Legal · Service Level Agreement
Service Level Agreement.
Uptime, response times and credits for a Nexxorex-managed deployment. Enterprise variations are agreed in the signed scope. Questions to legal@nexxorex.com.
Last updated · November 2026Governing law · Bangladesh
01
Scope
- This SLA covers the Nexxorex-managed engine running on the environment named in your signed scope: the conversation layer, the integration adapters we wrote, and the admin console.
- It does not cover your own phone system, ERP, CRM, carrier, SMS gateway or model provider. Where an outage of those systems degrades the service, we troubleshoot and coordinate but do not credit SLA against the outside outage.
02
Monthly uptime commitment
- Standard tier: 99.5% monthly uptime of the conversation engine.
- Enterprise tier (default for banks, insurance and telco deployments): 99.9% monthly uptime of the conversation engine, with a secondary region hot-standby included.
- On-premise tier: measured against the environment you operate; targets are the ones your operations team can commit to, and we support those targets, not override them.
- Uptime is measured per calendar month, from the admin console's health probes, and reported inside the first five business days of the following month.
03
Severity levels and response
- P1 (service down for all users or a sensitive data breach): first human response within 15 minutes, 24x7. Status update every 60 minutes until resolved.
- P2 (major feature unavailable, work-around exists): first response within 2 hours during business hours, 4 hours outside. Status update every 4 hours.
- P3 (minor defect, cosmetic, non-blocking): first response within 1 business day. Written plan inside 5 business days.
- P4 (change request, question, roadmap item): first response within 2 business days. Routed to the roadmap conversation rather than the incident queue.
04
Target time to resolve
- P1: workaround or full fix within 4 hours of acknowledgement. If a fix cannot be produced inside that window, status moves to a published incident with hourly updates until resolved.
- P2: fix inside 2 business days.
- P3: fix inside the next release window scheduled after the ticket.
- P4: scheduled in the next roadmap review you attend.
05
Scheduled maintenance
- Maintenance windows that may interrupt the service are announced 72 hours in advance to the operations contacts on your account.
- Default maintenance window is Fridays 02:00 to 04:00 Bangladesh Standard Time. Changes to the default are agreed in the signed scope.
- Hot-fix deployments that do not interrupt the service are not announced; the audit log records the deployment timestamp and the change hash.
06
Service credits
- If a month falls below the committed uptime, a credit is applied to the following month's retainer invoice: 10% credit for the first 0.5 percentage point below target, 25% for the next 1.0 point, 50% for anything beyond that.
- Credits do not apply to usage charges (voice minutes, model calls, SMS). They apply to the retainer and managed-service portion of the invoice.
- Credits are claimed in writing inside 30 days of the monthly SLA report; after that window the credit lapses.
07
What voids the SLA for a given incident
- Changes made to the environment by your staff or by a third party, outside a change window agreed with us.
- Outages of upstream providers (voice carrier, model provider, SMS gateway) for the duration of their outage; we credit nothing against someone else's problem.
- Force majeure, which applies equally to both parties.
08
Reporting and review
- Monthly SLA report: uptime, incident count by severity, MTTA (mean time to acknowledge) and MTTR (mean time to resolve), and the credit amount if any.
- Quarterly review: a one-hour call with your operations lead to walk through the quarter's incidents, trends and preventive actions.