Milena Traikovich has spent years at the intersection of performance optimization and lead generation, helping organizations navigate the high-stakes world of demand marketing. As an expert who thrives on data-driven results and the nuances of lead nurturing, she has a front-row seat to how modern teams are actually using artificial intelligence. In an era where every department claims to be “AI-powered,” Milena focuses on the tangible, often messy reality of how individual contributors are hacking their way to efficiency. Her perspective is shaped by the realization that while leadership often looks for enterprise-wide solutions, the real breakthroughs are happening in the quiet corners of marketing and operations departments. Today, she joins us to discuss why the most valuable AI innovations in a company are often the ones nobody knows about and how to bridge the gap between individual “power users” and a cohesive organizational strategy.
The conversation centers on the shift from top-down AI mandates to a “bi-directional flywheel” approach. We explore the phenomenon of the “hidden power user” who develops massive, proprietary prompt libraries in isolation and why traditional training programs often fail to capture these grassroots innovations. Milena breaks down the essential roles needed to surface these signals—such as the AI Lead—and explains why creating horizontal paths for knowledge is the only way to prevent teams from solving the same problems months apart. By focusing on the concepts of sparking, spreading, and scaling, she provides a blueprint for building a “common brain” that allows a marketing organization to learn faster than the market changes.
Non-technical staff in marketing and operations often become the heaviest AI users because they handle messy, judgment-heavy work. Why does this segment of the workforce see higher engagement than the technical teams?
It comes down to the nature of the friction they face every single day. While technical teams are often building the infrastructure, the marketers and operations folks are the ones knee-deep in the “messy” work of interpreting brand voice, navigating audience nuances, and managing complex campaign constraints. In a study of a major financial infrastructure company, we saw that the top 10 token users weren’t the engineers; they were the people in customer support and marketing who were using the tools to bridge the gap between rigid data and human judgment. These individuals are inventing their own workflows because they have to—it’s a survival mechanism for getting through a crushing workload. They aren’t looking for a “technical” solution; they are looking for a partner that can handle the nuance of a creative brief or the repetitive nature of lead filtering.
You’ve mentioned the “nine-page prompt” as a symptom of a larger organizational problem. What does it say about a company’s culture when such a powerful tool remains a secret held by a single employee?
It is a bittersweet realization because it shows incredible individual initiative, but also a total failure of organizational intelligence. Imagine a marketing manager sitting at her desk, having spent dozens of hours refining a Google Doc that covers every pet peeve her editor has and every audience constraint the brand faces. She’s reaching for this nine-page masterpiece every day to produce high-quality briefs, yet she’s the only one using it while her colleagues are starting from scratch. To her, this isn’t some grand innovation that deserves a ticket to the IT department; it’s just her “secret sauce” for not running out of tokens or losing her mind. When these breakthroughs stay buried in private documents, the company loses its “common brain,” and the individual is left feeling like they are the only ones figuring it out, which is a recipe for burnout and siloed knowledge.
Most AI enablement programs act as a one-way street, pushing information from the center out. Why is this top-down approach failing to capture the most important discoveries?
The problem with a center-out approach is that the people designing the training are often the furthest away from the daily “grind” where the actual discoveries happen. A central team might pick a tool and publish a playbook, but they aren’t the ones realizing that the tool fails when a certain brand constraint is applied or that a specific prompt structure works 10X better for B2B lead nurturing. When knowledge only flows outward, you end up paying for the same lesson over and over again because there’s no mechanism for the “front-line” people to feed their wins back to the top. We see organizations where three different business lines are all solving the same problem in isolation, simply because nobody thought to ask the marketing managers what they had already built. AI literacy shouldn’t be a lecture; it needs to be a flywheel where the people doing the work are actively contributing to the curriculum.
To fix this, you suggest the role of an “AI Lead” whose primary job is “noticing.” How does this role change the way a team identifies and scales innovation?
“Noticing” is a skill that is tragically absent from most job descriptions in 2026. An AI Lead isn’t just a technical supervisor; they are someone who keeps their ear to the ground to catch those casual mentions of a nine-page prompt or a clever Slack workflow. When a marketing manager mentions in passing that she’s found a way to automate her creative briefs, the AI Lead is the one who recognizes that “signal” and realizes it has value for the entire organization. Without this role, these sparks of genius just flicker and die in private folders or disappear when an employee leaves the company. The AI Lead’s responsibility is to capture that signal, refine it, and turn it into a standard practice that everyone can use, effectively turning an individual win into an organizational asset.
Once a signal is noticed, it needs a “landing spot” to be useful. What does a functional landing spot look like for a marketing organization?
A discovery that gets noticed but stays buried in a chaotic Slack thread or a random email chain is essentially useless. You need a named function—a specific person or team—with the authority to take a raw discovery, like a massive prompt, and “productize” it for the rest of the company. This means cleaning up the language, ensuring it aligns with current brand standards, and putting it into a format that is easily accessible to everyone. It’s about creating a repository of “standard practice” that isn’t a static PDF, but a living, breathing part of the workflow. If there is no formal place for these signals to land, people will eventually stop sharing them because they don’t see their contributions being scaled or valued.
We often see consumer marketing and B2B groups solving identical problems months apart. How can horizontal paths for patterns prevent this kind of waste?
This is the classic organizational silo problem amplified by the speed of AI development. If your consumer marketing group solves a complex prompt-engineering problem in March, but your B2B group doesn’t figure it out until July, you’ve effectively wasted four months of productivity. Horizontal paths are about ensuring that what one group learns is immediately visible and applicable to the other, regardless of their different target audiences. It requires a common language and a shared platform where discoveries are tagged and shared across departmental lines. When you have these paths in place, the entire organization starts to move at the speed of its fastest learner, rather than being held back by the slowest silo.
The concept of the “Common Brain” seems central to your philosophy. How does a marketing team know when they’ve successfully built one?
You know you have a “Common Brain” when you stop hearing people say they are “figuring it out on their own” and start hearing them ask, “What’s the current best practice for this?” It’s a shift in culture where the collective intelligence of the 40,000+ marketing professionals in the wider community is reflected in your internal operations. You start to see people building on top of each other’s prompts rather than starting from a blank page every morning. The emotional weight of the work changes; it feels less like a lonely struggle against a “messy” workload and more like a collaborative effort where everyone is contributing to a central, evolving library of expertise. When a new person joins the team and can immediately access the “nine-page prompt” that took someone else a year to perfect, that’s when you know the flywheel is truly turning.
What is your forecast for how marketing organizations will evolve as they move from scattered AI wins to this integrated flywheel model?
I forecast that within the next few years, the competitive advantage of a marketing organization will no longer be the specific AI tools they use, but the speed at which they can turn individual employee discoveries into collective institutional knowledge. We are moving away from the era of “AI as a tool” and into the era of “AI as a culture,” where the organizations that thrive will be those that have mastered the Spark, Spread, Scale, and Sustain framework. By the end of 2028, the companies that still rely on one-way training programs will find themselves buried under “AI debt,” while those with a bi-directional flywheel will be operating with a level of agility and creative precision that was previously unimaginable. The “Common Brain” will become the most valuable asset on the balance sheet, as it allows a company to adapt to market shifts in days rather than months.
