Industrial equipment manufacturing: modular platforms and the slow bleed of option sprawl

The platform decision precedes every order

Deciding what is common across a family and what varies is the highest-leverage engineering decision in this sector, and it is made before any of the orders it will serve exist. Get it right and a plant assembles hundreds of distinguishable machines from a modest set of stocked modules. Get it wrong and every order becomes a small engineering job. The useful test is whether variation can be applied late, after the expensive and long-lead content has been committed. Anything that forces a different frame, a different mounting pattern or a different control architecture is not really an option; it is a second product wearing the same name.

Option sprawl is a leak, not a leap

No single special ever looks expensive. Each one carries a drawing to maintain, a part number to stock, an assembly instruction to update, spares to hold for a decade, and a small probability of a build error. Multiply that across years of sales concessions and a plant ends up with a catalogue nobody can price accurately and a stockroom full of parts fitted to a handful of machines. The cure is arithmetic rather than policy: attach the true lifecycle cost of a variant to the order that requested it, review the option list on a schedule, and actively retire configurations that have not sold recently.

Distributors are the customer even when end users are the market

In much of this sector, the immediate buyer is a distributor who stocks, sells, installs and services. That changes what the factory is optimising. Distributors want predictable delivery, discount structures they can plan on, product that does not embarrass them in service, and protection from the manufacturer selling around them. End users, meanwhile, want the specification. Manufacturers who chase large direct accounts while relying on distribution for volume tend to damage both. The workable pattern is an explicit account split, honoured consistently, with defined technical support the distributor cannot supply on its own.

The installed base is an asset only if it is recorded

Aftermarket revenue depends on knowing what was actually shipped: which motor, which control revision, which seal kit, under which serial number. Plenty of manufacturers ship configured machines and keep only the sales order, then spend a decade guessing when service calls arrive. Serialised as-built configuration, retained and searchable, converts a stream of one-off machines into a parts annuity. It also makes engineering changes manageable, because the population affected by a defective component batch becomes a query rather than an investigation. This record is cheap to create at build time and expensive to reconstruct later.

Where lead time really goes

Customers assume delivery time reflects assembly time. It rarely does. Most of the quoted period is queue: waiting for a purchased component family with its own lead time, waiting for paint or plating capacity, waiting for a test bay, waiting for documentation. Mapping where a typical order actually sits, hour by hour, usually reveals that the assembly bay accounts for a small fraction of the elapsed time. That analysis is what allows a manufacturer to shorten delivery without adding people, usually by holding a strategic buffer of the specific long-lead items that appear in most configurations.

Frequently asked questions

How wide should a configurable product family be?

Wide enough to cover the applications that generate most of the volume, and no wider. A practical rule is that a variant earns its place if it sells regularly, applies late in the build, and does not force changes to structure or control architecture. Options that fail those tests should be quoted as engineered specials at engineered prices, which usually causes most of them to disappear from the pipeline without any policy argument being needed.

Is it better to sell direct or through distribution?

Distribution buys market coverage, local stock, installation capability and credit management that a manufacturer would otherwise fund itself. Direct selling captures more margin and keeps the customer relationship. The mistake is doing both without rules, because a distributor who loses a deal to the manufacturer stops investing in the brand. Manufacturers who succeed with hybrid models define named account boundaries in writing, keep pricing consistent, and pay the channel for work it genuinely performs.

Why are spare parts priced so much higher than the same content in a new machine?

A new machine is priced against competitive alternatives in an open tender. A spare part is priced against the cost of the machine being stopped, and the customer usually has no alternative source within an acceptable time. Beyond that, spares carry real cost the machine price does not: long-term inventory holding for parts that may never sell, obsolescence, packaging, and a distribution network sized for urgency rather than efficiency.