Industrial manufacturers often believe they are different from consumer businesses because many of their products are engineered or sold into long customer relationships. That is true, but it can also become an excuse for carrying too much complexity for too long. Plants accumulate slow-moving parts, lightly differentiated variants, customer-specific options that no longer matter, and legacy items preserved mainly because nobody wants to disrupt the installed base. The result is a catalog that is broader than the market truly requires and a supply chain that is more expensive and less responsive than it should be.
SKU rationalization in industrial manufacturing is therefore less about reducing visible choice and more about restoring discipline to a portfolio that has grown through years of exceptions, engineering carryover, and service commitments. A low-volume item may still support a machine family, a key OEM account, or a service promise that the company cannot ignore. Yet many industrial businesses still fund a tail of complexity that creates little incremental value. This chapter examines where that complexity comes from, how to distinguish standard, configured, and custom products, how to rationalize while protecting service levels and installed-base support, and where industrial manufacturers usually capture the most value first.
28.1 Complexity From Parts, Variants, and Legacy Items
Industrial portfolios become complex in a way that is often less visible than in consumer sectors. The product line may look coherent from the outside, but underneath it sits a long list of part numbers, assemblies, options, fittings, material variants, and service components. A single machine family can generate hundreds or thousands of active SKUs once the underlying components and configurations are counted. This is why many industrial companies underestimate the problem. They see one platform. Operations, procurement, and service see item-level complexity.
Parts proliferation: the growth of too many unique components, subassemblies, and service items across a product family. This often starts innocently. An engineering team creates a new bracket, connector, housing, coating, or software variant for a customer request or design refresh. Instead of reusing an existing part, the organization introduces another code. Over time, the bill of material becomes less modular, more fragmented, and harder to source or support economically.
Variant proliferation: the buildup of too many standard options or semi-standard versions of what is supposed to be a manageable industrial range. This is common in pumps, valves, motors, drives, electrical equipment, compressors, tooling, and packaging equipment, where customers want the product to fit their application precisely. Some variation is necessary. The problem begins when the business keeps adding frame sizes, material grades, pressure ranges, mounting styles, voltage versions, or software options without asking whether each one still earns its place.
Legacy-item burden: the cost and complexity created by older products, spare parts, and historical configurations that remain active long after their primary commercial role has faded. In industrial businesses, legacy items are often protected because of service commitments, installed base, or fear of customer dissatisfaction. That protection is sometimes justified. Often it is far broader than it needs to be. The company may still be buying or stocking components for a very old machine family when actual demand is sporadic and alternative service strategies are available.
Complexity is often worsened by acquisitions. The same functional need may be served by overlapping brands, overlapping part architectures, or duplicate legacy ranges. Without deliberate rationalization, the company carries historical complexity in parallel rather than converging toward a cleaner future-state architecture. Procurement manages more suppliers and smaller buys, and planning manages more low-signal demand patterns.
28.2 Distinguishing Standard, Configured, and Custom Products
One of the biggest mistakes in industrial rationalization is to treat all product variation as though it belongs in the same portfolio logic. It does not. Industrial companies usually operate across three broad categories of offer: truly standard products, configured products built from defined option sets, and custom or engineered-to-order products. Rationalization fails when the business does not distinguish clearly among these categories.
Standard product: an item that should be sold, planned, sourced, and produced repeatedly with limited or no variation. Standard products are where the business should seek the highest degree of simplification, reuse, and volume concentration. If too much complexity exists here, optionality has usually drifted into what should be the core catalog.
Configured product: an item assembled or specified from a controlled set of modules, options, or rules. This category is common in industrial markets because customers need some adaptation, but not a completely unique design. Good configuration logic can be a major advantage. It allows the business to meet application requirements while still limiting internal complexity. Poor configuration logic creates an enormous menu of options, many of them lightly used, that behave operationally like custom products while still being treated as standard catalog items.
Custom product: an engineered or heavily modified item created for a specific customer need that cannot reasonably be served through the standard or configured range. Custom products are not automatically bad business. In some industrial segments, they are central to strategy. The problem arises when the company blurs the line between true custom work and disguised catalog sprawl. A product may be labeled standard simply because it has its own SKU, even though demand is one-off and the company should really manage it through an engineered-to-order process.
Distinguishing among these categories changes the rationalization agenda. Standard products should be reviewed for duplication, low-volume variants, and unnecessary part diversity. Configured products should be reviewed for option-set discipline. Which options are frequently chosen and strategically important? Which options are rarely selected and create a disproportionate burden in planning, sourcing, or assembly? Custom products should be reviewed for business model fit and prevented from quietly reentering the standard catalog as permanent complexity.
