced content, Corres argues that a smaller amount of genuinely useful, perspective-driven content builds more trust and audience engagement. Volume without a real human point of view causes audiences to tune out; a principle with implications far beyond marketing.
- LLMO (Large Language Model Optimization) is the next frontier of brand discoverability. Corres introduces the concept of “citation equity,” which helps ensure a brand is consistently referenced across trusted third-party sources, enabling AI systems to confidently surface it in responses. His agency’s case study demonstrated a jump from 10% to 40% LLM discoverability while actually producing less content than competitors, underscoring that quality and credibility now matter more than sheer volume in the age of AI-driven search.
The Justin Corres Interview
STOCK describes itself as “AI-native, but not AI-only.” Where should AI take over marketing work, and where should human judgment remain non-negotiable?
AI takes over the repetitive, data-heavy, high-volume tasks well, such as bid management, keyword harvesting, performance reporting, A/B test execution, audience segmentation, and content scheduling. These are areas where speed and consistency matter more than intuition and where we feel AI is best at handling at its current stage.
Human review and judgment stay non-negotiable in strategy, brand positioning, client relationships, and creative direction. There are tools out there that can use AI to draft a hundred ad variations, but where it falls short is the direction that actually fits the brand’s voice or where the market is heading. This is where it requires someone who understands the business at a level AI doesn’t have access to.
The real risk isn’t AI taking over too much. It’s marketers outsourcing judgment to AI and calling it “strategy.” We use AI at STOCK, but every important and significant decision still runs through a team member with years of expertise and who owns the outcome. That accountability mattered before and still matters today.
Many marketing teams are using AI to produce more content faster. How do you keep that from becoming a volume strategy that overwhelms audiences rather than building trust?
The brands drowning audiences with content aren’t using too much AI. They’re using AI without a clear point of view and again, direction. Volume without relevance or a plan isn’t a content problem. It’s a strategy problem.
What keeps production from becoming just noise is having something real to say in the first place. AI can execute efficiently once you know what the core message is, but if you’re using AI to fill a content calendar because you feel like you need to post just to post, is where audiences tune out or never catch traction.
At STOCK, we think about content in terms of what it’s actually doing for the brand and its audience, not how much of it exists. A smaller amount of genuinely useful, valuable and specific content builds more trust than a high volume of polished and generic output.
STOCK emphasizes profit, incrementality, contribution margin, and sustainable scaling over vanity metrics. What marketing metrics should be retired, and what should replace them?
Some of them we believe are follower count, impressions, and raw click volume because these are visibility indicators, not business indicators. All they tell you is if people saw it or noticed something, not if it mattered.
What should replace them are metrics tied to actual business outcomes, such as cost per acquisition, contribution margin per channel, new customer revenue versus returning customer revenue, and incremental ROAS for paid media. Put simply, if a metric can look great while the business is stagnating, it’s probably a vanity metric and not worth putting a lot of weight on when determining the health of the business and progression of marketing campaigns and outreach.
The shift we push here at STOCK is from “how much attention did we get” to “what did that attention cost and what did it produce.” Those are different questions, and they lead to completely different decisions and directions.
Where should a brand realistically start with AI in 2026: content production, social listening, customer segmentation, retail media optimization, search/LLMO, or campaign testing?
It really depends on where the biggest inefficiency or opportunity is right now. But if I had to pick a starting point that delivers clear, fast value for most brands, retail media optimization is probably the strongest candidate. The feedback loop is short, the data is available, and the impact to most optimizations made show up in revenue directly.
Content production is the most accessible entry point, but also the most oversaturated. Everyone is using AI for content now, so the competitive edge there has compressed even more so recently.
For brands serious about long-term differentiation, LLMO is the area worth investing in early. We believe AI-driven discovery is changing how customers find brands, and most companies haven’t even started building for that yet. Getting in early means building what we call “citation equity” and brand discoverability in AI-generated answers before competitors get a chance to figure out what’s going on when they realize they are starting to lose market share.
AI can accelerate insights, testing velocity, and content production workflows. What does a good AI-enabled marketing workflow look like from brief to measurement?
The brief still has to be made by a person. You need someone who understands the brand, the customer, and the goal before AI touches anything. That context doesn’t come from a prompt template but instead from talking to the business owner, understanding where they come from, where they want to take the business and what’s truly important to them.
From there, AI accelerates the middle process, which includes research, content drafting, creative variations, audience building, and campaign setup. What used to take a team several days before can be completed in just a few hours. The time savings and improved efficiency is real here.
And then measurement closes the loop, and that’s where a lot of teams fall short. They let AI run fast at the front and then evaluate results the same way they always did. We can’t do this anymore. The better approach is building measurement into the workflow from the start, running faster tests, and shortening the feedback cycle between launches and learnings. The biggest mistake we see is treating AI as a production tool only, but the real value is in faster iteration and better decisions, not just faster output.
How is AI changing brand storytelling when buyers are increasingly skeptical of polished, automated, and obviously optimized content?
Overly polished and optimized content is losing trust because people are starting to recognize it more easily now. The tells are everywhere: the predictable structure, the generic imagery, the language that sounds like it was written by committee. Audiences, especially younger ones, have strong instincts for it.
