Last-Time Buys and Long Tails: Pricing Component Obsolescence Into UK Deep-Tech Valuations

Most diligence on a UK deep-tech hardware company concentrates on the things founders are proud of: the prototype, the patents, the pilot with a named customer. The bill of materials rarely gets the same attention. Yet a BOM is a slow-moving liability as much as it is a recipe, and the parts inside it have a habit of disappearing from production long before the product they sit in reaches the end of its service life.

The pattern is familiar to anyone who has run a hardware programme. A design is frozen around components that were available, affordable and well documented at the time. Two or three years later, the manufacturer issues an end-of-life notice, the last-time buy is placed under pressure, and the engineering team discovers that the replacement part is a different package, a different thermal profile and a different firmware interface. What follows is not a procurement problem. It is a redesign programme, and it lands at precisely the moment the company is trying to scale.

Where the exposure accumulates

The risk is not evenly distributed. Three areas deserve particular scrutiny in the current UK market.

Robotics and automation. A robot is a system of subsystems sourced from an unusually wide supplier base: motion controllers, servo drives, encoders, safety relays, vision modules. Many of these are produced in modest volumes by specialist firms, and a specialist firm with a thin order book is a candidate for discontinuation, acquisition or quiet withdrawal from a product line. A collaborative robot with a ten-year service expectation can easily contain a dozen parts whose manufacturers are planning a five-year horizon. Field-service obligations then become an unhedged commitment.

Materials processing and pilot plant. Equipment in this category often combines modern control hardware with legacy industrial interfaces and long-lived instrumentation. Pilot lines are frequently built from a mixture of new and refurbished kit, and the refurbished element introduces components that were already at end-of-life when they were installed. When a pilot line is the evidentiary basis for a scale-up investment, the question of whether it can be kept running for another three years is not academic.

Laboratory instrumentation. UK labs are full of instruments that are excellent at their job and no longer supported by their manufacturer. Detectors, sources, vacuum components and bespoke electronics can fall outside any vendor's roadmap. For a company whose scientific claims rest on data from a specific instrument, the loss of that instrument is a loss of continuity in the evidence base, not simply a capital expense.

The economics are longer than the warranty

The instinct is to treat obsolescence as a maintenance line item. It is better understood as a duration mismatch. Warranty periods, service contracts and customer expectations all run on different clocks, and the parts clock is usually the shortest of the three.

Consider a company selling capital equipment into a regulated environment. The customer expects a decade of support. The distributor commits to stocking for perhaps half that. The component maker, meanwhile, has already published a discontinuance notice. Between those three positions sits the manufacturer, holding a design it cannot easily change because the change would trigger re-qualification, re-certification or a fresh round of customer validation.

The result is a set of costs that are real but rarely modelled: the working capital tied up in a last-time buy, the storage and traceability burden of that inventory, the engineering time absorbed by an unplanned redesign, and the margin erosion from buying through brokers at a premium. None of these appear in a typical cap table conversation. All of them affect the cash profile of the scale-up years.

Counterfeit exposure follows scarcity

Scarcity creates a secondary market, and secondary markets in electronic components carry well-documented integrity risk. When a part is no longer available through franchised channels, purchasing moves to independent distributors and brokers. The counterfeit and refurbished-component problem is a persistent feature of that trade, and it is particularly unwelcome in safety-related, medical-adjacent or defence-adjacent applications where a failed component is not merely an inconvenience.

For an investor, the signal to look for is not whether a company has ever bought from a broker. It is whether the company has a documented process for doing so: incoming inspection, traceability records, specified test protocols, and a clear policy on which parts may never be sourced outside the authorised channel. A firm that can answer that question crisply has usually thought about the rest of its supply chain too.

A diligence checklist

The following questions are worth putting to any UK deep-tech hardware business with a multi-year service obligation.

• BOM longevity review. Has the company analysed its bill of materials for lifecycle status within the last twelve months? Which parts carry end-of-life or last-time-buy notices, and what proportion of the BOM by value and by criticality sits with single-source suppliers?
• Spares and stocking policy. Is there a stated stocking strategy for critical parts, and is it funded? A last-time buy that is discussed but not budgeted is a plan, not a position.
• Redesign readiness. For each at-risk subsystem, is there a qualified alternative, a documented change path, and an estimate of the re-qualification cost? Companies that have already done this work for one or two parts tend to have a repeatable method.
• Counterfeit controls. Are purchasing routes mapped, and are inspection and traceability requirements defined? Does the company distinguish between commodity parts and safety-critical parts in its sourcing rules?
• Maintenance economics. What is the cost of servicing the installed base today, and how does it trend as the fleet grows? Rising service cost per unit on a growing installed base is an early warning.
• Contractual alignment. Do customer support commitments exceed the company's ability to source parts? Where they do, is there a contractual mechanism to renegotiate or to charge for lifetime support?
• Data continuity. For instrument-dependent businesses, is there a plan for preserving measurement continuity if a key instrument becomes unsupportable?

What builders should do before the round

Obsolescence management is unglamorous, which is precisely why it differentiates. A UK deep-tech firm that can present a component lifecycle register alongside its technical roadmap signals a maturity that is difficult to fake. The register does not need to be elaborate. It needs to be current, owned by a named person, and linked to the engineering change process rather than to a spreadsheet maintained in isolation.

There is a broader point about the UK ecosystem here. Programmes such as the High Value Manufacturing Catapult and the network of testbed facilities give deep-tech firms access to serious equipment and expertise. Less accessible is the kind of supply-chain intelligence that larger manufacturers take for granted. Building that capability in-house, even modestly, is a competitive advantage in its own right — and a reassuring thing for an investor to find during diligence.

The quiet valuation risk in UK deep tech is not usually a failed technology. It is a product that works, sells, and slowly becomes impossible to support. Pricing that risk properly, before it is forced into view, is the difference between a company that scales and one that spends its growth years solving a problem it created at design freeze.