The publication of the Enterprise Data Health Study, a collaborative research project conducted by diginomica and Maureen Blandford of Serendipitus, has signaled a significant shift in how the technology industry perceives the current state of corporate information management. Based on 18 in-depth, anonymous interviews with senior enterprise practitioners, the study exposes a profound "trust collapse" within the data ecosystems of major organizations. This research identifies a phenomenon termed the "verification tax"—an invisible yet compounding organizational cost associated with the manual reconciliation and validation of data before it can be utilized for executive decision-making or advanced technology deployment. Following the release of these qualitative findings, several industry leaders and diginomica partners have provided on-the-record responses, confirming that the internal frustrations voiced by practitioners in private are reflective of a systemic crisis across the global enterprise landscape.
The Genesis and Methodology of the Enterprise Data Health Study
The Enterprise Data Health Study was conceived to bridge the gap between the optimistic marketing narratives of technology vendors and the lived realities of the professionals tasked with managing complex data architectures. By offering total anonymity to its 18 primary participants—all of whom hold senior roles in major enterprises—the study elicited a level of candor rarely seen in industry reports. The research focused on the qualitative experience of data management, exploring how silos, legacy infrastructure, and political maneuvering affect the utility of information.
The timeline of this study coincides with the aggressive push toward generative AI and agentic workflows in the enterprise sector. Throughout 2023 and into early 2024, organizations have invested billions into AI pilots, yet many have struggled to move these projects into full-scale production. The study serves as a diagnostic tool, identifying the underlying data pathologies that have stalled these high-stakes investments. By releasing the findings and inviting public responses from partners, the researchers have created a rare dialogue between the anonymous "buyers" and the public "sellers" of enterprise technology.
Quantifying the Verification Tax and the Burden of Manual Heroics
The most significant finding of the research is the "verification tax," which describes the 30% to 70% of professional time lost to the manual assembly, checking, and rebuilding of data sets. This tax is not a one-time fee but a compounding cost that erodes organizational agility. Raju Malhotra, Chief Product and Technology Officer at Certinia, noted that this pattern of "manual heroics" has become an absorbed operational cost that most organizations no longer question. According to Malhotra, when a majority of a team’s capacity is spent on verification, the ability to perform actual analysis is lost before the work even begins. This leads to a cycle where board preparation takes weeks, and parallel spreadsheets become the primary source of truth, further eroding confidence in centralized systems.
This erosion of confidence has direct implications for corporate governance. As the study highlights, the need for data to be "rebuilt" before it can be trusted suggests that data provenance and ownership are frequently undocumented or unclear. The result is a systemic trust deficit where decisiveness is replaced by revision rounds and skepticism.
Supporting Data: The AI Readiness Gap and Systemic Silos
The qualitative findings of the Enterprise Data Health Study are supported by broader industry metrics that suggest a widespread lack of data maturity. Gartner research indicates that approximately 63% of organizations currently lack the appropriate data management foundations required for successful AI implementation. Furthermore, analysts predict that up to 60% of AI projects will be abandoned by 2025 due to poor data quality and the absence of "AI-ready" data estates.
The diginomica study found that 94% of organizations continue to struggle with data silos. These silos are not merely technical barriers but are often the result of organizational fragmentation. Rupal Karia, Senior Vice President for North America, UKI, and MEA at Celonis, emphasized that the lack of a clear, trusted view of business operations is the primary reason AI initiatives struggle to move beyond the pilot phase. Karia argued that "Process Intelligence"—the ability to see interactions between disparate systems—is a prerequisite for AI functionality. Without this context, AI operates on incomplete information, rendering its outputs unreliable for enterprise-grade applications.
Official Responses: Reconciling Dedication with Dysfunction
The reactions from technology leaders suggest a consensus that the primary challenges facing data health are organizational rather than purely technical. Kathy Pham, Vice President of Open Technology and AI at Workday, provided a nuanced perspective on the "manual heroics" identified in the report. Drawing from her experience in healthcare data systems, Pham observed that fragmentation often arises from a place of professional dedication. When practitioners create "shadow data sets," they are frequently trying to fill gaps left by rigid legacy infrastructure to ensure immediate operational needs—such as patient care—are met.
