Risk allocation and guarantees
Recurring models fail when risk is priced badly or allocated to the party that cannot control it.
500+ cases•35+ industrial clients•Execution-led operating partner
What this page gives you
How to decide what you can guarantee - and what you cannot.
How to design guarantees using bands, caps, and staged commitments.
How to model downside exposure before scaling.
The commercial guardrails that protect cashflow.
How to move from pilot risk to scalable risk.
Core concepts
Practical decision logic
Identify outcomes and drivers (what causes success/failure).
Separate controllable vs uncontrollable drivers.
Guarantee only what you can influence and verify.
Use bands not binary thresholds.
Use credits first, penalties later.
Cap downside at contract level and portfolio level.
Model worst-case exposure - assume correlated failures.
Stage risk commitments through pilot and ramp clauses.
Build operational controls (monitoring, maintenance, escalation).
Standardise terms and pricing guardrails.
Common pitfalls
Guarantees before operating readiness.
No portfolio-level downside modelling.
No data rights or verification clauses.
Over-customisation creates unknown risk.
Binary thresholds that create all-or-nothing disputes.
No staging mechanism - full risk from day one.
Practical checklists
Guarantee readiness checklist
- Can you measure the outcome reliably?
- Do you control enough of the drivers?
- Is your operating model ready to deliver?
- Have you modelled worst-case exposure?
- Do you have data from pilots to validate assumptions?
Risk cap checklist
- Maximum credit/penalty per period
- Maximum liability per contract
- Maximum aggregate exposure across portfolio
- Clear termination triggers
- Force majeure and exclusion definitions
Where this shows up in deals
Availability credits with caps
Monthly exposure capped at percentage of contract value.
Shared savings with baseline governance
Upside shared only when both parties agree on baseline and measurement.
Output commitment with exclusions
Guarantee for input variability excluded - clear boundaries protect both sides.
Related content
Frequently asked questions
Q: How do you decide what to guarantee in an As-a-Service contract?
A: Only guarantee outcomes you can materially influence through your equipment, service, and monitoring capabilities. If the outcome depends on how the customer operates the asset, their raw material quality, or environmental factors outside your control, either exclude those variables explicitly or use a shared-risk model with gain/pain bands.
Q: What is the difference between performance bands and hard guarantees?
A: A hard guarantee sets a single threshold - if performance falls below it, the OEM pays. Performance bands create zones: a target zone (no adjustment), a minor shortfall zone (partial credit), and a critical zone (larger credit or penalty). Bands are more practical for industrial assets because they reflect operational variability without creating all-or-nothing risk.
Q: How do you model downside exposure before scaling an As-a-Service portfolio?
A: Run scenario analysis on each contract: what happens at worst-case utilisation, maximum maintenance cost, and full SLA credit payout simultaneously? Aggregate across the portfolio to find the maximum cash-at-risk in any quarter. If that number exceeds your risk appetite, tighten SLA bands, add caps, or restructure the financing before adding more contracts.
Q: How do you move from pilot-level risk to scalable risk management?
A: During the pilot, risk is managed deal-by-deal with close executive oversight. To scale, you need: standardised contract guardrails (caps, exclusions, measurement methods), a portfolio-level risk dashboard, actuarial-style data on failure rates and cost-to-serve, and financial reserves funded from contract margins. The transition usually happens between the 5th and 15th contract.
Want to design guarantees that scale?
Assess fit, execution risks, and the fastest path forward.