Support

Primary contact

Email: integrations@collectivex.health

Include in every ticket:

Your tenant ID (cxh-sandbox-oura, cxh-prod-oura, etc.)
request_id from your client logs (if the issue correlates to a specific request)
trace_id from the response body (if the response body had one — 422/500/502 do; 401/403/429 don't)
HTTP status code
Timestamp in UTC (ISO 8601) of when the issue occurred
A one-line summary in the subject

Example subject: [cxh-sandbox-oura] 502 persistence_failed recurring, trace_id 01HQRS4A...

Response-time expectations

During sandbox integration phase (initial onboarding, typically the first 4 weeks after you receive your sandbox key):

Severity	First response	Resolution target
URGENT — all traffic broken, key compromised, consent revoked mistakenly	1 hour (business hours) / 4 hours (outside)	4 / 8 hours
HIGH — specific endpoint broken, consistent error pattern, rate-limit-increase request for an upcoming launch	4 business hours	2 business days
NORMAL — question about docs, feature request, expected behavior clarification	1 business day	3–5 business days
LOW — nice-to-have, future-feature discussion	2–3 business days	best-effort

Business hours: Monday–Friday, 09:00–18:00 CET.

Outside business hours: URGENT tickets only. For a genuinely URGENT issue (key compromise suspected, all production traffic failing), include URGENT: in the subject and expect a reply within 4 hours; other severities wait until the next business day.

Escalation

If you've been waiting longer than the targets above:

Reply to your existing ticket with ESCALATION: prefix in the subject.
Include the original ticket ID and the first-response timestamp.
The ticket gets re-routed to the on-call engineer.

Do not open a second ticket for the same issue — it fragments the thread and slows resolution.

What to expect

Acknowledgement within the "first response" target above. The acknowledgement may be "we're investigating" — that's fine.
Triage update at the resolution target. If we can't resolve in the target window, we tell you where we are.
Post-mortem for URGENT incidents that affected production traffic — we share a brief write-up within 5 business days of resolution.

What speeds things up

A minimal reproduction (1–2 cURLs showing the issue) is worth 10 paragraphs of description.
request_id + trace_id let us pull the exact Langfuse trace and server-side logs in seconds.
Saying what you tried already ("I've rotated the key, issue persists") rules out a class of first-line-response suggestions.

What slows things down

Screenshots of terminal output instead of copy-paste text (we can't grep screenshots).
"Doesn't work" without status code / response body / timestamp.
Forwarded threads with 5 people CC'd and no clear owner on your side.

Not support

For contract, billing, DPA, or partnership questions, email partners@collectivex.health — that gets routed differently. The integrations@collectivex.health inbox is for technical integration issues only.

Status page

We do not currently publish a public status page. For sandbox, there is no SLA. Prod cutover will be accompanied by a status page — it'll be announced in the changelog when it launches.

Security issues

For suspected security vulnerabilities (not routine key rotation — actual vulnerabilities), email security@collectivex.health and include SECURITY: in the subject. We'll acknowledge within 24 hours and give a disclosure timeline. Please do not disclose publicly until we've had a chance to patch.