Can I run rtcstats.com locally?

This is one of the most common questions we get, and the answer is: partially, yes.

What you can run locally

The rtcStats ecosystem has two fully open-source components that you can run on your own infrastructure:

• rtcstats-js - the client-side SDK that collects WebRTC metrics from the browser. This runs entirely in the user's browser as part of your application
• rtcstats-server - the mediation server that receives data from rtcstats.js, stores it, anonymizes sensitive information, and extracts features into an SQL database

Together, these give you full data collection, storage, and basic feature extraction - all within your own environment. You own the data, you control the privacy, and you can query the database for BI and monitoring purposes.

The rtcStats enrichment engine

On top of the collected data sits the rtcStats enrichment engine, the layer where years of WebRTC expertise turn raw metrics into answers:

• Observations - our rules engine that automatically detects issues and flags interesting behaviors
• Deductions - the root cause analysis layer built on top of Observations
• Experience Score and other quality scores

Since v1.9 the enrichment engine no longer has to run on rtcstats.com. There are three ways to reach it:

• On rtcstats.com - upload a session and get the full analysis, the interactive viewer, and the AI Summary
• Through the enrich endpoint - POST /v1.0/enrich takes a dump you collected yourself and returns only the scores and Observations as clean JSON. The session is never stored. See How to enrich the sessions you collect yourself
• Inside your own network - the standalone rtcstats-enrichment server runs the same @rtcstats/rtcstats-processor engine next to your own stack, so no dump leaves your walls at all

What is exclusive to rtcstats.com

A few capabilities run only on the rtcstats.com SaaS platform:

• Visualization engine - the interactive dashboard with themed charts across Overview, Connections, Streams, Devices, and Logs tabs
• AI enrichment - AI-generated summaries of sessions
• Collaboration features - commenting, public sharing, multi-user accounts

So what does this mean in practice?

You can collect and store all your WebRTC metrics locally using our open-source tools. Many of our users do exactly that. From there you choose how much analysis you want and where it happens: enrich the sessions in place to get scores and Observations into your own database, and upload the individual session you actually need to look at to rtcstats.com for the deep analysis, visualization, and AI-powered insights. To make sure we keep the privacy of your users, rtcstats-server itself strips out all PII related information before uploading anything to rtcstats.com.

Think of it this way:

• Open source = your data collection and storage backbone
• Enrichment engine = the analysis, on rtcstats.com, through the enrich endpoint, or inside your own network
• rtcstats.com = the interactive dashboard, AI summaries, and collaboration

Can I get the enrichment engine on-premise?

Yes. The rtcstats-enrichment server is a small REST service that runs the same scoring engine rtcstats.com runs, inside your own infrastructure. It takes a WebRTC dump and returns the Experience Score, the audio, video and connectivity scores, and the Observations as JSON. No session is stored, and nothing leaves your network. Point your rtcstats-server at it, or call it directly.

It is not part of the open-source stack you can clone from GitHub today, so contact us if on-premise analysis is a requirement for your organization and we will work out access with you. If you want to see the response shape before committing to a deployment, POST /v1.0/enrich on the hosted API returns exactly the same payload - see How to enrich the sessions you collect yourself.

The interactive dashboard and AI summaries remain rtcstats.com cloud features either way.

See also

• How to enrich the sessions you collect yourself - the enrich endpoint and the standalone server
• rtcStats architecture - how the three layers work together
• Can rtcStats be used in closed environments?
• Data privacy and compliance