Closing the AI Execution Gap: Why Most Businesses Struggle to Scale and How Infor is Changing the Game

The dream of artificial intelligence in the enterprise has never been bigger, but the reality of actually deploying it at scale is proving to be a much steeper climb than many anticipated. While the buzz around generative AI continues to dominate headlines, a new study from Infor suggests that a significant number of organizations are hitting a wall. According to the newly released Infor Enterprise AI Adoption Impact Index, more than half of businesses are currently struggling to move their AI initiatives beyond the pilot phase and into full-scale production.

To address this persistent execution gap, Infor has introduced a series of major updates to its Infor Velocity Suite and announced the limited availability of its Infor Agentic Orchestrator. These tools are specifically designed to provide the industry-specific precision and governed execution that generic AI models often lack. By focusing on deep process intelligence rather than one-size-fits-all automation, Infor aims to help companies finally realize the ROI they’ve been promised.

The Paradox of AI Confidence

The research, which surveyed 1,000 business decision-makers across the United States, United Kingdom, Germany, and France, reveals a fascinating contradiction. On one hand, global confidence is high: 80% of leaders believe their organizations have the internal capability to manage an AI implementation. However, the data shows that 49% of these organizations are still stuck in the early stages of deployment. They are either running small-scale pilots or struggling with partial rollouts that haven't yet reached the broader enterprise.

Why the disconnect? The study highlights several structural barriers that act as a handbrake on progress. Data security, sovereignty, and compliance concerns were cited by 36% of respondents as a major obstacle. Meanwhile, 25% pointed to a lack of internal AI talent, and 23% admitted they are still struggling with an unclear ROI. It turns out that while the ambition is there, the infrastructure and expertise often are not.

Moving Beyond Generic AI

Kevin Samuelson, CEO of Infor, believes the problem lies in the lack of industry context. Generic AI models can write emails or summarize documents, but they don't understand the nuanced workflows of a healthcare purchasing agent versus a discrete manufacturer. Samuelson notes that Infor’s agentic AI isn’t just a "bolt-on" feature; it’s the result of two decades of building multi-tenant architectures and deep process intelligence.

This specificity is critical for defining and delivering measurable outcomes. As Mickey North Rizza, Group Vice-President of Enterprise Software for IDC, puts it, Infor’s clients are finding sustained economic value precisely because they have a clear path to becoming an "agentic enterprise."

Infor Velocity Suite: Bridging the Talent Gap

Recognizing that one in four businesses lacks the talent to build AI from scratch, Infor has expanded its Velocity Suite. This package is designed to be the simplest path for customers to see immediate value. It now includes a library of Infor Industry AI Agents that are pre-tailored to specific business impacts.

One of the standout additions is a new add-on for the Infor Warehouse Management System (WMS). This tool uses machine learning for "pick path optimization," guiding warehouse workers along the most efficient routes. The results are tangible: early adopters like Coram International have seen travel distances drop by 25%, leading to 15% faster picking. This kind of practical application allows warehouses to operate more efficiently without necessarily needing to hire more temporary staff during peak seasons.

The Agentic Orchestrator: Governance and Trust

Autonomous AI is a top priority for 32% of business leaders, but autonomy requires trust. This is where the Infor Agentic Orchestrator comes in. Currently in limited availability, this platform serves as the transparent infrastructure layer that coordinates different AI agents into unified workflows. It focuses on three main areas: visibility into agent performance, strict governance to ensure AI stays within its defined boundaries, and control mechanisms that allow humans to step in when needed.

For companies like AMADA America, this means a shift from manual searching to proactive intelligence. Zoaib Saifuddin, General Manager of IT at AMADA, explains that instead of service engineers hunting for answers, the intelligence now comes directly to them, streamlining the entire maintenance process.

Detailed Findings from the Impact Index

The Enterprise AI Adoption Impact Index provided a deep dive into several critical trends across sectors like Retail, Food and Beverage, and Industrial Manufacturing:

1. High Confidence vs. Structural Barriers: While leaders feel ready, the lack of operational infrastructure is stalling progress.
2. Data and Agent Distrust: Concerns over how data is handled and whether AI agents can be trusted with sensitive tasks are slowing the transition from deployment to actual value.
3. The Wish List: Decision-makers are prioritizing security, specialized agents, and a better fit for their specific industry over general-purpose AI tools.

Ultimately, the goal is rapid growth without a linear increase in resource consumption. Jamarl Scace, Digital and IT Lead at Kattsafe, noted that by starting with automation for customer order entry through the Velocity Suite, they found a simple, practical path to AI that freed their team for high-value engagement. As the enterprise landscape continues to evolve, the winners won't just be the companies with the most AI, but those with the most targeted AI.