Error tracking
for the early days.
For the products that haven't taken off yet — and the teams still figuring it out. Simple setup, honest pricing. $9/month, flat. When you grow past us, that's a good thing.
Three lines of code. That's the setup.
$ npm install tinymon import { init, capture } from 'tinymon'; init({ apiKey: process.env.TINYMON_KEY, env: 'production', }); // that's it. errors auto-capture. // manual capture if you need it: capture(err, { user: userId });
One screen. Every error.
TypeError Cannot read properties of undefined
47: const profile = await getProfile(user.id); 48: if (!profile) throw new NotFoundError(); 49: return res.json(profile); at getUserHandler handlers/users.ts:47 at middleware lib/middleware.ts:23 at Router.dispatch lib/router.ts:118
Made for the early days.
Enterprise tools
- Per-seat pricing. Your bill grows with your team.
- Feature depth. Distributed tracing, profiling, RUM — genuinely useful at scale.
- Enterprise-ready. SSO, SAML, audit logs, compliance certifications.
- 24/7 support contracts. For when uptime is revenue.
- Annual contracts. Procurement expects them.
- The right tool once you have a team and a budget that supports it.
tinymon
- One price. $9/month. Not per seat. Not per project.
- The basics, done well. Error capture, grouping, stack traces, alerts.
- No plan tiers. Everything is in the $9 plan.
- Email & Slack alerts. When something new breaks. Never noise.
- Monthly, cancel anytime. No contracts.
- The right tool until you need more — and then please go get more.
Three things. Nothing else.
Your app throws.
The SDK grabs the stack trace, captures breadcrumbs from the last 30 seconds, and ships it to tinymon. Under 1ms of overhead. No sampling, no queuing tricks.
You get pinged.
Email or Slack, within 10 seconds. Only for new errors, or for regressions — we do the dedupe so you don't have "inbox of 400 identical notifications" syndrome.
You fix it.
Click through, read the stack trace, see the breadcrumbs, ship the patch. Mark resolved. tinymon auto-detects if it comes back. No ceremony, no post-mortem template.
One plan. One price.
Everything, every month.
- Unlimited projects
- 50,000 errors per month
- Email + Slack alerts, instant
- Sourcemaps, breadcrumbs, releases
- Full error history
- No seats, ever — invite anyone
- Cancel any time, no questions
Commit for a year, save $18.
- Everything in monthly
- Save $18 vs monthly
- Priority support
- First-month access
Questions you're probably about to ask.
How is this different from the bigger error tracking tools?
The full-featured observability platforms are genuinely great — they just weren't designed for teams still figuring out if their product will exist in six months. Tinymon does the core of what they do (error capture, grouping, stack traces, alerts) at a price that works when you're early. Distributed tracing, profiling, and the rest — the bigger tools do that better. When you grow to where you need all that, please go use them. Until then, we're here.
What happens if I exceed 50,000 errors in a month?
First, we'll email you — you probably want to know. We'll also rate-limit and sample so you never get a surprise bill. Usually when error volume spikes, something's actually wrong: a runaway loop, a bad deploy, a regression. We'd rather help you track down what's happening than push you away over it. If it becomes a regular pattern and your real volume is just higher, reach out and we'll figure out a fair arrangement together.
What languages are supported?
JavaScript and TypeScript at launch (Node, browser, Deno, Bun, Cloudflare Workers). Python, Go, Ruby, and Rust are on the roadmap. Any language with HTTP can send errors via the REST API.
How long is my data kept? Can I self-host?
We keep your error data around for as long as we reasonably can — plenty of history to spot patterns and trends. If your volume is unusually high and storage becomes a real cost on a specific account, we'll reach out before changing anything and work it out with you directly. Self-hosting is on the roadmap — when it ships, it'll be a single Docker image. No Kubernetes, no Helm, no Terraform.
Joining the waitlist — what's the commitment?
None. Just your email. You'll get a note when early access opens, and early users get the $9 founding price. You pay when you're ready.
There's a gap in error tracking. The full-featured tools are excellent — distributed tracing, session replay, deep performance insights. But their pricing starts where most early-stage teams can't. If you're shipping a product that hasn't taken off yet, "starts at tens of dollars per developer" adds up fast.
Tinymon is built for that stage. Low cost. Honest pricing. The basics done well, so you can catch errors before your users do — without worrying about the bill.
When you outgrow it, go use the bigger tools. They're worth it at scale. Until then, we're here to help you get there.