How We Deliver Full-Stack Visibility with HyperDX

At FreeTech, we build and manage web applications for fast-moving startups and small businesses. Our clients rely on us to deliver resilient applications with minimal bugs or issues. Clean code, unit testing, and QA are essential, but bugs and edge case errors inevitably occur in production. Over the years, I’ve personally used a range of tools for analytics and real-time user monitoring (RUM), including Datadog, NewRelic, LogRocket, and fullstory—each with their own strengths and limitations. About a year ago, we discovered HyperDX and haven’t looked back. Below, learn why in 2024 we chose to standardize HyperDX for full-stack observability across all our projects—and how it helps deliver business value for each of our clients.

The Big Picture: Alerting & RUM (Real-time User Monitoring)

When delivering an application to hundreds or thousands of users, it’s essential to have real-time user monitoring (RUM) and alerting to give developers critical insights into production activity. When an error occurs, developers need to be notified immediately to ensure a fast resolution. Tying together a user's frontend session—visually, with cursor tracking—and backend API logs into a single unified interface enables rapid debugging and issue resolution.

In the past, I primarily used Datadog at the enterprise level and LogRocket for smaller clients. At FreeTech, we consistently explore open-source alternatives that offer equal or greater functionality at a lower cost to deliver more affordable solutions for our clients. In early 2024, I discovered HyperDX and scheduled a meeting with one of their founders, Michael Shi. HyperDX offered everything we needed—and more—for less than 10% of the cost compared to Datadog and LogRocket.

From RUM, metrics, and alerting to ease of setup for new clients, HyperDX was an ideal fit for our needs. In March 2025, HyperDX was acquired by ClickHouse, and we’re excited to continue using their platform under ClickHouse's ownership.

How We Leverage HyperDX for Full-Stack Observability

At FreeTech, we use several core HyperDX features daily to support our clients:

• Real User Monitoring (RUM) & Session Replays — Visual session tracking for real-time user insights and fast debugging.

• Dashboards — Custom developer-oriented dashboards to monitor key metrics like client-side errors, API request volume and CPU utilization.

• Live Alerts — Immediate alerts for server-side and client-side errors, as well as anomaly detection.

• Infrastructure Metrics — In addition to cloud-native billing and resource utilization alerts, we use HyperDX to track CPU, memory, and serverless infrastructure performance.

Whether we’re working with React, Next.js, Express, or Firebase, we integrate HyperDX across the full stack—from browser to backend. The result: complete context around every error and performance issue, without needing to switch tools. We've also built an internal FreeTech library that allows us to deploy HyperDX out-of-the-box for each new client.

Real-Time Response

The moment something breaks, we’re notified. We’ve integrated HyperDX with Slack and SMS alerts to instantly catch server and browser errors—often before clients even notice.

When a client does report an issue, we don’t rely on guesswork. Using RUM data and session recordings, we review the actual user session in real time to quickly identify and resolve the problem.

Seamless End-to-End Correlation

In full-stack applications, tracing issues across fragmented across frontend and backend systems can be difficult, particularly when trying to associate user-reported problems with underlying system behavior. Traditional setups often force teams to manually cross-reference browser logs, API traces, and database events without a reliable link between them.

Utilizing HyperDX we're able to seamlessly associate frontend logs and session recordings with backend server logs, allowing us to efficiently track down and resolve issues or improve user experiences. For authenticated sessions, we inject identifiers—such as userId and email—into frontend events. This enables us to accurately correlate user activity with backend API requests and system logs, providing our engineers with complete visibility across the stack.

This allows us to:

• Correlate frontend performance issues (e.g., delayed client-side rendering or interaction stalls) directly with backend API traces in Express or serverless functions.

• Follow a user’s session from the browser, through middleware and service layers, into database operations in PostgreSQL or Firebase, all tied to a unified userId.

• Quickly filter and investigate production incidents by user, tenant, or organization without manually stitching logs across systems.

HyperDX aggregates all session, trace, and log data under a common timeline, enabling our team to perform targeted, efficient root cause analysis without context switching between multiple tools.

This structured correlation significantly reduces investigation time, supports faster client support responses, and improves overall platform reliability.

Fine-grained Permissions and Unlimited Users

A critical requirement for our observability stack was the ability to manage fine-grained permissions within a single environment. We needed precise control over which users could access specific services, ensuring that engineers and clients could only view data related to their own environments.

HyperDX provides a simple method to restrict access through service tags, allowing us to restrict access at a granular level based on service name. Using a company name based naming convention, we create a new group for each of our clients using * wildcard, allowing us to easily provide access to engineers and company stakeholders.

Another critical requirement for our observability tool was unlimited users without additional licensing or per-seat fees. This allows us to scale access across engineers, QA teams, product managers, and client stakeholders without cost increases—enabling us to offer this service free with all of our service packages.

Open Source and Cost-Effective

HyperDX is built on top of two core open-source technologies: ClickHouse high-speed analytics storage and OpenTelemetry for tracing, logging, and metrics collection across distributed systems. Users can self-host HyperDX on virtual machine starting at $10/month, with the recommend minimum 4 GB of RAM and 2 CPU cores. If you're comfortable managing occasional server maintenance tasks like SSL certificate rotation, their self-hosted option is straightforward to deploy.

While HyperDX offers a fully open-source deployment option—which we would typically gravitate toward—we choose to support the HyperDX team by using and paying for their managed cloud platform. Their hosted solution allows us to deliver full-stack observability to our clients without the operational overhead of maintaining telemetry infrastructure, while supporting the ongoing development of an exceptional product.

Client Experience: Datadog Nightmare

In late 2023, I was working with a smaller FreeTech client who had been using Datadog. At the time, I wasn’t yet aware of HyperDX. During the project, a server was left with debug logging enabled in production by one of their team members, resulting in a Datadog bill of approximately $3,000-4000. Datadog refused to adjust the charges, and the client had to absorb the cost while FreeTech and the client’s development team corrected the logging configuration to prevent it from happening again.

Had this client been using HyperDX, their bill for the same log volume (nearly 1 TB) would have been approximately 10% of the cost (HyperDX Pricing).

Included with All of Our Service Packages

Thanks to HyperDX’s cost efficiency, we’re able to include full-stack observability free of charge with all of our service packages. Previously, clients would pay $100+/month for services like LogRocket or FullStory, or $1,000+/month for Datadog or New Relic for an equivalent service. By efficiently managing our logging infrastructure, we keep our total observability costs typically below $100/month—across more than a dozen applications with hundreds or thousands of daily active users.

HyperDX is now a standard part of our deployment stack. Whether we’re building a new application or onboarding an existing one, we implement end-to-end observability from day one—so our clients can move faster, with greater confidence.