OpenReplay is an open-source session replay and product analytics platform for teams that fix what users actually experience. It records user sessions with developer context so engineers, product managers, and support staff can watch issues happen instead of reconstructing them from reports.[1][2]
Quick verdict
Product and engineering teams that debug real user sessions benefit most from OpenReplay. Its pixel-perfect replays carry console, network, and funnel context, and the free self-host tier keeps session data on your infrastructure. Teams without operations capacity or needing SSO on a free plan should choose a managed vendor instead. If you can run Docker or Kubernetes and want the data in-house, OpenReplay is the conditional pick; otherwise start with its Serverless cloud trial.[1][2][3][4]
Capabilities
Pixel-perfect session replay Replays reproduce clicks, scrolls, and interactions exactly as the user saw them, with a timeline of user activity for jumping to moments of friction.[1]
DevTools context per session Each replay carries console logs, network activity, and performance signals, so an error report becomes a watchable session with backend context attached.[1][2]
Co-browsing for live support Support agents can join a user's live session inside the app to see the problem as it happens rather than trading screenshots.[1]
Funnels and journey analysis Conversion funnels and journey views show where users drop off across multi-step flows, connecting replay evidence to product decisions.[1]
Heatmaps of real interaction Aggregated click and scroll heatmaps reveal which page areas attract attention and which go ignored, grounding redesign choices.[1]
Product and web analytics Usage analytics sit next to replays so teams can move from an aggregate metric dip to the individual sessions behind it.[1]
Mobile and web SDKs Trackers cover JavaScript, iOS, Android, and React Native, with a beta Flutter SDK, so web and mobile sessions land in one place.[1]
Self-hosted or managed cloud Run it free on your own infrastructure via Docker, Kubernetes, Ubuntu, or source builds, or pay for managed Dedicated or usage-based Serverless cloud.[1][2]
Who will love it
Product and engineering teams that debug reported UX issues: replay plus DevTools context turns vague bug reports into reproducible sessions.[1][2]
Operators with data-residency requirements: free self-hosting with Docker, Kubernetes, or Ubuntu keeps session data on their own infrastructure across 50 residency regions.[1][2]
Support-led teams wanting live context: co-browsing lets agents see the customer's session instead of reconstructing it from descriptions.[1]
Who should avoid it
Teams without operations capacity should weigh the managed tiers: Docker deployment is marked experimental and self-hosting still needs a remote server you maintain.[1]
Budget-sensitive teams needing SSO or long retention should cost the paid tiers: single-sign-on, custom retention, and dedicated support sit behind Dedicated and enterprise plans.[1]
Teams wanting a permissively licensed core should read the license split first: the default is AGPLv3 with MIT parts and a separate proprietary ee/ directory.[1][2]
Positioning
OpenReplay is defined against hosted replay vendors: the same tracker-capture-replay loop, but deployable on operator infrastructure with an open-source core.[1][2]
Its closest open-source peer is Highlight, which pairs session replay with error monitoring and Docker self-hosting; both cover the install-tracker, capture, replay debugging workflow. Highlight's hosted future is unsettled since highlight.io redirects to LaunchDarkly.[1][2]
Questions worth asking
What is the best OpenReplay alternative for non-technical users?
A managed specialist fits better: a hosted replay vendor for hands-off operation, since OpenReplay's value assumes SDK instrumentation and either self-host operations or a paid cloud tier. Highlight, the closest open-source peer, shares the self-host tradeoff, so it is not the non-technical answer either.[1][2]
Is OpenReplay really no-code?
How much does OpenReplay cost per month?
Self-hosting the software itself is free; you pay only for your own infrastructure. Managed Dedicated starts at $199 per month billed hourly, and Serverless cloud is usage-based on recorded sessions. Pricing checked October 12, 2026.[1]
Can you self-host OpenReplay?
Sources and editorial review
Source research. Reviewed on October 12, 2026. Written by Sanjib Sah. AI assisted; reviewed by a human editor.
We have not run or tested this product.
- OpenReplay homepage
official · Observed October 12, 2026
Official homepage: self-hosted session replay and product analytics positioning; Self-Host, Dedicated, and Serverless paths; GitHub org link.
- OpenReplay pricing
official · Observed October 12, 2026
Pricing: free self-host; Dedicated from $199/mo billed hourly with SSO, custom retention, 50-region residency; usage-based Serverless; enterprise SAML/SCIM path.
- OpenReplay session replay
official · Observed October 12, 2026
Session-replay feature page: pixel-perfect replays of clicks, scrolls, and interactions; official compare pages against FullStory, LogRocket, PostHog, Hotjar.
- openreplay/openreplay repository
official · Observed October 12, 2026
Repository openreplay/openreplay: monorepo for self-hosted session replay, cobrowsing, and product analytics; multi-license layout.
- OpenReplay LICENSE
official · Observed October 12, 2026
LICENSE text: AGPLv3 default, MIT parts, separate ee/ enterprise license; copyright Asayer Inc dba OpenReplay.
- OpenReplay Docker deployment docs
official · Observed October 12, 2026
Docs deployment guide: Docker Compose (experimental), Kubernetes, Ubuntu, from-source paths; JS/iOS/Android/RN/Flutter SDKs; API, Integrations, MCP sections; v1.28.0 latest.
- Highlight homepage (redirects to LaunchDarkly)
official · Observed October 12, 2026
Competitor anchor: highlight.io redirects to the LaunchDarkly homepage; Highlight catalogue record documents open-source session replay with Docker self-hosting.
Compared with these products
No explicit proprietary baseline is recorded.