However, Pham acknowledged that this commitment inadvertently creates the very silos that prevent broader organizational trust. Workday’s response emphasizes that moving toward an era of agentic AI requires honoring that dedication by providing a foundation that incorporates organizational context, rather than just raw data. This involves moving away from tools as a panacea and focusing on transparency, risk management frameworks, and "confidence signals" that allow users to understand the reliability of AI-generated outputs.
Cathy Mauzaize, EMEA President of ServiceNow, echoed the sentiment that complexity has been layered upon complexity over decades. She noted that the accumulation of systems has led to a breakdown in ownership. In her view, scaling AI responsibly is dependent on connecting governance and workflows into a single foundation. Without this, AI is not a solution but "expensive, and often unreliable, advice."
The Political Economy of Data Silos
One of the most provocative aspects of the study’s aftermath came from Rowan Tonkin, Chief Marketing Officer at Planful, who addressed a topic often avoided in corporate discourse: the deliberate maintenance of data silos. While many view silos as accidental byproducts of growth, Tonkin suggested that they are often maintained as strategic assets.
"If you own the data, you control the narrative," Tonkin remarked. He argued that some departments protect their silos because they provide negotiating leverage during budget cycles and can hide internal inefficiencies or risks. When data is fragmented, it is harder for executive leadership to hold specific functions accountable to a single version of the truth. This revelation suggests that the "trust collapse" mentioned in the report may be a design feature for some stakeholders rather than a flaw. Overcoming this requires more than a software update; it requires a fundamental realignment of corporate incentives and metrics to ensure that transparency is rewarded rather than punished.
Technical Strategies for Restoring Data Integrity
While the consensus leans toward organizational change, technical leaders also highlighted the necessity of a modernized data architecture. Fred Lherault, Field CTO for EMEA and Emerging Markets at Everpure, argued that businesses are suffering from a "systemic trust deficit" that cannot be fixed with another dashboard or point solution.
Lherault proposed a transition to an "Enterprise Data Cloud"—a single, intelligent data plane that automatically discovers, classifies, and governs data across its entire lifecycle. This technical approach aims to treat data as a managed asset rather than a byproduct of various applications. By automating governance and sovereignty policies, organizations can theoretically reduce the "verification tax" by making data trustworthy at the point of origin. However, Lherault maintained that such a technical shift must be accompanied by a willingness to collaborate on a "single source of truth," a goal that remains elusive for the 94% of organizations currently operating in silos.
Broader Impact and Future Implications for the Enterprise Market
The Enterprise Data Health Study serves as a critical reality check for an industry currently infatuated with the promise of Artificial Intelligence. The collective responses from Certinia, Celonis, Workday, ServiceNow, Acumatica, Planful, and Everpure validate the study’s core thesis: the "AI revolution" is currently being built on a foundation of sand.
The broader implications for the market are twofold. First, there is an increasing demand for "honesty as a service." Buyers are becoming weary of "performed confidence"—the tendency of vendors and practitioners to ignore broken data structures in favor of chasing the next technological trend. The study suggests that admitting to a trust collapse is a necessary starting point for any meaningful digital transformation.
Second, the redistribution of power within organizations is likely to become a central theme of the next decade. If the "verification tax" is to be eliminated, the strategic silos used for political leverage must be dismantled. This will require a different kind of conversation between C-suite executives and their IT departments—one that prioritizes data integrity and organizational transparency over the mere acquisition of new tools.
As organizations move forward, the "Enterprise Data Health Study" provides a framework for evaluating their readiness for the future. The findings suggest that those who succeed will be the ones who invest in trusted data foundations first and AI second. Until data is treated as a managed, governed, and transparent asset, the "verification tax" will continue to drain corporate productivity, and the full potential of enterprise AI will remain a pilot-phase ambition. The study and the resulting industry responses mark the beginning of a necessary, albeit difficult, transition toward a more honest and effective enterprise technology market.