1. Availability Target

99.9% monthly uptime for the production web application and API, excluding planned maintenance.
Statement extraction and categorization are asynchronous. Availability means the ability to upload, queue, review, and report; processing completion time is a target, not a guarantee.
Typical extraction completes within minutes; large batches or provider throttling may extend this.

2. Planned Maintenance

Maintenance windows are announced via the public status endpoint and in-product banner, with a countdown and live updates.
We aim for at least 24 hours' notice; urgent security patches may proceed with shorter notice.
Maintenance is scheduled outside North American business hours where practical.

3. Support Response Targets

Severity 1 — outage or data-integrity risk: initial response within 1 hour, continuous effort until mitigated.
Severity 2 — material degradation (e.g., extraction backlog, invoice delivery failures): initial response within 4 hours.
Severity 3 — defect or how-to question: initial response within 1 business day.

Contact support@ledgerpilot.com or /support. Flag security concerns in the subject line for priority triage.

4. Operational Resilience

LedgerPilot is built on Microsoft Azure with resilience designed in:

Queue-based processing with visibility timeouts, retry, and poison-message isolation so a failed document does not block others.
Automatic recovery of stale or stuck statements, with throttled re-queue and administrator requeue controls.
Multi-endpoint document intelligence with health tracking, backoff, and cooldown failover.
Invoice delivery retries (up to three attempts) with administrator notification on failure.
Isolated PDF rendering service with health checks and browser recovery.
Horizontal scaling via Azure Container Apps, Azure SQL managed backups, and Blob Storage durability.
Monitoring through Application Insights with request, dependency, and exception telemetry.

5. Data Protection and Recovery

Database point-in-time restore via Azure SQL managed backups.
Uploaded statements and generated PDFs stored in durable Azure Blob Storage.
Recovery objectives for production: RPO ≤ 1 hour, RTO ≤ 8 hours for regional service incidents.

6. Exclusions

This commitment does not cover downtime or degradation caused by:

planned maintenance announced as described above;
outages of third-party providers (Microsoft Azure, Stripe, SendGrid, AI services) beyond our control;
customer networks, browsers, or integrations;
unsupported, malformed, or non-PDF documents;
suspension for breach of the Terms or non-payment;
force majeure events;
early-access, pilot, or beta features and non-production environments.

7. Service Credits (Enterprise Agreements)

Where an executed enterprise agreement provides for credits, the following schedule applies unless the agreement states otherwise. Credits are issued as processing credits, must be requested within 30 days of the affected month, and are the sole remedy for availability shortfalls.

99.0% – 99.9%: 10% of that month's committed fees.
95.0% – 98.99%: 25% of that month's committed fees.
Below 95.0%: 50% of that month's committed fees.

Self-serve credit pack customers are not eligible for service credits unless required by law.

8. Status and Transparency

Current version information and maintenance status are published at /status. Material incidents are summarized after resolution on request from affected customers.

Service Level & Reliability Commitment