How a fintech unicorn monetised customer-facing dashboards handling 1000+ QPS concurrency

Background

A high-growth B2B fintech unicorn with a customer base of 18000+ merchants recognized an opportunity to monetize its rich transaction data. The idea was to offer its merchant clients self-service analytics dashboards built on that data, turning internal information into a revenue-generating service. Achieving this at scale meant the solution had to be highly elastic, deliver near-real-time insights, and meet strict latency SLAs for every query. 

‍

At this scale, a customer-facing analytics dashboard isn’t just an add-on—it’s a core feature that needs to be architected to perform under high concurrency and low latency.

‍

“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.”
- Staff Data Engineer

‍

Challenge: 1000 QPS Concurrency

The company’s existing query engine struggled to meet these new demands. In test runs, infrastructure costs quickly ballooned as they scaled up capacity, yet performance still lagged. At peak loads (approaching 1,000 QPS), the legacy system began to buckle, unable to keep p95 query times under the required 2-second threshold. 

‍

In short, the old stack could not handle the required high concurrency and low latency simultaneously without crashing or becoming extremely expensive. This put the analytics project at risk, as an unreliable, slow dashboard would be unacceptable to merchants and would ruin the product offering’s value proposition.

‍

Solution: Lakehouse Compute Engine Built for High QPS 

The fintech team piloted with e6data alongside their existing compute engines for the selected use case. The pilot offered the following core propositions: 

‍

• High concurrency support- The platform could scale to handle thousands of customers & their requirements without performance degradation and additional costs.
• Autoscaling for peak workloads- The engine provided an elastic execution layer that auto-scales granularly to handle bursts of query traffic, then scales back down to save resources whenever possible. 
• Sub-second insights- e6data’s architecture also enabled near-real-time querying on fresh transactional data, so merchants could get up-to-the-second insights.

‍

Results

The impact was immediate and helpful. With e6data powering the dashboards, the fintech unicorn achieved:

‍

• 12x faster query completion across a mix of OLAP and near-real-time workloads
  
• 1000+ QPS sustained with p95 latencies <2s even for queries with complex joins and aggregations
  
• 50%+ lower TCO versus the previous solution, making the new analytics offering a new revenue stream
• The company now offers its customer-facing analytics as a premium feature for merchants — a new offering unlocked by e6data’s performance and cost-efficiency