Service Level Agreement (SLA)
Last updated: 26 February 2026
1. Definitions
- Monthly availability: the percentage of time the Service is accessible and functional, calculated over a monthly measurement period (calendar month).
- Incident: any unplanned event causing unavailability or a significant degradation of the Service.
- Planned maintenance: scheduled work announced at least 72 hours in advance (except in emergencies).
- Measurement period: the full monthly period used to calculate availability (from the 1st to the last day of the calendar month).
- Severity: the incident impact level according to the matrix defined in the Support section.
- Service: the Numezis SaaS platform accessible through official URLs, excluding test, beta, sandbox, or pre-production environments.
- Downtime: the period during which the Service is unavailable or does not meet the defined operating criteria.
2. Scope
This SLA applies exclusively to:
- production environments of eligible paid plans (Starter, Growth, Business, Enterprise);
- features and modules explicitly included in the subscribed plan; and
- documented public APIs and standard user interfaces.
Express exclusions:
- beta, pilot, preview, early access, or experimental services/features/modules;
- staging, sandbox, development, or test environments;
- interruptions caused by factors outside Numezis' reasonable control (see Exclusions); and
- customer failure to follow configuration best practices or contractual limits.
3. Availability target
Numezis commits to the following minimum monthly availability, depending on the subscribed plan:
- Starter: target 99.0%; guaranteed 98.5%.
- Growth: target 99.5%; guaranteed 99.0%.
- Business: target 99.9%; guaranteed 99.5%.
- Enterprise: target 99.95%; guaranteed 99.9%.
Calculation method:
Availability (%) = (Total time in the measurement period – attributable downtime) / Total time in the measurement period × 100
The calculation excludes planned maintenance and downtime falling under the Exclusions section.
4. Planned maintenance and communications
4.1. Maintenance windows. Numezis prioritizes low-activity windows (generally between 22:00 and 06:00 CET on Saturday or Sunday). Any planned maintenance likely to cause more than 15 minutes of unavailability will be announced at least 72 hours in advance via:
- in-app notification (banner + notification center); and
- email to designated administrators.
4.2. Emergency maintenance. In case of an imminent threat to security or stability (for example, a critical vulnerability), Numezis may intervene without prior notice. An ex post notification will be sent within 24 hours.
5. Technical support
Channels, coverage, and objectives (CET).
- Starter: email + help center; business days 09:00–18:00; first response ≤ 24 business hours; no guaranteed resolution time.
- Growth: email + in-app chat; business days 08:00–20:00; first response ≤ 12 business hours; no guaranteed resolution time.
- Business: email + chat + priority phone; 24×5; first response ≤ 4 business hours; indicative resolution ≤ 48 hours for critical incidents.
- Enterprise: email + chat + dedicated phone (+ Slack option); 24×7; first response ≤ 1 hour (critical) / ≤ 4 hours (high); indicative resolution ≤ 8 hours (critical) / ≤ 24 hours (high).
Incident severity matrix.
- Critical (P1): Service fully unavailable for all users; first response ≤ 60 minutes (24×7); resolution: best efforts.
- High (P2): major degradation—key functionality unavailable for many users; first response ≤ 4 hours; target resolution ≤ 24–48 hours.
- Medium (P3): moderate degradation—limited impact; first response ≤ 8 hours; target resolution ≤ 5 business days.
- Low (P4): minor/cosmetic issue; first response ≤ 24 business hours; resolution: next sprint or release.
6. Incident management procedure
- Detection: 24×7 monitoring (uptime probes, synthetic transactions, logs/alerts).
- Qualification: triage by the engineering/on-call team based on the severity matrix.
- Handling: creation of an incident ticket in our internal incident management system.
- Communication: updates every 2–4 hours for P1/P2 via status.numezis.com and email.
- Resolution: fix deployment or workaround.
- Post-incident review (blameless post-mortem): mandatory for any P1/P2 incident.
7. Service credits
Service credits apply only to Business and Enterprise plans and only if actual monthly availability falls below the guaranteed threshold.
Credit schedule:
- Actual availability < guaranteed threshold and ≥ 95.0%: 10% credit on fees for the affected month.
- Actual availability < 95.0% and ≥ 90.0%: 25% credit.
- Actual availability < 90.0%: 50% credit.
Conditions:
- Credits are calculated on fees actually paid for the affected month (excluding variable fees, overages, and taxes).
- Credits are granted as a credit note applied to the next invoice.
- A written request is required (support@numezis.com) within 30 days after the end of the affected month.
- Credits are the Customer’s exclusive and sole remedy for SLA non-compliance (except in cases of willful misconduct or gross negligence).
8. Exclusions (non-attributable downtime)
Downtime is not counted toward availability where unavailability results in particular from:
- announced planned maintenance;
- emergency maintenance for security reasons;
- force majeure;
- a third-party provider failure outside Numezis' reasonable control (for example, a global outage, Stripe outage, widespread internet disruption);
- customer acts or omissions (misconfiguration, intentional overload, non-compliant use);
- exceeding contractual limits or quotas; or
- targeted attacks (DDoS, brute force) despite reasonable protections implemented by Numezis.
9. Monitoring and reporting
Numezis maintains continuous monitoring through:
- external uptime probes (every 60 seconds from multiple continents);
- synthetic monitoring on critical user journeys; and
- internal metrics (API latency, 5xx error rate, resource saturation).
A public status page (status.numezis.com) displays in real time:
- overall Service status;
- ongoing incidents; and
- history of maintenance and major incidents (last 12 months).
10. Changes to the SLA
Numezis may update this SLA. Any material change (reduction of availability commitments or credits) will be notified at least 60 calendar days in advance. The Customer may terminate without penalty within 30 days of the notice.
11. Governing law and jurisdiction
Swiss law.
Exclusive jurisdiction: ordinary courts at the registered office of Numezis SA (Lausanne).