Trade concerns in 2026 present a real paradox for manufacturers: many are operating some of the most sophisticated production environments in the world, yet are likely managing costs and margins the same way they did a decade ago, with spreadsheets and manual reconciliation.

Although the industry landscape remains dominated by super-sized manufacturers, there's a real opportunity for small and mid-size businesses to level the playing field by adopting AI innovations. But here's the catch: in a world of rapidly changing supply chains and regulations, maintaining margins feels like guesswork, which goes against everything engineering-driven manufacturing stands for.

The numbers tell the story. Fewer than one-third of manufacturers have deployed AI at scale, despite mounting margin pressure. Unmanaged change orders, rework, and production variances erode 5-10% of margins. And while AI can help interpret constantly changing laws and fees across supply chains, most manufacturers are focusing on integrating AI where they can actually control it: their data and operations.

The Real Problem: Data Trapped in Silos

The most cited challenge? Data locked in systems that can't talk to each other. Engineering works in CAD and PLM. The shop floor runs on MES, SCADA, and MRP. Finance lives in the ERP. Quality control operates in its own database.

Each system holds critical pieces of the profitability puzzle, but extracting insight requires significant labor from highly paid engineers to manually aggregate and reconcile everything. By the time finance reconciles actual costs against estimates, production is often complete, and the margin damage is done. Material overruns surface weeks after procurement. Rework gets buried in production reports. Leadership is flying blind.

So, where should a mid-size manufacturer actually apply AI?

Aligning Data: Where AI Really Excels

Manufacturing has been slow to adopt AI, and for understandable reasons. Unlike consumer-facing industries where AI can be deployed in isolated use cases, manufacturing requires AI to work across deeply integrated, mission-critical systems. The stakes are higher, the technical complexity is greater, and the tolerance for error is lower.

But here's what's changing: AI is no longer just about automation or predictive maintenance. The real breakthrough is using AI to unify fragmented data sources and deliver real-time visibility across your entire operation.

Here's a practical example: That beam you designed in CAD with specified structural strength? AI can automatically translate it into a BOM that fits your PLM requirements, align those requirements to your ordering part number in the ERP, and connect shop floor quality reports back to the original design specifications.

This eliminates the hours engineers and analysts spend on tedious table comparisons. Work that's both time-consuming and error-prone. Instead, they focus on the critical decision points that actually add value.

You're not replacing your existing systems. You're creating an intelligence layer that translates between them so leadership can see the complete picture while there's still time to act.

Where Manufacturers Are Seeing Results

The manufacturers achieving the fastest ROI aren't trying to transform everything at once. They're solving one specific pain point completely before moving on to the next. Here's where manufacturers are seeing wins:

• Production Variance Analysis: AI identifies patterns in cost overruns across MES, ERP, and quality systems, helping operations teams address root causes before they compound.
• Cross-System BOM Management: AI automatically reconciles bills of materials across CAD, PLM, and ERP, eliminating manual errors and freeing engineers for higher-value work.
• Real-Time Margin Visibility: AI connects engineering changes, material costs, labor hours, and quality issues to show actual margins versus plan, whether you're tracking job costs, production runs, or work orders.
• Quality Control Integration: Inspection data flows automatically into cost tracking and production planning, making rework costs and schedule impacts immediately visible.

Start Small and Scale What Works

Here's the good news: manufacturers are already experts at continuous improvement. These businesses were built on making processes more efficient. It isn’t about reinventing the entire operation overnight, it is about improving it systematically, piece by piece.

Apply that same discipline to AI adoption. Don't try to replace an entire process. Fix one data reconciliation workflow, one cost tracking gap, one quality reporting bottleneck, one planning cycle. Prove the value in a contained environment, then scale what works.

Looking Ahead

The manufacturers who move first on AI-enabled visibility will protect margins and gain a competitive advantage with the ability to compound over time. While competitors are still reconciling last month's variances in spreadsheets, early adopters will be making real-time decisions on production mix, capacity allocation, and pricing based on actual performance data.

In an uncertain world, maintaining margins shouldn't feel like guesswork. The question for 2026 isn't whether AI will transform manufacturing operations, it's whether you'll lead that transformation or spend years catching up.

Johnny Than is the Founder & CEO of Appficiency, a global IT consulting firm providing enterprise digital services to manufacturing firms. With 18 years of manufacturing technology experience, including 7 years at Oracle helping transform NetSuite's manufacturing capabilities from distribution-focused to a comprehensive manufacturing platform, Johnny has implemented systems for Fortune 500 manufacturers and global mid-market to enterprise companies. Appficiency's AI Enablement Consulting (AIEC) service helps manufacturers identify high-impact AI use cases and deploy solutions that unify fragmented systems for real-time operational visibility. For more information, please visit: https://appficiency.com/.