Which error tracking, application monitoring and observability services still have a genuinely free tier in 2026. 25+ compared — Datadog, Grafana Cloud, New Relic, Better Stack, Sentry, Checkly, SigNoz, HyperDX, Elastic — free data ingest, retention and APM limits, and what monitoring costs at 10/50/100/500 hosts.
Free Tiers for Error Tracking, Monitoring & Observability 2026
Published 2026-04-03 · Reviewed 2026-09-12, corrections outstanding · Figures compiled 2026-04-03 · 12 of 13 figures in the “Pricing Change Timeline” table come from our records for 1,580 developer tools · 25+ monitoring services compared
Quick verdict:Grafana Cloud offers the best overall free tier for monitoring — 10,000 metrics series, 50GB logs, 50GB traces, and 14-day retention with full dashboarding and alerting. New Relic has the most generous unified platform ingest at 100GB/month with full APM, but limits you to 1 full platform user. Axiom offers 500GB/month raw ingest free. Better Stack is the best combined uptime + logging starter. For self-hosted: SigNoz is the modern all-in-one alternative; Prometheus + Grafana is the most battle-tested.
The observability cost trap: Monitoring is the #1 developer spending complaint. Datadog bills can go from $0 to $10,000+/month because pricing compounds across per-host fees, separate product charges (APM, logs, synthetics are each billed independently), and cardinality penalties. At 100 hosts with APM and logs, Datadog costs ~$5,500/mo while Grafana Cloud Pro runs ~$300-800/mo. The worst trap: Datadog’s free tier retains metrics for only 1 day — you literally cannot investigate yesterday’s incident.
Free tier: 5 hosts, 1-day metric retention, 500MB logs (1-day retention), 2 alert monitors. The most feature-rich platform but the most restrictive free tier. The 1-day retention makes the free tier practically useless for anything beyond real-time dashboards — you can't investigate yesterday's incident or spot weekly trends. APM, Synthetics, and SIEM are all paid add-ons.
Free tier: 10,000 metric series (Prometheus), 50GB logs (Loki), 50GB traces (Tempo), 14-day retention, 3 users, unlimited dashboards, alerting included. By far the most generous free tier in full-stack observability. Built on open-source Grafana, Prometheus, Loki, and Tempo — so your queries and dashboards are portable. 3-user limit is the main constraint for teams.
Free tier: 100GB/month data ingest across all telemetry types, 1 full platform user + unlimited basic users (read-only dashboards). 8-day metric retention in NRDB, 30-day log retention. Full APM, infrastructure, and browser monitoring included. The catch: the 1 full user limit means only one person can create dashboards, set up alerts, or query data — a bottleneck for any team larger than one.
Free tier: 500GB/month ingest with 30-day retention. Unlimited users. Supports logs, metrics, and traces. Modern query language (APL) with a fast columnar engine. The most generous raw ingest allowance on this list. Good for startups that generate a lot of telemetry data and need longer retention.
Free tier: 500MB/day logs (7-day retention), 5 hosts for infrastructure monitoring (30-min metric retention), basic alerting. Offers both logs and infrastructure in one platform. The 30-minute metric retention is extremely short — only useful for live monitoring.
Free tier: Unlimited (self-hosted, AGPL-licensed). Elasticsearch for search and analytics, Kibana for visualization, Elastic APM for application performance monitoring, Filebeat/Logstash for log ingestion. The most feature-complete open-source observability stack. Supports logs, metrics, traces, and APM in a single platform. Cloud (Elastic Cloud) has a 14-day free trial, then ~$95/mo for the cheapest tier. The self-hosted cost is pure infrastructure. Downside: resource-hungry — Elasticsearch needs significant RAM (16GB+ recommended for production).
Free tier: 1 host, 1-day data retention, full-stack observability including APM, logs, and infrastructure. Modern OpenTelemetry-native platform. The single host limit makes it suitable only for development or very small projects.
Free tier: 1GB log ingest, 3-day retention, 10 uptime monitors (3-min checks), status pages. Combines logging with uptime monitoring. Clean UI. The log retention is short but the combined logging + uptime monitoring makes it a good starter choice for teams that want uptime + logs in one platform without setup complexity.
Error Tracking & APM
Specialized tools for catching application errors, performance issues, and user-facing problems.
