Frequently asked questions
What RewindRewind does, what it costs, what happens to your data, and where it stops. If something is missing, email hello@rewindrewind.com and it will get answered here.
The basics
What is RewindRewind?
A hosted service that records what your app does and what breaks. Tracked events, exceptions, page visits, health rules, metrics, and uptime checks all live in one place, on one bill.
Who is it for?
Teams running web apps and web games who want to see business health and errors without running an observability platform. It is built for small and mid-size teams, not for large infrastructure estates.
What is a signal?
One tracked action such as cart.checkout, or one captured exception. Both cost the same, so there is nothing to work out about which kind of data is cheaper to send.
How is this different from an error tracker?
An error tracker shows you what broke. RewindRewind shows what broke next to what people were doing, so you can tell whether a spike in errors is costing you orders or is noise from a third-party widget.
Does it replace my analytics tool?
Only if your reporting needs are light. There are no funnels, cohorts, or retention reports. You get an event timeline, an API, health rules, and metrics.
Setup
How do I get started?
Create an account, create a project, then open Setup on the project to copy its keys. The browser SDK is one script tag in the head. Server SDKs are one configure block. Most people are sending data in a few minutes.
Which languages and frameworks are supported?
Browser JavaScript, Node and Bun, Ruby and Rails, Python, and Go. Anything else can post JSON to the V1 API directly.
Is there a build step or an npm account to set up?
No. Every SDK installs straight from rewindrewind.com, and the browser SDK is a plain script tag with no bundler involved.
Can I keep the Sentry SDK I already have?
Yes. RewindRewind accepts the Sentry envelope format, so you can point an existing Sentry DSN at RewindRewind instead of swapping SDKs.
Why are there two kinds of keys?
The public project key starts with rrpub_ and authenticates ingestion, so it is safe in browser code. The admin key starts with rr_ and is a secret for the management API and CLI. You can also limit ingestion to specific browser origins and IP addresses.
Can an AI agent set this up for me?
Yes. Run the CLI with npx github:rewind-rewind/rewindrewindcli. The init command finds the project and prints setup for front-end exceptions, back-end exceptions, and events, and the verify command sends test data and confirms it landed. The file at /llms.txt describes the service for agents.
Pricing and billing
How much does it cost?
Every account gets 100,000 signals a month at no charge. Signals past that cost 20 cents per 1,000, which is $200 per million. There is no minimum and no monthly fee.
Is the free tier limited?
Only in volume and history. Free accounts get every feature, unlimited team members, unlimited projects, and 90 days of history, with no card required.
Do you charge per seat or per project?
No. Add your whole team and as many projects as you want. You pay for signals and nothing else.
Do page visits and uptime checks cost anything?
No. Visits are stored as daily counters rather than events, and uptime checks are not metered. Only tracked events and captured exceptions count as signals.
What happens when I use up the free signals?
Ingestion pauses and we email you, and warnings go out on the way there at 50, 80, and 95 percent, so it does not arrive as a surprise. Nothing is charged without a card on file. Add a card to keep sending past the included 100,000.
Can I get a surprise bill?
The spending cap is on by default and you choose the number. Usage emails land at 50, 80, and 95 percent of it, and at 100 percent ingestion pauses rather than running up the tab. Raise the cap to start again straight away.
When am I billed?
At the end of each monthly billing period, for exactly the signals used in it. The invoice is emailed to you.
Can I cancel?
Yes, from billing settings, at any time. The account goes back to free rules: the included monthly signals and 90 days of history. Older history is removed after a warning and a grace period.
Your data
How long is data kept?
Free accounts keep 90 days of history. Paid accounts keep their history for as long as they stay.
Where is the data stored?
On Cloudflare. Each project gets its own SQLite database inside a Durable Object, so one project's volume cannot slow another one down. Older rows move into compressed cold storage and still read back through the same queries.
Can I export my data?
Yes. The management API and the CLI export events and issues, and /openapi.json is the full reference for both.
Can I delete data?
Yes. Deleting a project removes its telemetry, and retention removes aged data on its own schedule without you asking.
What about personal data?
You decide what gets sent. Identity fields such as a user id or email are optional, and event properties are yours to choose. If you do not want something stored, do not send it.
Features and limits
How are exceptions grouped?
Exceptions with the same fingerprint become one issue that keeps its stack trace, occurrences, affected users, raw payload, comments, and lifecycle history. Resolve, ignore, or snooze an issue, and it reopens on a new release or a surge in recurrences.
Will I get alerted when something breaks?
Yes. New issues email you, with settings per project and per environment, a repeat threshold, and a cadence. You can silence a project for yourself without changing it for the team.
Are stack traces readable?
Yes, if you upload a source map for the release. The browser SDK also attaches breadcrumbs for the navigations, clicks, console output, and network calls that led up to a crash.
What are health rules and metrics?
A health rule turns a query over your own events into a pass or fail check, such as orders per hour staying above a floor. Metrics chart the same kind of query over time. Both are evaluated every five minutes.
Does it monitor uptime?
Yes. Add an HTTP or HTTPS URL and RewindRewind sends a GET request every five minutes. Two consecutive failures open an incident and email account admins, and a successful check closes it. You get 24-hour and 30-day percentages and a 90-day check log.
Does it do session replay, tracing, or feature flags?
No, and there are no plans to pretend otherwise. If you need those, Sentry, PostHog, or Datadog are the better fit. The comparison pages say plainly where each one wins.
Can I self-host it?
No. It is a hosted service, and every SDK sends to the collector at rewindrewind.com. There is no self-hosted distribution to install.
Accounts and support
How many people can I add?
As many as you like, on any plan, as admins or members. Team size never changes the price.
Can I run several projects in one account?
Yes. Each project has its own keys, settings, health rules, and data, and there is no limit on how many you create.
How do I get help?
Email hello@rewindrewind.com. The docs cover setup for every SDK, and the API reference is at /docs/api. Email support is included with paid accounts.
Still deciding?
The pricing page has a calculator for your own volume, and the comparison pages put RewindRewind next to the tools people usually weigh it against.