The rapid deployment of artificial intelligence within go-to-market (GTM) functions is colliding with the reality of inadequate underlying data infrastructure. According to a new survey from LeanData, while 79% of revenue operations, marketing operations, and sales professionals are currently scaling or deploying AI agents, 70% of respondents report that poor data quality is actively undermining their results. This discrepancy suggests that the primary barrier to successful AI adoption is not the capability of the models themselves, but rather the structural readiness of the enterprise environments in which they operate. The findings, derived from a study of 157 GTM practitioners, highlight a growing "AI reckoning" where ambitious technological implementation is being constrained by legacy data and process issues.
Infrastructure Gaps Stalling AI Transformation
The survey results indicate that the friction points in AI adoption are rooted in organizational and structural deficiencies rather than the performance of AI tools. LeanData reports that 55% of respondents identified bad data and a general lack of AI readiness as the primary obstacles to their digital transformation. When specifically asked why AI initiatives underperform or stall, the respondents identified three distinct categories of failure: 45% cited inconsistent or incomplete CRM data, 37% pointed to undocumented business processes, and 32% highlighted the impact of siloed teams.
These figures suggest that the "intelligence" of an AI agent is effectively limited by the quality of the inputs it receives. LeanData CEO Katy Keim notes that while executives have pushed for rapid AI movement, teams are finding that business processes often exist only in the minds of employees rather than in documented systems. This lack of formal structure, combined with fragmented views of the customer across different departments, creates a scenario where AI agents lack the necessary context to execute reliably. Consequently, the problem is being framed not as a failure of AI technology, but as a failure of the GTM infrastructure required to support it.
Growing Executive Anxiety Over AI Agent Control
Beyond the technical challenges of data integrity, the survey reveals a significant psychological and operational barrier: a lack of trust in automated decision-making. A majority of GTM professionals, totaling 60%, expressed fear that AI agents might act upon inaccurate data, process incorrect records, or violate established ownership rules. This sentiment indicates that the current enterprise appetite is shifting away from a desire for more autonomous agents and toward a demand for greater control and transparency.
The data suggests that for AI to move from the experimentation phase into reliable execution, organizations must prioritize the creation of a complete audit trail. The fear of agents violating ownership rules or making errors based on "dirty" data points to a critical need for orchestration layers that can validate AI actions against business logic. LeanData is positioning its Intelligent GTM Orchestration platform as a solution to this tension, suggesting that a "thoughtful backbone" allows teams to maintain speed without sacrificing the ability to trust the output. This shift in focus from model capability to execution infrastructure is becoming a central theme for revenue operations leaders attempting to stabilize their AI-driven workflows.
Key Takeaways
- 79% of revenue, marketing, and sales professionals are currently scaling or deploying AI agents, yet 70% claim poor data quality is undermining their results.
- The primary drivers of stalled AI initiatives are inconsistent or incomplete CRM data (45%), undocumented business processes (37%), and siloed teams (32%).
- 60% of GTM practitioners expressed concern that AI agents could act on inaccurate data or violate established ownership rules.
TechInsyte's Take
In our view, the LeanData findings signal a critical pivot point for enterprise IT and RevOps leaders. For the past several years, the strategic focus has been on selecting the most powerful LLMs and agents; however, this survey signals that the "intelligence" of these tools is secondary to the integrity of the data pipeline. We are seeing the emergence of a "garbage in, garbage out" crisis at an enterprise scale. If 60% of practitioners are afraid to let agents run autonomously, the ROI on AI investments will remain theoretical rather than realized. This suggests that the next wave of enterprise spending will likely shift away from pure AI model procurement and toward data cleansing, process documentation, and orchestration layers that provide the necessary guardrails for automated execution.
Questions & Answers
How is poor data quality specifically impacting AI GTM execution according to the research?
The research indicates that 70% of respondents feel poor data quality is undermining their results. Specifically, 45% of practitioners identified inconsistent or incomplete CRM data as a primary reason why AI initiatives stall or underperform, suggesting that AI agents lack the reliable inputs required for accurate execution.
What are the primary non-technical barriers to successful AI transformation in GTM teams?
Beyond data quality, the survey identifies undocumented business processes (cited by 37% of respondents) and siloed teams (cited by 32%) as major structural barriers. These issues suggest that AI failure is often a result of organizational coordination and process maturity rather than a deficiency in the AI models themselves.
Why are GTM professionals hesitant to increase the autonomy of AI agents?
A significant 60% of surveyed professionals expressed fear regarding AI agent autonomy. Their concerns are centered on the risk of agents acting on inaccurate data, processing wrong records, or violating established customer ownership rules, highlighting a critical need for auditability and control.
What does the survey suggest about the current stage of AI adoption in revenue operations?
The data suggests that AI has moved from the experimentation phase into the execution phase, as evidenced by the 79% of professionals already scaling or deploying agents. However, this transition has triggered an "AI reckoning" where the limitations of existing GTM infrastructure are becoming apparent.
Source: Businesswire
