BARC x Actian Research: Why Data Products are Key to Scaling AI
Zusammenfassung
- Organizations with mature data products are far more likely to scale AI, with 85% running multiple AI projects in production.
- Agentic AI adoption is significantly higher in these organizations, with 77% deploying autonomous systems versus limited adoption elsewhere.
- Data products provide structured, governed, and reusable data foundations required for reliable production AI.
- Strong data ownership, quality standards, and contracts enable safer and more scalable AI operations.
Erkenntnisse aus der weltweiten Studie von BARC und Actian aus dem Jahr 2026 unter mehr als 300 Führungskräften im Bereich Unternehmensdaten
Unternehmen aus allen Branchen investieren massiv in KI. Dennoch haben viele Initiativen nach wie vor Schwierigkeiten, über das Experimentierstadium hinauszukommen und einen nachhaltigen geschäftlichen Nutzen zu erzielen.
Um besser zu verstehen, was skalierbare KI-Initiativen von solchen unterscheidet, die ins Stocken geraten, hat sich Actian mit BARC, einem weltweit führenden Analyseunternehmen für Daten und Analytik, um eine weltweite Studie unter führenden Datenexperten in Unternehmen.
Auf der Grundlage der Erkenntnisse von mehr als 300 Befragten aus verschiedenen Branchen und Regionen untersucht die Studie, wie Unternehmen Datenprodukte und Datenverträge einführen und umsetzen und wie sich diese Ansätze auf den Erfolg von KI-Initiativen auswirken.
One of the clearest insights emerging from the research is the strong relationship between data product adoption and the ability to scale AI into production.
BARC × Actian – Globaler Forschungsbericht (2026)
Organizations With Mature Data Products Run More AI in Production
Many organizations today are experimenting with AI. However, moving AI initiatives from experimentation into production remains difficult.
The research reveals a clear pattern: organizations that deploy data products at scale run significantly more AI systems in production.
Among organizations that deploy data products company-wide, 85% report running multiple AI projects in production. By contrast, organizations that are not using—or only experimenting with—data products are far less likely to operationalize AI beyond pilot initiatives.
This finding highlights the role data products play in enabling reliable AI systems. By organizing data assets with clear ownership, defined quality expectations, and reusable interfaces, organizations create the structured data foundation required to support production AI.
Correlation between AI project adoption and data product usage – © BARC 2026
Agentic AI Adoption is Highest in Organizations With Mature Data Products
The same pattern appears when looking at agentic or autonomous AI systems.
Agentic AI systems—capable of acting autonomously, orchestrating workflows, and interacting with multiple data sources—require highly reliable and well-governed data environments to operate safely.
The research shows that 77% of organizations with company-wide data products already have at least one agentic or autonomous AI system in limited or full production. The inverse also holds true: 77% of organizations that are only experimenting with data products—or not using them at all—have not yet deployed agentic AI in production.
This finding reinforces an important insight: as AI systems become more autonomous and integrated into operational workflows, the quality, governance, and reliability of underlying data become even more critical. Data products provide the structure needed to manage these dependencies and support the safe deployment of increasingly complex AI capabilities.
Correlation between agentic AI adoption and data product usage – © BARC 2026
Why Data Products Enable AI at Scale
The correlation observed in the research reflects the role that data products play in modern data architectures.
By organizing data assets as products—with defined ownership, quality standards, and clear interfaces—often reinforced through data contracts—organizations create a reliable foundation for analytics and AI systems.
This approach helps ensure that AI systems rely on trusted, well-managed data, reducing the risks associated with inconsistent inputs, unclear responsibilities, or fragile data pipelines.
As a result, data products are emerging as a critical capability for organizations seeking to operationalize AI at scale.
Den vollständigen Forschungsbericht herunterladen
Dieser Artikel stellt einige wichtige Erkenntnisse aus der weltweiten Studie von BARC und Actian vor.
Der vollständige Bericht, Datenprodukte und Datenverträge im Jahr 2026: Die Grundlage für den Erfolg der KI, untersucht, wie Unternehmen verschiedener Branchen und Regionen Datenprodukte und Datenverträge einführen und operationalisieren – und wie diese Praktiken die KI-Reife, die Governance und die Ergebnisse in der Praxis beeinflussen.
Laden Sie den vollständigen Bericht herunter, um die Ergebnisse zu entdecken und zu erfahren, wie führende Unternehmen KI mithilfe von Datenprodukten skalieren.
Vollständigen Bericht herunterladen
Bleiben Sie in Verbindung
Datenanalysen, die Ihnen geliefert werden.
(z. B. sales@..., support@...)
