Most steel service centers in North America still run on-premise desktop software designed in the 1990s or early 2000s. A small minority have adopted cloud-based systems. AI adoption is in early innings. This is an honest assessment of where the industry stands and what the barriers to change look like.
The Current Technology Landscape
MSCI tracks over 2,000 steel service centers in the United States. Based on industry surveys, conversations with operators, and vendor market share data, the technology breakdown looks roughly like this:
About 55% to 65% of service centers run legacy on-premise systems. MetalTrax, INVEX (Invera), custom-built Access databases, and in some cases, actual DOS-based applications that have been running since the Clinton administration. These systems work. They are stable. The operators know them inside out. And they are functionally frozen in time.
About 20% to 25% run mid-market ERPs adapted for distribution. Epicor Kinetic, Sage, and NetSuite are the most common. These are cloud-capable (or cloud-native in NetSuite's case) and offer more modern interfaces. But they are horizontal tools that require significant customization to handle steel-specific workflows like dimensional inventory, remnant tracking, and heat number traceability.
About 10% to 15% use newer, industry-specific platforms. EOXS, WeSteel, and a handful of newer entrants are building cloud-native software designed specifically for steel distribution. This segment is growing fast but starts from a small base.
The remaining 5% to 10% run entirely on spreadsheets, paper, and phone calls. These are typically smaller operations with 5 to 15 employees where the owner manages inventory from memory and quotes from a calculator.
Cloud Adoption: Slower Than Every Other Industry
Gartner projects that 60% of new ERP deployments globally will be cloud-native and industry-specific by late 2026. Steel distribution is well behind that curve. The reasons are structural, not irrational.
Steel service centers handle sensitive financial data, customer pricing, and supplier contracts. Owners who have operated for decades with data on a server in their back office are understandably cautious about moving it to the cloud. The security question ("Where is my data and who can access it?") is legitimate, even if the answer (modern cloud infrastructure is more secure than an unpatched server in a warehouse office) is straightforward.
Implementation risk is the bigger concern. A failed ERP migration can cripple a service center for months. The stories are real: lost inventory records, corrupted customer data, six months of parallel systems, warehouse employees who refuse to use the new system. Every service center owner has heard at least one horror story. That collective memory creates inertia.
Cost perception also plays a role. Legacy systems are often fully amortized. The monthly software cost is zero (ignoring the substantial hidden costs of maintenance, lost productivity, and workarounds). A cloud platform with per-user-per-month pricing looks like a new expense, even when the total cost of ownership is lower.
Where AI Adoption Actually Stands
Despite the marketing noise, AI adoption in steel distribution is in very early stages. Most of what vendors call "AI" today falls into three categories:
Basic analytics. Dashboards that show trends, charts, and KPIs. Useful, but not artificial intelligence. This is reporting with a better label.
Rule-based automation. Systems that automatically reorder when inventory drops below a threshold, or flag quotes below a margin target. Again, useful, but these are business rules, not AI. They have existed for decades under different names.
Genuine AI applications. Pricing recommendations based on market data and customer behavior analysis. Quote generation from natural language descriptions. Demand forecasting using machine learning models trained on historical order patterns. These exist but are deployed at fewer than 5% of service centers.
The gap between marketing and reality is wide. But the trajectory is clear. The service centers deploying genuine AI today are seeing measurable advantages in quote speed, pricing accuracy, and inventory optimization. As these advantages become visible to the broader market, adoption will accelerate.
The Barriers to Change
Cost and implementation risk are the obvious barriers. But three less-discussed factors matter more:
Workforce readiness. Many service center employees have used the same system for 15 to 20 years. The prospect of learning a new platform is genuinely daunting, especially for workers who are not digital natives. Training is not a one-week event. It is a multi-month process that requires patience, support, and a willingness to tolerate temporary productivity dips.
Data quality. A new system is only as good as the data migrated into it. Many legacy systems have years of accumulated data inconsistencies: duplicate customer records, inconsistent product descriptions, incomplete inventory histories. Cleaning this data before migration is tedious but essential work.
Leadership bandwidth. Service center owners and GMs are busy running their businesses. Evaluating software, managing an implementation, and driving organizational change requires time and attention that competes with daily operations. This is why many transformations stall: not because the technology fails, but because leadership cannot sustain focus on the project while also running the business.
What the Next Three Years Look Like
Cloud adoption in steel distribution will follow the same S-curve it has followed in every other industry: slow start, rapid acceleration, eventual ubiquity. The inflection point is approaching.
Three forces are converging. Workforce turnover is bringing in employees who expect modern tools. Competitive pressure from consolidated players is forcing smaller operations to find efficiency gains. And the software options available to the industry are significantly better than they were even three years ago.
By 2028, we expect the majority of new software deployments in steel distribution will be cloud-native and industry-specific. The service centers that move first will have a meaningful head start in data quality, operational efficiency, and workforce capability. The ones that wait will face a more urgent and more expensive transition later.
The question is not whether digital transformation happens. It is whether you lead it or react to it.