Real-time analytics for an EU fintech
Client disclosure: EU fintech, 50-200 FTE (k>=5 cohort)
- p99 latency
- 4.2s380ms
- monthly cost
- 3800 EUR1100 EUR
Use the live console to inspect this record, compare it with your workflow, or start a fit check.
A mid-sized European fintech approached us to overhaul their data-pipeline ingestion. They were running batch ETL on hourly cadences and struggling to meet sub-second analytics SLAs for their internal dashboards.
Situation
The client's existing pipeline relied on cron-driven batch jobs orchestrated by a deprecated Hadoop installation. p99 read latency hovered at 4.2s. Monthly infrastructure cost: 3800 EUR.
Approach
We migrated their ingestion to a streaming architecture (Kafka + ClickHouse), retaining their existing dashboarding tool. Engagement length: 6 months, with the Lavx team embedded for the first 8 weeks.
Outcome
- p99 latency: 4.2s -> 380ms
- Monthly infrastructure cost: 3800 EUR -> 1100 EUR
- Operator handle-time per data ops issue: 8min -> 2min