Sole-Source Housing Supplier Asks for 9% While a Second Source Waits

Incumbent, effective cost:

• Price: $18.40
• Disruption cost actually incurred: ($340k + $180k + $95k) ÷ 240,000 = $2.56/unit
• Effective: $20.96/unit

Alternative:

• Price: $21.10
• Tooling: $400k ÷ 5 years ÷ 240,000 = $0.33/unit
• Effective at full volume: $21.43/unit - before qualification, transition inventory, PPAP runs and learning-curve scrap.

So the two are within about 2% - the TCO does not, by itself, justify a switch. Two caveats matter more than the headline number:

1. The $2.56 is a one-year realization of a low-frequency, high-severity event. Two outages in one year is a small sample. The right input is expected annual disruption cost (probability × severity), not last year's invoice - if the true frequency is one outage every three years, the incumbent's effective cost is closer to $19.3 and the alternative looks expensive.
2. Tooling amortization is volume-dependent. At a 30% allocation, the tooling contribution becomes ~$1.11/unit, so the alternative's per-unit cost rises as its volume share falls. The 'move 100%' and 'dual source' options do not share a cost basis, and comparing 100% alternatives against a split is a common analytical error here.

Qualify the alternative for 20–30% of volume; keep the incumbent on the majority.

• Rationale: the real exposure is not the $0.70 price gap, it is a sole source on a part with no alternate and a demonstrated three-week recovery time on a line that stops. A second source is insurance, and the premium should be paid only on the volume needed to keep that source viable and tooled - roughly 50,000–70,000 units.
• Structure: minimum annual volume commitment to the alternative so it holds capacity and the tool; a firm 20–30% allocation with a review trigger (incumbent misses delivery or PPM for two consecutive quarters → allocation shifts automatically).
• Do not split 50/50. That halves the incumbent's absorption on a supplier already running thin margins, raising the probability of the failure the split is meant to prevent. A healthy majority/minority split keeps both alive.
• Avoid the 100% move: it maximizes transition risk and tooling cost at the moment the incumbent is most fragile, and it converts a benchmark into a new sole source.
• Qualify the cost of the split honestly: the 20–30% volume will cost more per unit than either quoted number. Frame it as an insurance premium with a stated annual cost, so the decision can be approved on risk grounds rather than argued on price.

Neither accept it flat nor reject it - decompose it, and treat supplier financial distress as the supply risk it is.

• Request an open-book cost breakdown with the drivers behind the 9% (energy, labor, alloy, freight) and their share of cost. A thin-margin supplier asking for 9% is often describing a real cash problem; if so, refusing does not avoid the cost, it moves it to the outage line.
• Convert the ask into an indexed mechanism: a basket of published indices with a cap and a collar, reset semi-annually. This gives the supplier relief without locking in a permanent 9% on a request that may be a one-time input spike, and it gives the buyer a downward path.
• Attach conditions: verified cost data, a documented capacity and recovery plan, delivery and PPM commitments, an inventory or consignment requirement to buffer the next outage, and a right to audit.
• Use the second source as context, not as a threat. Say plainly that a qualified alternate is being developed for a minority share - that is credible only if it is true and already in motion. Never signal the split as retaliation while simultaneously asking the same supplier to invest.
• Test the worst case: model the 9% against the cost of a repeat of last year's $615,000 disruption. If the increase costs about $400,000 a year on 240,000 units, it is roughly the same order as one bad outage - which is exactly why the answer is a negotiated, monitored increase plus a second source, not a unilateral refusal.