Milena Traikovich is a powerhouse in the demand generation space, known for turning complex analytics into high-performing lead generation engines that don’t just capture clicks, but drive actual revenue. As organizations move past the initial “shiny object” phase of artificial intelligence, her perspective on bridging the gap between busywork and bottom-line impact has never been more critical. Milena specializes in helping businesses navigate the transition from experimental AI pilots to sustainable, value-driven operations. In this discussion, we explore how marketing leaders can stop chasing the next tool and start building a disciplined AI portfolio that secures a lasting competitive advantage in an increasingly automated world.
Many organizations find themselves caught in a cycle of high AI activity with very little measurable value to show for it. How should marketing leaders shift their mindset to ensure they are chasing outcomes rather than just launching more pilots?
The shift starts by acknowledging that we are moving out of the “exploration” phase and into a period of heavy accountability. In the early days, it was enough to just experiment with tools to see what stuck, but that approach has left many teams with a fragmented mess of AI activity that doesn’t actually move the needle on revenue or mission success. You have to stop asking what tool you should try next and start asking where AI can create a durable, sustainable competitive advantage. This requires a disciplined reversal of the traditional adoption sequence; instead of starting with a vendor’s new capability, you must start with the specific business problem you are trying to solve. If you aren’t identifying the business outcome or the specific process improvement first, you’re just adding noise to an already crowded tech stack.
When evaluating new AI use cases, what are the most common “hidden costs” that teams overlook, and how do these factors impact the long-term feasibility of a project?
People often think the cost of AI is just the subscription fee or the implementation hours, but that is a dangerous oversimplification that leads to projects stalling out. The real weight of an AI investment lies in the “before and after” work—the grueling tasks of cleaning new data sets, conducting rigorous accuracy testing, and setting up ongoing model monitoring to prevent drift. You also have to factor in the heavy lift of change management and staff training, which can take months to truly take hold within a culture. If you ignore these elements, you’ll find that your “productivity tool” actually creates more work than it saves because your team is constantly troubleshooting or questioning the output. You have to be prepared for the reality that governance and ethics aren’t just checkboxes; they are active, resource-intensive parts of the operational lifecycle.
There is a lot of palpable anxiety among marketing professionals regarding job displacement and the rapid evolution of required skills. How can leaders build “human and AI team intelligence” while addressing these very real fears?
The anxiety is real, and it’s something you can feel in the room when you mention automation—people worry that their years of craft are being reduced to a prompt. Leaders have to address this head-on by pivoting the conversation toward “context engineering” and the importance of human judgment, which AI still lacks. While traditional tasks like basic summarization or translation are becoming less central, we are seeing a massive surge in the value of customer understanding, business acumen, and ethical governance. We are moving toward a model of tiny, agile teams that can deliver results at a scale previously reserved for massive departments, but only if those team members feel safe enough to experiment. Managers need to become “value storytellers,” helping their people see that AI isn’t there to replace them, but to elevate their roles so they can focus on the high-level strategy that actually drives growth.
You’ve suggested managing AI like a value portfolio rather than a collection of tools. Can you walk us through the “Defend, Extend, and Upend” framework and why it’s necessary to have all three?
Think of your AI initiatives as a diversified investment portfolio where you’re balancing immediate wins with long-term bets. The “Defend” use cases are your operational foundation—they are about reducing manual effort and speeding up production, which builds the internal confidence and “AI muscle” your team needs. Once you’ve secured those wins, you move to “Extend” use cases, which is where you start seeing direct impact on things like conversion rates, lower acquisition costs, and better personalization. Finally, you have the “Upend” category, which is where you take bigger risks to create entirely new value propositions or change how customers experience your brand. If you only focus on efficiency, you’ll get marginal gains but no real growth; if you only focus on the big “Upend” bets, you’ll likely take on too much risk before your organization is ready to handle the technology.
Measuring the success of AI can be notoriously difficult since the benefits are often intangible at first. What specific metrics should leaders be tracking to prove that their AI investments are actually paying off?
Measurement has to be tailored to the specific type of value you’re trying to capture, otherwise you’ll end up with a set of “vanity metrics” that don’t satisfy the C-suite. For your “Defend” initiatives, you should be looking at operational KPIs like output per hour, cycle time reduction, or the clearing of a persistent backlog. When you move into “Extend” territory, the conversation shifts to financial and marketing performance, such as pipeline contribution, sales impact, and overall revenue growth. For those high-level “Upend” projects, you need to watch leading indicators like customer switching behavior or early signals of new demand in markets you haven’t touched before. By defining these success metrics early and tracking them with the same rigor you would any other major capital expenditure, you remove the ambiguity that often surrounds AI and replace it with a clear story of business impact.
What is your forecast for the future of AI in marketing departments, particularly regarding team structure and the scale of operations?
I foresee a dramatic shift toward what I call “hyper-agile, tiny teams” that leverage AI agents and shared services to do the work of forty-person departments with just a fraction of the headcount. We are already seeing a community of over 40,000 marketing professionals who are actively looking for ways to integrate these tools, and that collective intelligence is going to accelerate the move away from siloed roles. In the next few years, the most successful marketing organizations won’t be the ones with the biggest budgets, but the ones that have mastered “AI agent management” and can pivot their strategies in real-time based on AI-driven insights. The competitive advantage will belong to those who treat AI not as a vendor plugin, but as a core competency that empowers humans to do more meaningful, creative, and strategically sound work than ever before.
