Milena Traikovich is a powerhouse in the world of demand generation and revenue operations, known for her ability to transform chaotic lead pipelines into high-performance revenue engines. With a background deeply rooted in analytics and performance optimization, she has spent years helping B2B organizations navigate the complexities of data-driven marketing. As we move through 2026, the landscape has shifted from simple automation to a complex ecosystem of autonomous AI agents. Milena has been at the forefront of this transition, observing how the rapid adoption of AI is both solving old problems and creating entirely new categories of operational risk. In this conversation, we explore the current state of AI readiness, the hidden dangers of “agent sprawl,” and why the future of go-to-market success depends more on human governance than ever before.
Many organizations currently struggle to track exactly how many AI agents are modifying their customer records. From your perspective in revenue operations, how did we reach a point where nearly one-third of professionals can’t even count the tools automating their data?
It is a classic case of innovation moving significantly faster than the infrastructure designed to support it. We have seen a staggering 93% of go-to-market teams deploy at least one AI agent, but because these tools are so easy to “toggle on” within existing software, they have slipped under the radar of traditional IT procurement. I have walked into organizations where the leadership thinks they are managing a single bot, only to discover three or four different agents buried in their tech stack, each making autonomous decisions. When nearly 30% of teams report finding actions taken on their records without any audit trail, it creates a “ghost in the machine” environment that is incredibly unsettling for any RevOps professional. We are essentially automating our decision-making at a pace that has outstripped our ability to document or even identify the source of those decisions.
The data suggests that over half of teams see data hygiene as their primary hurdle for AI transformation. Why does AI seem to amplify existing data problems that MOps teams have been wrestling with for years?
The reality is that AI agents are consumers of the same messy customer data and fragmented workflows that humans have struggled with for a decade. While 70% of practitioners admit that poor data hygiene has actively degraded their go-to-market execution, the stakes are now much higher because an AI doesn’t have the “common sense” filter a human does. If an agent is fed poor data, it will execute a bad decision with a level of speed and volume that a human could never achieve. This is exactly why 45% of AI initiatives are currently stalling; leaders are realizing that you cannot build a sophisticated AI layer on top of a broken foundation. We are seeing a shift where 37% of teams are forced to pause because of undocumented processes, proving that AI is acting as a giant magnifying glass for every flaw in our operational logic.
We are seeing reports of “agent collisions,” where multiple tools contact the same prospect or marketing sequences fire while a rep is mid-negotiation. How can leaders prevent these overlaps from destroying the trust they have built with their audience?
These collisions are the ultimate nightmare for customer experience because they make the brand look uncoordinated and robotic. When 27% of organizations report that multiple tools or agents are hitting the same prospect simultaneously, it signals a complete breakdown in internal communication. I have seen instances where 17% of marketing sequences fire while a sales rep is actively working a deal, which effectively sabotages the human relationship being built. To stop this, we have to move away from siloed agent deployment and toward a unified set of business rules that every bot must follow. It is about creating a central “air traffic control” system where agents are required to check a prospect’s status across all platforms before they are allowed to send a single email or update a field.
With 69% of teams using AI embedded in existing tools and over 60% building custom applications, the “agent sprawl” is a growing concern. What are the hidden costs of having so many different platforms—like HubSpot, Outreach, or custom LLM APIs—deploying agents simultaneously?
The primary hidden cost is the fragmentation of the truth within your CRM. When you have 46% of teams using agent platforms like Agentforce or Gemini Enterprise alongside custom-built apps, you end up with different agents operating on different versions of the same data. One agent might be enriching a lead based on outdated info while another is scoring it based on real-time behavior, leading to a massive discrepancy in how that lead is handled. This confusion is why 31% of practitioners now say their top priority is a complete, cross-system audit trail. Without that visibility, you aren’t just losing track of your agents; you are losing the ability to tell a coherent story to your customers, which eventually leads to a degraded brand reputation and lost revenue.
Ownership seems to be a major point of contention, with 42% of teams using cross-functional committees and 19% having no formal owner at all. Who should be steering the ship when it comes to AI go-to-market strategy?
While the cross-functional committee approach is popular, it often lacks the tactical teeth required to manage daily operations. It is deeply concerning that 19% of organizations have no owner at all, leaving AI in an ad-hoc, “Wild West” state. I believe RevOps should be the natural home for this, but they are currently the designated owners at only 18% of companies, largely because they are already spread too thin. When 66% of operations teams say they are at or over capacity with zero room for strategic projects, it is clear we have a resource problem. We cannot expect teams that are already drowning in daily tasks to suddenly govern a complex fleet of autonomous agents without additional headcount or a major shift in priorities.
What is your forecast for the future of AI in Go-to-Market operations?
I forecast that the next 18 months will be defined by a “Great Rationalization” where companies stop collecting agents like shiny toys and start focusing on centralized governance. Currently, only 8% of organizations describe their AI operations as fully optimized, which means the vast majority of us are still in the experimental phase. We are going to see a massive push for tools that provide a “single pane of glass” view of every automated action, making agents follow the exact same rules as human teams. Success will no longer be defined by how many agents you have running, but by how well those agents are synchronized to provide a seamless, human-like experience for the prospect. The organizations that win will be those that prioritize data hygiene and clear documentation as the essential fuel for their AI engines.
