05Design patternsPayment Systems & Financial Consistency
05 · Reusable pattern
Payment Systems & Financial Consistency
Use ledgers, idempotent commands, reconciliation, audit trails, and explicit settlement states.Lesson spine
What you need to understand.
Payment systems combine a strongly consistent internal ledger with asynchronous, sometimes ambiguous external settlement.
Payment lifecycle
Model created, authorized, captured, settled, failed, voided, and refunded as explicit versioned transitions.
Idempotent commands
Use merchant-scoped keys and request fingerprints so retries return the original payment result.
Double-entry ledger
Record balanced debit and credit entries with exact decimal amounts; correct mistakes by compensating entries, not edits.
Provider ambiguity
After a timeout, query status or reconcile before retrying a charge that may already have succeeded.
Reconciliation
Compare internal records with processor and bank statements, classify mismatches, and retain an auditable repair trail.
Webhooks and security
Sign events, retry at least once, minimize PCI scope through tokenization, encrypt sensitive data, and enforce least privilege.
Before the boxes
Frame the decision.
What must work
Use ledgers, idempotent commands, reconciliation, audit trails, and explicit settlement states.
What changes the design
Exact decimals · currency rules · settlement windows · audit · zero imbalance
What owns the truth
Identify the component that commits authoritative state, then separate synchronous confirmation from derived work.
What stays simple
Do not add global coordination, multi-region writes, or a specialized store until a requirement earns the complexity.
Architecture map
Trace ownership, not just traffic.
Follow the decision from left to right. Every arrow should have a reason.
Carries financial intent
Tracks authorization and settlement
Commits balanced entries
Calls rails
Orders callbacks
Compares statements
Retains evidence
Walk one representative request across every arrow. Say whether the handoff is synchronous or asynchronous, what identity makes a retry safe, and which step changes authoritative state.
- 01
Command — Carries financial intent Define the output contract before moving to the next owner.
- 02
Workflow — Tracks authorization and settlement Define the output contract before moving to the next owner.
- 03
Ledger — Commits balanced entries Define the output contract before moving to the next owner.
- 04
Provider adapter — Calls rails Define the output contract before moving to the next owner.
- 05
Webhook inbox — Orders callbacks Define the output contract before moving to the next owner.
- 06
Reconciler — Compares statements Define the output contract before moving to the next owner.
- 07
Audit system — Retains evidence Confirm the result and emit the evidence needed to reconcile it.
Decision table
Make the trade-offs explicit.
| Decision | Defensible position | Cost to acknowledge |
|---|---|---|
| Primary mechanism | Strong ledger consistency coexists with eventual external settlement; do not conflate them. | The stronger guarantee usually adds coordination, latency, state, or operational work. |
| Sync vs. async | Keep only correctness-critical confirmation synchronous. Move derived views, notifications, analytics, and cleanup behind a durable boundary. | Async work needs idempotency, lag monitoring, replay, and a product definition for partial completion. |
| Simple vs. scaled | Begin with one logical owner and a clear API. Partition or replicate only the resource proven to be the first bottleneck. | Migration requires stable identities, versioned contracts, backfill, and a rollback path. |
Failure review
Design the recovery path.
Topic-specific risk
Partial provider success and duplicate callbacks create unexplained money.
ResponsePersist enough identity and state to distinguish retry, resume, compensation, and operator repair.
Dependency timeout
A timeout is ambiguous: the remote side may have failed, succeeded, or still be running.
ResponseUse deadlines, bounded backoff with jitter, idempotency keys, and a status or reconciliation path.
Overload or skew
Average capacity can look healthy while a tenant, key, partition, region, or expensive request saturates one owner.
ResponseExpose queue depth and hot-key share, apply backpressure, isolate tenants, and degrade optional work before correctness.
Evidence + level bar
Prove the design can be operated.
Health of the promise
Measure user-visible latency or freshness, correctness drift, saturation, retry volume, and time to recover. Alert on the failed promise—not only CPU.
Complete and clear
Finish the happy path, identify the state owner, choose reasonable building blocks, and explain one scale mechanism.
Trade-offs and failure
Separate read and write paths, define consistency, explain partitioning, and make duplicate or partial failure safe.
Evolution and operations
Discuss multi-region boundaries, migration, tenant isolation, capacity, observability, and how the architecture changes over time.
Interview language
Open the deep dive with a claim.
“For Payment Systems & Financial Consistency, the decision I want to make explicit is this: Strong ledger consistency coexists with eventual external settlement; do not conflate them. I’ll trace the state-changing path first, show where the result becomes durable, then test the design against the highest-risk failure and our target scale.”
08 · Retrieval check
Can you defend it without the page?
- For Payment Systems & Financial Consistency, where is the correctness boundary and which failure would you test first?
- Which component owns committed truth, and what event or response proves the commit?
- Where is the first scaling or coordination bottleneck under the stated envelope?
- What happens after an ambiguous timeout or duplicate operation?
- Which complexity would you remove at one hundredth of the scale?