A 48-inch by 120-inch sheet of A36 plate is not a box of screws. But if you look at how most ERP systems handle it, you would never know the difference.
SAP, Oracle, NetSuite, and every mid-market ERP on the planet were designed around a simple assumption: inventory is a thing with a SKU, a quantity, and a price. That works fine for consumer goods, electronics, even automotive parts. It falls apart the moment you walk into a steel service center.
Mismatch #1: Dimensional Inventory
Steel is not counted in pieces. It is described in dimensions: width, gauge, length, weight, and sometimes diameter. A single coil of 16-gauge CRC might weigh 40,000 pounds and yield dozens of different products after slitting. Generic ERPs have no native concept of dimensional inventory. They force service centers to create workarounds, custom fields stacked on custom fields, just to describe what they sell.
The result is a system that technically holds the data but makes it nearly impossible to search, filter, or report on inventory the way a steel person thinks about it. Try asking Oracle to show you all available HRC inventory between 0.060" and 0.075" gauge with a minimum width of 48 inches. You will be writing a custom report before lunch.
Mismatch #2: Pricing by CWT
Steel is priced per hundredweight (CWT), not per unit. The price depends on grade, gauge, coating, quantity, current market conditions, and the customer relationship. A single item might have a base price from the mill, a processing adder, a freight component, and a margin target that varies by account.
Generic ERPs understand "price per unit" and "quantity discounts." They do not understand CWT-based pricing with market-linked adjustments, grade extras, and gauge extras that change when the mills publish new price sheets. Service centers running SAP or NetSuite end up maintaining parallel pricing spreadsheets because the system cannot handle the math.
Mismatch #3: Heat Number Traceability
Every piece of steel traces back to a specific heat of metal produced at the mill. That heat number follows the material from receiving through inventory, processing, and shipment. It connects to the Mill Test Report (MTR) that certifies the chemical and mechanical properties. For DFARS-compliant orders, for nuclear, for pressure vessel applications, losing track of a heat number is not an inconvenience. It is a compliance failure.
Generic ERPs treat lot numbers as an afterthought, an optional field buried in a submenu. Steel service centers need heat number tracking woven into every transaction: receiving, inventory transfers, processing inputs and outputs, shipment documentation. When a customer calls asking for the MTR on a specific piece, the answer needs to come in seconds, not after a 15-minute search through filing cabinets.
Mismatch #4: Remnant Management
After processing, you have remnants. The drop from a shear cut. The edge trim from a slit. The short end of a bar after sawing. Some remnants are valuable and saleable. Others are scrap. The difference between the two can be thousands of dollars per month.
Generic ERPs have no concept of remnants. They understand finished goods and raw materials. A remnant is neither. It is a new inventory item, born from processing, with its own dimensions, weight, and potential value. Service centers that track remnants properly recover 3% to 8% of inventory value that would otherwise sit forgotten in a corner of the warehouse. Service centers running generic ERPs usually do not track remnants at all.
Mismatch #5: Processing as Manufacturing
A steel service center is not a factory, but it is not purely a distributor either. Slitting, shearing, sawing, plasma cutting, and blanking are manufacturing operations with yield rates, machine schedules, setup times, and scrap. Generic ERPs offer manufacturing modules designed for discrete or process manufacturing. Neither maps cleanly to steel processing.
The manufacturing module in a horizontal ERP expects a bill of materials, a routing, and a work order that produces a known output. Steel processing starts with a coil or plate and produces variable outputs depending on the customer order, the material condition, and the machine capabilities. Yield is calculated after the fact, not before. Scrap is an inherent part of every job, not an exception to be investigated.
The Compound Effect
Any one of these mismatches is manageable with workarounds. All five together create a compounding problem. The service center spends more time fighting the software than using it. Data quality degrades because workarounds introduce inconsistency. Reporting becomes unreliable. Decision-making slows down because nobody trusts the numbers.
The steel industry does $2.28 trillion in global revenue. The companies that move material through the supply chain, the service centers, deserve software that understands their business. Not software that forces them to pretend they are something they are not.