Kurly Pay · 2022.09 — 2023.12

Building Statistics Monitoring and Researching Scalability

  • Built Redash-based dashboards and researched a scalable statistics architecture

Background

For non-developers (operations/planning) to check key metrics, they had to request them from developers every time.
Quick metric checks required a self-service dashboard, and over time this expanded into researching a statistics architecture that went beyond simple query sharing.

Outcomes

  • Enabled non-developers to check key metrics directly through Redash-based dashboards
  • Consolidated frequently-used queries and lookup patterns for operations/planning teams
  • Researched a scalable statistics architecture considering the data model, aggregation methods, query performance, and operational convenience

Details

Redash-based monitoring setup

Payment method / share chart

Redash chart showing payment method breakdown and share.

Gift card basic statistics dashboard

Redash dashboard of basic gift card statistics for operations.

Gift card issuance time-series graph

Time-series graph tracking gift card issuance over time.

Research into scalable data statistics

Building on hands-on Redash operations, researched a scalable statistics architecture covering the data model, aggregation methods, query performance, and operational convenience.

Background (discussion / summary of existing approach)

Background discussion summarizing the existing query-sharing approach.

Goal (defining the statistics to automate)

Goal definition for the statistics to be automated.

Plan (ETL/ELT, AWS Pipeline study)

Plan studying ETL/ELT and AWS Pipeline for scalable statistics.