Payroll has to run on time, every time. This page describes how Payrollix keeps the service available — the infrastructure it runs on, how we monitor it, the uptime target we hold ourselves to, and how we handle incidents and maintenance. We document this honestly: Payrollix launched in 2026, so we describe the commitments we are making rather than claiming a long historical track record we have not yet earned.
Payrollix runs on Amazon Web Services in the United States. The platform is built so that no single host is a point of failure and so that customer data is protected against loss.
Compute: Application services run on managed compute behind a load balancer; application hosts are not directly reachable from the public internet
Database: A managed relational database with automated, encrypted backups, so data can be restored if a host fails
Isolation: Services run in isolated environments, so a fault in one component does not cascade across the platform
2. Monitoring
We watch the platform continuously so that problems are caught before they spread.
Automated monitoring: Continuous automated checks cover both the web application and the background jobs that process payroll, tax filings, and payments
Error tracking: Application errors are captured and tracked centrally, with the context needed to diagnose them quickly
Real-time alerting: Issues surface to the team in real time, so an emerging problem is investigated before it affects more customers
3. Uptime Commitment
Payrollix targets 99.9% uptime for the application. As a platform that launched in 2026, we are stating this as a target we are committing to and managing toward — not a long historical average we are claiming to have already achieved. We would rather be honest about that than publish a number we cannot back up.
Just as important, payroll-critical operations are designed to survive a brief web interruption. Tax filing and direct deposit are processed through durable background queues rather than being tied to a live browser session. If the web application is briefly unavailable, in-flight work is held in the queue and processed once service is restored — it is not dropped or lost.
4. Incident Response
If an incident affects service, we communicate status by email to affected customers so you are not left guessing about what is happening and when it will be resolved.
For any incident that involves personal or payroll data, we commit to notify affected customers within 72 hours of discovery, consistent with the timeline described in our Security page.
After an incident is resolved, we review what happened, identify the root cause, and remediate it so the same failure does not recur.
5. Scheduled Maintenance
Necessary maintenance is performed in low-traffic windows, and we give advance notice when a planned change could affect availability.
We deliberately avoid scheduling maintenance during payroll-critical periods — such as quarter-end filing weeks — when our customers can least afford an interruption.
6. Enterprise SLA
Enterprise plans include a formal uptime service-level agreement (99.9%) backed by service credits if we fall short. The specific terms — including the credit schedule, measurement methodology, and exclusions — are part of the Enterprise agreement and are negotiated with each Enterprise customer.
To review SLA terms for your organization, contact sales.
7. Contact
For questions about Payrollix reliability, uptime, or our incident-response process, or to request additional information for a vendor review:
Reliability questions: security@payrollix.com
This page describes the reliability posture of Payrollix as of the "Last updated" date above. As the platform matures, our practices and commitments may evolve; material changes will be reflected here.