Realtime Analytics Dashboards for a Multi-tenant Application

Location

USA

Client

The customer is a US-based care management platform provider offering multi-tenant web applications to healthcare administrators and stakeholders to improve care coordination and decision-making.

Challenge

The customer needed to enhance existing legacy monthly reporting capabilities historically managed by a separate team by enabling real-time analytics inside their multi-tenant application. The core requirements were to provide real-time, anonymized reporting with strict row-level security enforcing access control per tenant and various user roles. They also needed functionality for filtering, drill-downs, varied chart types, and downloadable reports, all operating without impacting production database performance and available on-demand. Updates to dashboards had to be deployed with zero downtime despite constrained enterprise IAM policies and network restrictions.

Partner Solution

To meet these requirements, the partner implemented direct connectivity between AWS QuickSight and the customer’s RDS for PostgreSQL cluster, specifically using a read replica to offload complex queries and avoid production load impact. The dashboards were built with AWS QuickSight’s embedded analytics features, supporting filters, charts, tables, histograms, widgets, and callouts. Row-level security was enforced within QuickSight datasets using tenant-specific filters and parameters complying with IAM constraints.

Role-based access was managed through a combination of application-level mappings of user roles to dashboard ARNs and infrastructure-level IAM roles enabling fetching tenant-specific dashboards via QuickSight APIs and anonymous URLs. The datasets and dashboards were templated via JSON export/import, enabling easy replication and zero-downtime updates across tenants. The solution was deployed first to UAT for business user acceptance, followed by a production rollout.

Results and Benefits

The solution delivered real-time, on-demand dashboards to five tenants without downtime during updates in comparison to legacy monthly reports. Business users approved the MVP with minimal rework. Interactive features like export and advanced filtering enhanced user productivity. The architecture was designed for straightforward scaling of QuickSight authors and readers as user volume grows.

A total cost of ownership (TCO) model isolated AWS QuickSight subscription costs from supporting infrastructure expenses, enabling clear budget management and forecasting. Initial deployment avoided reserved instance commitments to maintain flexibility amid rapid scaling expectations, a strategy validated by project success.

Please provide your contact details, and we will get back to you promptly.

First name*

Last name*

Email*

Job title

Company*

Phone

What would you like to discuss?*

How did you learn about DataArt?

What is your role?

I agree to receive DataArt's monthly marketing newsletter.

I have read and agree to the Privacy Policy*