Spare parts obligations are negotiated at the start of a programme, when the production terms are the focus and the aftermarket is a schedule somebody agrees to in the final week. They then become relevant a decade later, when the people involved have moved on and the only record of what was agreed is the document. By that point the supplier's own commercial team is usually reading the clause for the first time as well.
That timeline is the defining characteristic of this clause and the reason it is worth more attention than it usually gets. Most contractual obligations are exercised while the relationship is active and both parties have commercial reasons to cooperate. Parts support is exercised after the programme has ended, when the supplier has no further volume to win and every incentive to treat small orders as an inconvenience.
The clause therefore has to work without goodwill. It needs to be specific about duration, price, lead time and what happens when a component goes obsolete, because none of those will be resolved generously in year nine. It is a long-dated commitment of the kind that only survives if it is tracked, in the same way that PPAP and quality commitments have to be tracked past approval.
The elements of a workable commitment
A parts clause that holds up over a long support term addresses five distinct risks, each of which materialises at a different point in the programme's life.
- Support duration, counted in years from end of series production, with the end-of-production date itself defined and notified.
- Lead time and minimum order, set realistically for small batches, with a separate expedite provision for field-down situations.
- Pricing basis, using a volume-banded table or a capped escalation index, rather than leaving low-volume pricing to later agreement.
- Tooling rights, covering ownership, location, maintenance duty, notification before movement and a right to take possession on trigger events.
- Obsolescence and substitution, requiring notification of sub-component unavailability and defining which substitutions need approval.
Pricing is where most of these agreements come under strain. A part made in tens of thousands annually during production may be ordered in dozens in year eight, and the real unit cost can rise by an order of magnitude. A supplier facing that economics without a contractual mechanism will either quote a price the buyer considers opportunistic or decline the order. Both outcomes are predictable at signature and both are avoidable with a volume-banded table. The bands do not need to be precise; they need to exist, so that the conversation in year eight is about which band applies rather than about what the price should be.
A support term counted from end of production only works if end of production is a defined, notified event. State who declares it, what notice is given and how it is recorded, otherwise the clock has no agreed start and the obligation becomes arguable at exactly the point it is needed.
Tooling rights that can actually be exercised
Buyer-owned tooling is standard and frequently useless in practice, because ownership on paper is not the same as the ability to recover and use the tool. The point at which the right matters most is also the point at which it is hardest to exercise, which is a supplier insolvency where an administrator is deciding what belongs to whom. Five provisions make the right exercisable.
- Documented location. The plant and address where each tool is held, maintained as a register rather than recorded once at launch.
- Movement notification. Advance notice before a tool moves between plants or to a sub-tier, with the new location recorded.
- Maintenance obligation. A duty to maintain the tool in production-capable condition, with a condition report available on request.
- Marking and segregation. Physical identification of buyer-owned tooling, which matters considerably in an insolvency.
- Access and removal rights. A right to enter, inspect and remove on defined triggers, with the cost allocation stated in advance.
A five-step implementation roadmap
Step 1, Inventory existing commitments by programme
Pull the parts support clauses from the current supply base into a register keyed to programme, with the support term, its start trigger and the expiry date. Most organisations find that several terms have already expired without anyone noticing.
Step 2, Record the end-of-production date when it happens
Make declaring end of production a deliberate step that updates the obligation record and starts the support clock. Where the date has passed without being recorded, reconstruct it from production data and confirm it with the supplier in writing.
Step 3, Convert pricing language into a table
Where the clause references later agreement on low-volume pricing, negotiate the volume bands now rather than at first order. The leverage to do this is highest while the programme still carries production volume.
Step 4, Build and maintain the tooling register
Record each tool, its owner, its physical location, its condition at last inspection and the supplier holding it. Reconcile the register against supplier confirmation annually, because tools move and the movements are rarely notified.
Step 5, Schedule the last-time-buy decision
Set an obligation record ahead of support term expiry so the forecast and final order decision are made with time to spare. The alternative is discovering the expiry when a field order is rejected, at which point the options are poor.
Track commitments past the programme
Support terms, tooling locations and pricing bands held as records that outlive the project team.
Why these obligations go missing
The practical reason parts commitments lapse is organisational rather than contractual. The programme team disbands at end of production, the commodity manager moves to an active programme, and the supplier record stops being reviewed because there is no current spend against it. The obligation survives in the document and nowhere else. Nobody decides to stop managing it; the supplier simply drops out of every report that drives attention, because every one of those reports is built on current spend. Holding it as a dated record with an owner who still has a job after the programme closes is the whole fix, and it is the same reason index-linked terms need a standing settlement cycle rather than ad hoc attention, as set out in controlling cost pass-through clauses.
Commitment elements and when they bite
The table shows when each element becomes relevant across the life of a programme. Terms vary widely by industry and product life, and should be set against the expected field support period.
| Element | Negotiated at | Exercised at | Record to maintain |
|---|---|---|---|
| Support term and start trigger | Programme award | End of production onward | Expiry date with declared start |
| Volume-banded pricing | Programme award, best leverage | Falling volume years | Price table by annual quantity |
| Lead time and expedite | Programme award | Every field order | Performance against committed lead time |
| Tooling location and condition | Launch, maintained annually | Transfer or supplier failure | Tooling register with inspection date |
| Last-time-buy notice | Programme award | Approaching term expiry | Scheduled decision before expiry |
The question worth asking of any mature portfolio is how many products currently supported in the field are covered by a parts commitment that is still in force, and whether anyone can name the expiry date for the top ten. Where the answer is unclear, the exposure is already live; it simply has not surfaced yet because no customer has needed an awkward part. The cost of finding out late is rarely the part itself. It is the engineering time spent qualifying an alternative source for a component that was supposed to be covered.
For related guidance, see IATF 16949 supplier quality agreements, a control guide, Price adjustment clauses in direct material contracts and ESG and supply chain due diligence clause tracking.
