Customer story · Rawda Health
Rawda Health passed security review in a week and caught a real failure the same afternoon
A healthcare data team needed monitoring that never touches patient data.
- to connect
- 20 min
- fewer alerts to triage
- 70%
- from security review to production
- 1 week
The problem
Rawda Health runs 46 primary care clinics across the UAE and Saudi Arabia. Its data team of four reports on appointments, waiting times and staffing from a Postgres operational database replicated into Snowflake.
The team had tried two monitoring tools before. Both wanted to sample table contents, which meant patient data leaving the warehouse, when health data in the UAE generally has to stay in the country. Both security reviews stalled. Meanwhile alerts came from three places and nobody owned them.
What changed
Tessel reads metadata, run logs and code, never query results, and stores that metadata in its UAE region, inside the country. Rawda’s security team reviewed the SOC 2 Type II report (demo), the data flow diagram and the Data Processing Addendum, and signed off in a week.
Amani Yousif, Data Lead, connected dbt Cloud, GitHub and Snowflake in 20 minutes. That afternoon Tessel caught a failed incremental load in the appointments model and opened a fix. Grouped incidents cut the alerts the team triages by 70%, and on-call now runs through PagerDuty with a clear owner for every incident.
Stack: dbt Cloud, Postgres, Snowflake, GitHub, Slack, PagerDuty.
The rollout
-
Week 1
Security review using the SOC 2 report, data flow diagram and DPA.
-
Day 1 live
Connected in 20 minutes; first fix pull request the same afternoon.
-
Month 1
Owners routed through PagerDuty; alert volume down 70%.
Setup took twenty minutes. It caught its first real failure that afternoon.
Get the same on your stack
We’ll run Tessel on a copy of your project during the demo, so you see your own incidents.