Hardware DTC supply-chain resilience — the second-source playbook

The cost-benefit math

Second-sourcing increases per-unit cost typically 5-15%. The trade-off:

• No second source: ~3-5% probability per year of a meaningful disruption (1-3 month inventory gap). Expected loss: 1-2 months of revenue.
• Second source at 70/30 split: ~1-2% probability per year of full disruption. Expected loss: ~2-3 weeks of revenue (recovery time).

For a business at $5M revenue/year, the expected-loss math:

• Unilateral supplier: ~$400k/year expected loss (long tail).
• Second-sourced: ~$60k/year expected loss + ~$250k/year in higher unit costs.
• Net cost of insurance: ~negative $90k for a meaningful risk reduction.

The math justifies second-sourcing for any SKU representing >20% of revenue. The math doesn't justify it for long-tail SKUs unless they're also critical to brand identity.

The category of failure most founders don't model

Single-region exposure. Your primary supplier is in China; your secondary is also in China. A regional event (typhoon, regulatory shift, geopolitical) takes out both. Geographic diversification matters as much as supplier diversification.

The minimum: at least one qualified supplier in a different country. For US-based DTC, that often means a Mexico/Vietnam secondary alongside the China primary. The unit cost is higher (10-25%) but the geographic-independence value is real.

What kills second-sourcing

• "It's too expensive." Compute the expected-loss math; it's almost always net-positive at scale.
• "Our primary supplier won't like it." They actually will, eventually. Tell them honestly: "We're growing to a scale where we need supply-chain resilience. Your business with us doesn't shrink; it's still 70%."
• "We don't have the engineering bandwidth." True at small scale. Second-source one critical SKU per quarter; in 18 months you've covered the top 6.
• "The market won't accept the quality variance between suppliers." Quality has to be controlled at spec, not at supplier. If your QA depends on knowing the supplier, you have a different problem.

What to do today

1. List your top-10-revenue SKUs. For each, name the supplier. If any has only one, you have exposure.
2. Compute the dollar value of revenue at risk if your single-source supplier goes dark for 90 days.
3. Identify the top 1-3 SKUs to second-source first. Start the qualification process for the riskiest one this quarter.
4. Document the supplier-disruption runbook even before you have a second source — at minimum it forces clear thinking about what would actually happen.