Case Study Deskcasestudydesk.com
InsuranceSourced

Case Study: Superscript serves real-time data in seconds with Materialize

Superscript Case StudySourced & dated by Case Study Desk
Key facts · TL;DR
Company
Superscript
Industry
Insurance
Challenge
Warehouse synced only twice daily, too slow for live use
Headline result
Materialize turns twice-daily warehouse syncs into second-level live data and ML serving

Key results

1-5s
End-to-end data latency
From hours
~50ms
User-ID lookup latency
Indexed tables

The challenge

Superscript needed to operationalize a warehouse data model for real-time ML predictions and live dashboards with sub-second latency. Its prior warehouse synced data only twice a day, making it unsuitable for time-sensitive conversion work.

The solution

Superscript used Materialize as a real-time transformation and serving layer, connecting Segment data through Kafka and building materialized views in PostgreSQL-compatible SQL for dashboards and ML scoring services.

Every week, we're closing customers that we wouldn't even have had the opportunity to contact before using Materialize

TC
Tom Cooper
Data Team Lead, Superscript

The results, in context

End-to-end latency from source to dashboard dropped to 1-5 seconds, down from hours, with roughly 50ms responses for indexed user-ID lookups.

Products used

Materialize Materialize