SLA — Service Level Agreement

1. Availability (Uptime)

99.5% monthly for the web interface (app.capapp.ro), REST API (/api/*) and email outbound sync (SMTP). Calculated on minutes per calendar month, with scheduled maintenance windows excluded.

Does not count as downtime: (a) scheduled maintenance announced in advance, (b) force majeure (massive DDoS, regional Cloudflare/hosting incident), (c) issues with client network or browser.

Uptime statistics published monthly in `/admin/billing/current` (future endpoint) or on request in monthly report.

2. Incident response time

Priority (severity) levels:

P1 (critical) — app completely inaccessible or core feature (login, skill matrix, IDP) broken. Initial response: 4 hours in-hours (Mon-Fri 09:00-18:00 EET). Resolution: best effort continuous.

P2 (major) — secondary feature broken, no workaround (e.g. DSP report fails to generate, PIN reset broken). Initial response: 1 business day.

P3 (minor) — bug with workaround (e.g. button triggers refresh instead of inline save). Initial response: 5 business days.

P4 (cosmetic) — typo, colour, spacing. Resolved in next release (~2-4 weeks).

Clock starts when ticket is received at support@cleanthes-apps.com or reported through the support system during specified hours.

3. Support channels

Standard plan: email only at support@cleanthes-apps.com, in-hours Mon-Fri 09:00-18:00 EET.

Enterprise plan: email + dedicated 24/7 phone (+2000 EUR/year extra), plus dedicated account manager for escalations.

Average email reaction time: under 2 hours in-hours for all tickets.

4. Backup and recovery

Daily automated backup of the database (SQLite) and app configuration, retained 90 days. Backups stored in the EU (Cloudflare R2 or equivalent), encrypted at rest.

RTO (Recovery Time Objective) — 24 hours from critical incident to complete application and data restoration.

RPO (Recovery Point Objective) — maximum 24 hours of data can be lost in the worst-case scenario (last daily backup).

Monthly restoration test on parallel environment, report available on request.

5. Scheduled maintenance

Notification with minimum 48 hours in advance for any maintenance implying downtime > 15 minutes.

Preferred window: Sunday 02:00-06:00 EET (lowest traffic in RO factories).

Notification via email to all Admin-role users in the tenant, plus in-app banner 24h before.

6. Penalties for SLA non-conformance

If monthly uptime falls below 99.5%, the client receives 10% credit on the next invoice for each 1% uptime lost below threshold. Example: 98.4% uptime in a month → 11% credit.

Maximum credit per month: 50% of that month's invoice (no cash refund, only credit applied to next invoice).

Claim must be submitted to support@cleanthes-apps.com within maximum 30 days from the end of the affected month.

If the app is down consecutively more than 72 hours in a month, the client can terminate the contract without penalties and receives pro-rata refund for the remaining days.

7. What this SLA does not cover

Wrong client configurations (e.g. notifications sent to non-functional email addresses, misconfigured roles).

Training the client users — inclusion of training sessions happens separately via setup fee or Enterprise plan.

Custom integrations (ERP, HR, other APIs) — provided through separate projects with specific SLAs.

Data recovery over 90 days — older backups are automatically deleted.

8. SLA modifications

Notice 90 days in advance for any modification that could negatively affect the client (e.g. changed penalty threshold, extended maintenance window).

Modifications that improve the SLA (e.g. P1 response reduced from 4h to 2h) take effect immediately, with notification.