What’s working is the opposite. Raw, specific, perspective-driven content such as a founder talking directly to camera about a real problem they saw and how they solved it, or a behind-the-scenes moment that wasn’t staged. These are content that have a distinct point of view, even when that view is uncomfortable or can be polarizing.
AI can still play a role here, but it’s in the infrastructure, distribution, and targeting, not in authenticity and trust. The storytelling that builds brand trust has to feel like it came from a person with an actual opinion, not a machine.
Your work stresses human connection and relatable narratives. How does a brand preserve authenticity when AI is helping write, target, personalize, and optimize the message?
Authenticity doesn’t come from the absence of tools; it comes from having something real underneath them. A brand that has a clear point of view and message, a genuine relationship with its audience, and content grounded in real human experience can use AI to help support scaling efforts, without losing that authenticity. A brand without these things will sound hollow, whether AI is involved or not.
What we tell clients is to think of AI as production infrastructure, not as the source of the message. The human has to supply the actual perspective, and if that perspective is clear and distinct, AI can help amplify it without diluting it. But if it isn’t, no amount of optimization will fix that.
The brands getting in trouble are the ones trying to use AI to manufacture a voice they haven’t built yet or just doesn’t make sense, That’s the tell audiences are picking up on.
AI search and LLM-driven discovery are changing how people find brands. What should marketers do differently when customers may encounter a brand first through an AI-generated answer rather than a search result or social post?
The first thing marketers need to accept is that the path to brand discovery has been changing ever since the mass adoption of AI. A customer’s first touchpoint with your brand may now be an AI-generated answer that you didn’t write, didn’t control, and may not even know happened. In some cases, the customer may not even know that it was an AI-generated answer. And that’s a fundamentally different problem than SEO or paid media.
What we focus on at STOCK through our LLMO workflows is making sure brands are building the kind of credibility and citation presence that AI systems draw from. What this means is being consistently referenced across trusted third-party sources, having a clear and specific category presence, and maintaining the kind of brand signal that makes an LLM confident in surfacing you over another competitor’s content.
This tactical shift is thinking about share of voice in AI responses rather than just rankings. If a customer asks an AI to recommend the best product in your category and your brand isn’t showing up, that’s a distribution and credibility problem, not just a content problem, and in the end needs to be treated as one.
Retail media, paid media, social, SEO, and product content often sit in separate teams. What breaks when those functions remain disconnected, and how should AI help connect them?
What breaks most visibly is the budget. Channels running in silos optimize for their own metrics, and there’s no one accountable for what the whole system is producing together, and this is important to understand. With channels siloed, you end up with situations where a brand is winning on ROAS in paid media while losing on organic share, or pouring spend into social media while product listing content is working against conversion.
At STOCK, we’re aware of all the moving parts that need to be understood and the importance of having a clear and cohesive marketing message across all channels. A campaign can be running well from an ads perspective while inventory issues, weak PDPs, or inconsistent brand signals are undermining everything downstream.
AI can help connect those functions by surfacing cross-channel signals that humans in different departments wouldn’t see together. For example, AI can help analyze data from one channel, cross-reference it to performance on another, and flag what’s worked, what hasn’t, and recommend next steps. But the underlying issue is organizational, not technical, because AI doesn’t fix a structure where five teams are optimizing independently and no one owns the overall outcome.
Give me one good example of AI driving measurable results in marketing where the value came from better decisions, not just faster production.
The clearest example I can remember is an LLMO engagement where we used AI to analyze competitor content and identify which topics were actually being favored by LLMs in that category. Some competitors were publishing 200+ articles a month, but we went the opposite direction and produced 8 to 10 highly optimized articles per month, each one built specifically to earn credibility with the models rather than just fill a content calendar.
The AI analysis wasn’t producing content but instead was informing the decision about what to write, how to position it, and where the real citation opportunities were.
In about four months, the client went from a discoverability score below 10% to over 40% across the major LLMs. The competitor publishing at 20x our volume didn’t move meaningfully in that same window and actually started to lose market discoverability and share of voice.
That’s the version of AI value we care about at STOCK. It wasn’t faster production; it was a better read on what the models reward, and the discipline to build toward that instead of just chasing volume.
About Justin Corres,
Co-Founder and Managing Director of Paid Media, STOCK marketing agency
Justin Corres is the Co-Founder and Managing Director of Paid Media at STOCK marketing agency and one of the agency’s founding experts. Over the past decade, he has managed more than $55 million in e-commerce advertising spend and helped multiple brands reach their first $1 million in revenue through strategic paid media and retail media management. At STOCK, Justin leads the agency’s performance advertising practice across retail media networks, paid social, and search. His work focuses on how paid media strategies must evolve alongside AI-driven product discovery, helping brands structure their advertising and data to build “machine confidence” with AI systems that increasingly influence purchasing decisions.
For more serious insights on AI, click here.
Did you find this interview with Justin Corres useful? If so, please like, share or comment. Thank you!
The cover image is AI-generated from the author’s prompt and Aravind’s source photos.
Leave a Reply