The goal is not to force every industrial business into a tiny catalog. It is to protect standardization where it belongs, control configuration where it creates customer value, and isolate custom work so it does not contaminate the standard portfolio with permanent low-volume burden.
28.3 Rationalizing While Protecting Service Levels and Installed Base Support
Industrial rationalization becomes more sensitive when it touches the installed base. In many sectors, a large share of customer trust comes not from the initial sale alone but from the manufacturer’s ability to support equipment in operation for years or even decades. Spare parts, replacement components, software updates, consumables, and service documentation are therefore central to the value proposition. This is why industrial companies often preserve legacy SKUs far longer than consumer businesses would tolerate.
Installed base support: the obligation, explicit or implicit, to maintain serviceability for equipment already in the field. This support may be contractual, commercially strategic, or simply necessary to preserve brand credibility. Rationalization has to respect that. A company cannot simply delete a low-volume spare part because the revenue is small if doing so would strand customers, jeopardize safety, or damage future equipment sales.
But protecting the installed base does not mean preserving every legacy SKU in its current form forever. The smarter question is how the support obligation can be met with less complexity. In some cases, multiple old parts can be superseded by one redesigned replacement. In others, a stocked item can move to make-to-order or service-center stocking rather than network-wide inventory. Some items can be supported through repair kits, refurbishment, aftermarket conversions, or qualified substitutes rather than exact historical duplicates.
Service-level protection: the discipline of preserving uptime, field responsiveness, and maintenance confidence while simplifying the portfolio underneath. This usually requires segmentation. Not every spare part deserves the same treatment. Items linked to critical uptime, safety, or major installed populations may require higher stocking and slower exit. Items with tiny installed populations, long lead-time tolerance, or viable substitutes can often be handled differently. Rationalization should therefore classify service items by criticality, installed-base exposure, failure consequence, and replacement path rather than by sales volume alone.
It is also important to separate customer expectation from actual obligation. Many industrial teams assume they must support every legacy item indefinitely because “customers expect it.” Sometimes that is true. Sometimes the actual contract or warranty is narrower. A part may need to remain available for a defined period, after which the business can move customers to a retrofit, upgrade kit, or next-generation platform. These transitions are easier when planned proactively rather than triggered by shortage. Field-service and technical-support input is especially important because they understand which items are genuinely critical and which substitutions are workable.
28.4 Practical Priorities for Industrial Portfolios
Industrial manufacturers usually create the most value when they focus first on the forms of complexity that are both low in customer distinctiveness and high in internal burden. The first priority is often component standardization. If multiple products use slightly different fasteners, brackets, boards, housings, or subassemblies without a strong technical reason, the company should converge them. Component-level simplification often creates larger savings than end-SKU pruning because it improves sourcing leverage, reduces bill-of-material diversity, and simplifies service stocking at the same time.
The second priority is to clean the semi-standard tail. These are the catalog items and configured variants that remain technically available but have weak demand, high supply-chain burden, or no longer reflect a strategic need. In many industrial portfolios, this includes minor option combinations, special finishes, unusual voltage or material variants, and old platform derivatives that still appear in the catalog but generate little healthy business. These are often easier to rationalize than true service parts because the customer need can frequently be met through the current standard or configured range.
The third priority is to tighten configuration rules. Many industrial companies do not need fewer platforms as much as they need fewer allowed combinations inside each platform. That means identifying the options customers truly select, the combinations that drive most of the business, and the rare combinations that create complexity out of proportion to their value. A disciplined configuration model can preserve application fit while sharply reducing quoting complexity, lead-time variability, and low-volume supply exposure.
The fourth priority is to redesign legacy support economics. Rather than treating all spare parts as permanently stocked SKUs, the company should segment them by criticality and installed-base value. Some should remain core service items. Others should move to make-to-order, repair, refurbishment, or superseded replacement logic. This is often where the largest working-capital gains appear.
A practical sequence usually follows a clear pattern. First, identify and remove obsolete or inactive items and duplicate acquired parts. Second, standardize components and simplify overlapping catalog variants. Third, tighten configuration logic so custom behavior stops leaking into the standard range. Fourth, redesign installed-base support around criticality and replacement strategy rather than around historical item preservation. The companies that do this well make the portfolio easier to engineer, source, produce, service, and navigate. In industrial manufacturing, that is often the real prize: not just fewer SKUs, but a business that can support complexity where it matters and eliminate it where it no longer pays.
