According to a recent Censuswide research project, 41% of construction and engineering (C&E) projects are currently overrun on both budgets and target completion dates. Furthermore, more than one in two (54%) projects are failing to meet their anticipated margins. For an industry historically characterized by thin margins, these inefficiencies are dangerous. Small disruptions — whether from inflation, supply chain issues, labor shortages, or extreme weather — can quickly turn small profits into losses, and in some cases, threaten the existence of the business.
One area receiving increased attention is the use of AI agents to help reduce the back-office pressures of keeping major projects on time and on budget. Censuswide research revealed that delivering projects quickly (35%) and on budget (35%) are the most important differentiators for construction and engineering businesses. For progress to be made on this front, firms must first replace chaotic data silos with a unified Enterprise Resource Planning (ERP) backbone to create the “single source of truth” required for Industrial AI to build the foundations for industry change.
AI adoption is accelerating
Facing these headwinds, the industry needs a strong technological foothold. In its “2026 Engineering and Construction Industry Outlook,” Deloitte reveals “firms are increasingly leveraging advanced digital tools to boost productivity, protect margins, and adapt to rapidly changing market conditions. Leading organizations are deploying technologies such as AI-driven analytics, real-time project management platforms, and connected job-site solutions to streamline operations, enhance decision-making, and stand out in a competitive landscape.”
IFS Research into the “Invisible Revolution” of AI in industrial settings reveals that 91% of C&E firms expect to increase their AI investment in 2025 and beyond — a figure that ties for the highest among all industries surveyed. The optimism is backed by early results: 89% of early adopters already report profitability gains, and 83% cite improved product and service quality.
The ERP challenge
However, a critical gap remains between ambition and execution. While valuable, bricklaying robots, automated design tools, or smart glasses for site safety, are isolated tools and only add to the problems because they do not address the fundamental business process issues dragging down profitability. Moreover, many C&E organizations don’t have the digital foundation in place to commence AI projects.
To exploit the true benefits of AI, construction companies must first address their data landscape. Currently, many firms operate on fragmented systems, trying to manage complex global projects using a complex web of spreadsheets locking away key data in a siloed manner. But AI models are only as effective as the data that feeds them. If a company lacks a single version of the truth regarding projects, materials, workforces, and subcontractors, AI cannot function effectively. A strong ERP system is not just a system of record — it is the prerequisite foundation for AI.
The industry recognizes this need for structural modernization. The Censuswide study found that more than 63% of C&E businesses state they are in looking to replace their existing ERP systems within the next two years. Once that digital foundation is in place, organizations can begin applying AI to specific operational challenges where it can deliver measurable value.
Where AI can deliver value
Once the data foundation is in place, the most transformative initial AI use cases likely won’t be found on the physical job site, but in the back office. These “unsexy” processes are the gears that drive productivity and labor usage, and there is a new breed of digital workers willing and able to lend support.
Unlike traditional automation, robotics, and generative AI, digital workers powered by Agentic AI manage complex workflows, make context-driven decisions, and continuously improve operational performance while working right alongside employees to reduce manual tasks and amplify their impact. Whether embedded in core ERP or even running alongside legacy systems, these digital workers provide immediate ROI and operational impact. Three operational areas where AI agents could provide measurable value include:
The most critical control process for any construction business is project cost control, predicting where a project will end financially. Currently, this is an extremely manual process done month-to-month, where many staff members spend weeks manually compiling data to re-forecast costs and revenue. AI agents can revolutionize this by continuously analyzing real-time data to predict end-of-project outcomes autonomously.
Here’s where AI can make a huge difference. Instead of looking in the rear-view mirror once a month, AI can provide a forward-looking prediction, allowing project managers to mitigate margin fade before it happens.
2. Smart spend authorization
Controlling spending often involves a bureaucratic nightmare where a purchase order or subcontract might require 10 different signatures to get out the door. This often leads to “rubber stamping,” where busy managers sign off on requests without proper review just to keep the project moving. AI agents can introduce rigorous, non-emotional logic to this process.
But this is to undervalue the power of AI. An agent can review a request against budgets, contracts, and project requirements instantly, ensuring strict compliance without the bottleneck of human fatigue or time constraints.
3. Procure-to-pay efficiency
Construction is a high-volume and high-value procurement environment — the perfect scenario for AI. AI agents can drastically reduce the friction in the procure-to-pay cycle by automating invoice matching and payment processing. This not only reduces administrative overhead but also ensures better supply chain relationships and cash flow management.
The path forward
AI can only deliver real value in construction when built on a strong ERP foundation. Fragmented systems and siloed spreadsheets prevent AI from functioning effectively, making ERP modernization a critical first step. With 63% of firms planning ERP upgrades within two years, the industry is moving toward consistent processes and a unified data backbone.
The future isn’t just about robots on-site — it’s about intelligent AI agents managing business operations. By securing a robust ERP core and layering Industrial AI, construction firms can shift from fragile spreadsheets to a resilient, AI-driven model that enhances efficiency, protects margins, and ensures control in a volatile market. As AI capabilities continue to evolve, organizations with connected business systems and reliable data will be better positioned to evaluate where the technology can provide measurable operational value.
