1,000+ QPS customer-facing dashboards, monetised at 50%+ lower TCO
A high-growth B2B fintech with 18,000+ merchants turned petabytes of transaction data into a premium self-service analytics product, priced on performance it could finally afford.
"With e6data in the mix, we finally hit p95 <2s on our customer-facing dashboards without doubling our bill. It's rare to see cost go down and performance jump that dramatically."
From cost center to premium feature
The dashboards are now sold to merchants as a premium offering, priced on performance the old stack could not deliver at any budget.
The opportunity
Petabytes of transaction data across 18,000+ merchants, and a plan to sell self-service analytics on top of it. The catch: every query had to meet strict latency SLAs at high concurrency, or the product was dead on arrival.
The wall
Approaching 1,000 QPS, the incumbent engine could not hold p95 under 2 seconds. Scaling capacity ballooned infrastructure cost while performance still lagged: high concurrency and low latency, but never both.
The evaluation
A structured proof of concept on their own data: candidate engines were scored on the criteria that decided the product - p95 latency under real concurrency, cost at scale, security, and ease of adoption - with production queries replayed rather than synthetic benchmarks.
The drop-in
e6data ran the pilot alongside the existing engines, on the same S3 tables and Glue catalog. Granular autoscaling absorbed traffic bursts and scaled back down, and near-real-time querying kept merchant insights fresh.
The result
12x faster query completion across OLAP and near-real-time workloads, 1,000+ QPS sustained with p95 under 2 seconds, and 50%+ lower TCO. The analytics product shipped as a new revenue stream instead of a write-off.
Book a demo on your own workloads
Reach out to book a demo, share challenges you're facing, and tell us how this fits into what you're currently working on or thinking about.
Prefer to self-serve? Problems we're solving →