Error trackers, language-focused APM profilers and open-source trackers you host yourself are the three real replacements for AppSignal. Pick the type by what you lean on most today: exception alerts, slow-request traces, or keeping error data on your own servers.
Two developers leave AppSignal in the same week. One runs a single Rails app, cares about slow endpoints and background jobs, and wants a profiler that reads Ruby like a native speaker. The other runs a Node API, a Python worker and a mobile app, and mostly needs to know the second something throws. They'll end up with very different tools, and both will be right. That's why this list mixes pure error trackers with performance profilers instead of pretending they're one category.
AppSignal bundles errors, performance, logs and uptime into one product across a long list of languages. Most replacements do one or two of those jobs very well, so the first decision isn't which brand. It's which of those jobs you'd miss on Monday morning.
Error trackers, profilers and self-hosted trackers: strengths and drawbacks
Pure error trackers catch exceptions, group duplicates into one issue, attach the stack trace and the request that caused it, and alert the right person. Their strength is breadth: they usually support many languages and front-end frameworks, so a mixed stack can sit in one place. Some have added session replay, release tracking or suggested fixes on top. The drawback is that a fast, quiet app that's slowly getting slower won't trip them. You'll know when something breaks, not when a page that used to take a moment now takes several.
Language-focused APM tools and profilers go the other way. A profiler built for Ruby and Rails understands controllers, ActiveRecord queries, view rendering and background jobs, so the trace reads like your code rather than a generic span tree. That's the upside: you see the repeated query inside a loop, the slow partial, the job that's ballooning. The downside is scope. A Rails-only tool is no help for the Go service a teammate starts next quarter, and some keep error tracking thin or leave it out.
Open-source trackers you can self-host appeal to teams who want error data on their own servers or a bill that doesn't jump with traffic. The ones that accept Sentry client SDKs are the easiest to try, because the instrumentation you add is the same either way. The trade is your time: upgrades, backups, a database that grows, and nobody else to call when the ingest queue backs up at night. Hosted plans from the same projects exist for teams who like the code but not the pager.
If you used AppSignal mainly for its all-in-one dashboard, no single type here replaces all of it. I'd happily pair a profiler with an error tracker, because each tool then stays sharp at one job.
Built around Ruby, Rails and their neighbours
Skylight
Rails teams chasing slow endpoints get a profiler that only speaks Ruby, with traces organised the way a Rails request runs.
RoRvsWild
A one-person Swiss company makes this Ruby on Rails monitor, which combines APM and error tracking without asking you to become a DevOps engineer.
Scout Monitoring
APM with traces, logs and errors that reaches beyond Rails into Python, PHP, Elixir and Node.js applications.
Honeybadger
Error tracking first, with logging, uptime checks, cron monitoring and lighter performance views for Rails, Elixir, PHP and Python apps.
Error monitoring for mixed stacks
Sentry
A hosted product that puts error monitoring, tracing, session replay and logs for web, mobile and backend code in one place.
Rollbar
Groups errors, plays back the session around them and can draft a fix for a developer to review.
Bugsnag
Mobile apps sit alongside web apps here, with error and performance monitoring built for teams shipping to both.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Airbrake
For error monitoring tied to deploys, this tracker adds performance insights and shows which release introduced a problem.
GlitchTip
Open source and compatible with Sentry client SDKs, it covers errors, performance and uptime, either self-hosted or as a hosted service.
How the bill grows as traffic does
Monitoring tools don't all count the same thing, and that's where a cheap-looking plan turns expensive. Some charge by requests or transactions: every web request and background job the agent sees uses up the allowance, so a busy but healthy app costs more than a quiet one. Some Rails-focused tools bill on requests and jobs rather than seats. Error trackers usually count events or occurrences instead, which means a quiet app is cheap and a bad deploy that throws the same exception thousands of times can eat a month's quota in an afternoon. Others add a per-user element on top of the event volume.
So before comparing tiers, pull last month's numbers from AppSignal: total requests, background jobs and the count of error occurrences, including your worst day. Then hold each candidate's pricing page against those figures, not against the headline plan. Check what happens when you go over: does the tool stop accepting data, sample it, or bill the overage? Free plans and trials vary from vendor to vendor, so confirm the current terms on the page itself. Self-hosting removes the vendor bill but swaps it for server costs and your hours, which are easy to forget in a spreadsheet.
For the wider picture of tracing tools beyond this list, my guide to the best APM tools goes further into full-stack monitoring.
An hour on a staging app
Don't decide from feature grids. Pick two candidates, one of each type you're torn between, and install both agents on a staging copy of your app at the same time. Then do four things. Trigger a deliberate exception from a controller and from a background job, and time how long each tool takes to alert you and whether the alert lands somewhere you'll see it. Hit your slowest page a few dozen times and look for the database query you already know is the culprit: if the trace doesn't point at it, the profiler isn't earning its place. Deploy a small change and check the tool marks the release. Finally, open the issue view on your phone, because that's where you'll read it at night.
While you're there, look at what the agent costs your app. A monitoring agent that adds noticeable memory or slows boot time on a small dyno is a poor trade, whatever the dashboard looks like.
Do you make error tracking or APM tools? Tool makers can ask to be considered for a featured place on this list.
Moving off without a blind spot
Run the new tool alongside AppSignal for a week or two rather than switching in one deploy. Re-create your alert rules by hand, including who gets paged and when, because those rarely carry over between tools. Re-add deploy tracking in your CI pipeline, upload source maps if you have a front end, and move uptime and cron checks somewhere before you cancel, and check whether the new tool covers them at all. If logs lived in AppSignal too, decide where they'll go next; a dedicated tool from my list of log management tools may suit better than squeezing them into an error tracker.
Only cut the old agent once you've seen the new one catch a real error and a real slow request in production. Then remove the gem or package, so you aren't paying two vendors to watch the same app.
Want your tool on this list? Featured spots and reviews here are paid partnerships. Here's how to partner with me.


