Managing single-source risk

Sole source and single source are different problems

The distinction is worth keeping straight because the remedies differ entirely.

• Sole source: only one supplier in the world can provide it. Patent, proprietary process, unique material, or a specification written — sometimes accidentally — around one company's product.
• Single source: several suppliers could provide it; you have chosen to buy from one. Volume leverage, qualification cost, relationship inertia, or simply nobody ever revisited the decision.

Single source is a choice you can revisit at a known cost. Sole source is a constraint you must design around. Most businesses discover, on auditing their spend, that a large share of what they assumed was sole-source is actually single-source held in place by habit.

Finding your real exposure

Spend is a poor proxy for risk, and ranking suppliers by invoice value will mislead you. The question is not what you pay them but what happens if they stop. Work through your supplier list against four questions:

1. How long could we operate without them? Count actual usable inventory, not what the system says. Days, not a comfortable adjective.
2. How long to qualify an alternate? Sourcing, sampling, testing, regulatory approval, customer sign-off where required, first article inspection. Be pessimistic; this number is almost always underestimated by half.
3. What does the gap cost per week? Lost margin, expedite premiums, penalty clauses, and the customers who do not come back.
4. How likely is disruption? Financial health, geographic concentration, labour situation, their own single-source dependencies, and how much of their capacity you represent.

Where the qualification time exceeds the inventory cover and the weekly cost is material, you have found a genuine exposure. That list is usually short — often five to fifteen items — which is what makes the exercise affordable to do properly.

Five responses, cheapest first

1. Buy time with inventory

Safety stock is an insurance premium paid in working capital and warehouse space. It is the fastest control to implement and the easiest to size: hold enough to cover the qualification time for an alternate, not an arbitrary number of weeks. For a component with a six-month qualification path, ordinary inventory targets are irrelevant.

2. Qualify an alternate without buying from them

The expensive part of switching is qualification, not purchasing. Run the alternate through sampling and testing, get them approved, and then buy nothing — or buy a token volume annually to keep the approval live and the relationship warm. You have converted a six-month outage into a two-week one for a fraction of the cost of dual sourcing.

3. Split the volume

Genuine dual sourcing at something like an 80/20 split keeps both suppliers active and gives you real price discovery. It costs you volume leverage and adds management overhead, so reserve it for items where the exposure justifies the ongoing expense. An uneven split is usually better than an even one: the minority supplier stays qualified without either party pretending the relationship is equal.

4. Design the dependency out

The most durable fix and the slowest. Where a specification was written around one company's part, rewriting it to a performance standard can open the market to several. This is engineering work rather than procurement work, which is why it rarely happens without someone specifically asking for it during design review.

5. Contract for continuity

Where none of the above is practical, buy protection in the agreement: a capacity commitment, an obligation to hold buffer stock on your behalf, notice periods on discontinuation, access to specifications and tooling on insolvency, and a right to audit their continuity plan. Paper does not manufacture anything, but it improves your position and, more usefully, forces the conversation while everyone is calm. The relevant clauses are covered in what belongs in a supply agreement.

What concentration is worth

None of this argues for spreading every purchase across three vendors. Concentration buys real things: better pricing, priority during shortages, engineering support you would not otherwise get, and a supplier who knows your business well enough to tell you when you are asking for the wrong part. Fragmenting spend to feel safe usually produces higher prices, more administration and worse service across the board — three mediocre relationships instead of one good one.

The discipline is to concentrate deliberately, know exactly which concentrations could stop the business, and hold a specific, costed plan for each of those. Everything else can stay single-sourced with a clear conscience.

Reviewing it on a schedule

Exposure moves. A supplier gets acquired, a second source exits the market, your volume doubles until you are 40% of a small firm's revenue, a component goes end-of-life. Put the critical-item review on the same quarterly cycle as your vendor scorecard and the two reinforce each other: the scorecard tells you a supplier is deteriorating, the exposure list tells you how much that deterioration is worth. Where terms are being dictated to you rather than negotiated, as described in negotiating supplier payment terms, that is concentration showing up on the invoice.