Support SLAs

Contractual commitments only. Internal escalation theatre lives in Incident response where it belongs.

Response times (Enterprise plan)

Applies only to customers on a signed Enterprise contract. Growth customers get best-effort business-hours response.

Severity	Response	Target resolution
S1 — Critical (prod down, data loss risk)	1 hour, 24/7	4 hours
S2 — High (major feature broken)	4 hours, business hours	1 business day
S3 — Medium (workaround exists)	1 business day	5 business days
S4 — Low (question, enhancement)	3 business days	Best effort

Growth response: S1 = 4h business hours, S2 = 1 bd, S3 = 3 bd, S4 = best effort. Not contractual — best-effort targets.

Channels

Channel	Who uses it
support@tappass.ai	All paid customers
Slack Connect	Enterprise only
Dedicated contact	Enterprise only

Uptime SLA

• Enterprise: 99.9% on /v1/chat/completions, measured monthly, excluding customer-caused outages.
• Growth: best effort.
• Credits: 5% of monthly fee per 0.1% below target, capped at 25%.

Measured against the /api/health/live uptime check plus a synthetic chat.completions probe. See Monitoring & alerts for the real probe setup.

Exclusions

• Customer-caused (misconfigured policy, invalid key).
• Upstream LLM outage where TapPass is still up (we return a passthrough 502 with the upstream error).
• Scheduled maintenance announced >48h in advance.

Ticket flow

1. Customer emails support@tappass.ai or files via Slack Connect.
2. Triaged by whoever picks it up.
3. If S1, run the incident triage.
4. Comms back to the customer at the SLA cadence above.
5. Closure requires customer confirmation, not just our fix.

Also see

• Incident response — what to actually do when something breaks.
• Customer onboarding — what happens after signing.