According to Valtech’s latest survey of over 100 senior manufacturing decision-makers, 2025 marks a turning point, the year where companies shift from AI aspiration to real-world AI application.
Data readiness: The foundation is still under construction
Despite strong enthusiasm, many manufacturers are still working to get the data basics right.
More than half (55%) of respondents say their data is only partially integrated. Silos remain a persistent issue. The complex and disconnected nature of data sources — from ecommerce platforms to legacy ERP systems and installed product bases — makes it difficult to unify and leverage data effectively.
Without clean, integrated and accessible data, ambitions like predictive maintenance or AI-driven personalization remain out of reach.
In other words, manufacturers can’t unlock the full value of AI in manufacturing until they fix their fragmented data ecosystems.
A shift toward practical AI in manufacturing
There’s good news, too. Manufacturers are becoming more strategic about how they implement AI in manufacturing.
Last year, the focus was on broad operational efficiency. In 2025, it’s about targeted high-impact use cases:
-
29% of respondents aim to improve decision-making with AI and machine learning.
-
17% want to create new value-added services using AI.
-
12% are pursuing a full transition to being data-driven AI-powered businesses — mostly among the largest organizations surveyed.
Instead of chasing hype, manufacturing companies are starting small and focusing on tangible business outcomes. This phased use-case-first approach reduces risk while enabling iterative learning.
The top barriers to AI in manufacturing
But significant hurdles remain. The three most cited barriers to AI in manufacturing are:
-
Lack of internal AI expertise
-
Poor data quality
-
Budget limitations
Security concerns, unclear governance structures and compliance challenges also create friction. For many organizations, there’s a clear desire to move forward, but uncertainty about how to begin.
Start small, think big
The report includes candid advice from digital leaders who have started their AI in manufacturing journeys. Here are a few takeaways:
-
Focus on data quality first. Without clean, reliable data, AI initiatives won’t gain traction.
-
Divide and conquer. Build a proof of concept in one business area, show results and then scale.
-
Hire the right people. A strong data team, particularly a skilled master data manager, is critical.
-
Be transparent. Acknowledge gaps in data and use them as a catalyst for future improvements.
The path forward: pragmatic innovation with AI in manufacturing
The overarching theme of this year’s findings is pragmatic progress. Manufacturers are embracing incremental ROI-driven innovation with AI in manufacturing.
Becoming a connected manufacturer — Valtech’s term for organizations delivering seamless data-enabled customer experiences — requires more than technology. It demands cultural change, cross-functional collaboration and a relentless focus on aligning data, tools and teams around business goals.
AI in manufacturing is no longer a question of if, but how. The winners will be those who clean up their data, hire smart, start small and scale with confidence.
