Names and aliases
transaction failure rate, failed outcome share
Formula
failed_eligible_transactions / eligible_started_transactions × 100%
Metric card
| Unit | % |
|---|---|
| Data source | Transaction-state log, terminal status, failure reason, start time, and observation-window closure. |
| Frequency | By transaction-start cohort after the permitted fulfilment window closes. |
| Segments | micro-market, category, stage, failure reason, payment method, fulfilment method, cohort |
Interpretation errors
- Combining technical failures, voluntary cancellations, and still-open transactions.
- Including attempts that never reached the defined starting state in the denominator.
- Treating retries as independent transactions without a deduplication rule.
Example
Illustrative example: 9 classified failures among 180 eligible transactions produce a 5% rate. The figures describe no real platform.
Use
Freeze the transaction state machine and mutually exclusive terminal states first. Report causes and stages beside the aggregate rate, and treat still-open observations as censored until the window closes.
Anti-example
Reliability has not improved if unresolved transactions merely stop receiving a terminal failure status and remain open.
Limitations
- Failure taxonomy and the starting state depend on the platform archetype.
- A low rate does not establish matching quality or participant satisfaction.
Related metrics
cancellation-rate, payment-failure-rate, fulfilment-failure-rate, transaction-completion-rate
Context sources
These works provide two-sided-market context; the formula above remains an author-defined working definition.
- Jean-Charles Rochet, Jean Tirole. Platform Competition in Two-Sided Markets (2003).
- Mark Armstrong. Competition in Two-Sided Markets (2006).