Free tier (Developer): 5,000 errors/month, 50 session replays, 5M spans/month, 5 GB logs, 1 user. 30 days retention — the 90-day lookback is the Team plan. Supports 30+ platforms. Source map support, breadcrumbs, and issue grouping included. The de facto standard for error tracking. Self-hosted option available for unlimited scale.
Free tier: 5,000 events/month, 30-day retention, 1 user. Error monitoring with automatic grouping, deploy tracking, and real-time alerts. Strong REST API and 40+ platform integrations. Deploy tracking links errors to specific releases. The 1-user limit and lower event count vs Sentry makes it a secondary choice for most teams.
Free tier: 7,500 events/month, 7-day retention, 1 project, 1 user. Error monitoring with stability scoring, release health tracking, and breadcrumbs. Particularly strong for mobile apps (iOS/Android). Performance monitoring included on free tier (basic). The single-project limit is the main constraint.
Free tier: 5 browser checks (Playwright-based), 10 API checks, 1-min check interval from 20+ global locations. Monitoring-as-code: define checks in JavaScript/TypeScript, version in git, deploy via CLI. Integrates with Vercel, GitHub Actions, and Terraform. The only free synthetic monitoring tool with full Playwright browser automation — test real user flows, not just pings.
Key takeaway: UptimeRobot dominates on monitor count (50 free). Checkly is the best choice for synthetic browser testing (Playwright-based, monitoring-as-code). Hyperping has the fastest simple check interval (1 min). Better Stack combines uptime with logging for a two-in-one free tier. For most small projects, UptimeRobot's 50 monitors at 5-min intervals is more than enough.
Incident Management
Tools for on-call scheduling, incident response, and status communication.
Free tier: 5 users, unlimited integrations, on-call scheduling, email/SMS/push notifications, escalation policies. The industry standard for on-call management. Free tier is genuinely usable for small teams. No phone call alerts on free tier (email + push only for escalations).
Free tier: 5 users, Slack-native incident management, on-call scheduling, post-mortems, status pages. Modern UX built around Slack workflows. Strong for teams already centered on Slack for communication.
Cron & Job Monitoring
Dead simple monitoring for scheduled tasks, cron jobs, and background workers — alert when a job doesn't check in on time.
Key takeaway: Healthchecks.io is the clear winner — 20 free checks, 3 team members, 20+ integrations, and BSD-licensed self-hosted option. Cronitor and Dead Man's Snitch are more limited but have slicker UIs.
Self-Hosted / Open Source
Zero vendor lock-in, unlimited scale, but you own the infrastructure and operations.
Prometheus + Grafana
Cost: Free (infrastructure costs only). The industry-standard metrics pipeline. Prometheus scrapes and stores time-series metrics; Grafana visualizes them. Default 15-day retention (configurable). PromQL is the most widely supported query language in monitoring. Add Alertmanager for alerting. Operates at massive scale (millions of time series). Downside: no built-in log management — pair with Loki or ELK.
Cost: Free (infrastructure costs only). CNCF-graduated distributed tracing platform. OpenTelemetry native. Supports multiple storage backends (Elasticsearch, Cassandra, Kafka). Essential for microservice architectures. Pair with Prometheus for metrics and Loki for logs to build a full observability stack.
Cost: Free self-hosted (open-source, Apache 2.0 or AGPL depending on features). Single binary with metrics, logs, and traces powered by ClickHouse and OpenTelemetry. The closest open-source equivalent to Datadog — single pane of glass with correlated metrics/logs/traces, APM dashboards, and alerting. OpenTelemetry-native (no vendor SDK). Cloud tier starts at $199/mo. Best for teams wanting a Datadog-like experience without Datadog pricing.
HyperDX
Cost: Free self-hosted (open-source, MIT). Unified platform for logs, metrics, traces, and session replays. Correlates frontend session replays with backend traces — unique capability among open-source tools. OpenTelemetry-native. Built on ClickHouse. Cloud free tier available (limited). Best for full-stack debugging where you need to see exactly what the user did and how the backend responded.
Cost: Free self-hosted (unlimited). Cloud free tier: 5 nodes, 14-day retention. Real-time infrastructure monitoring with zero-config auto-discovery of 800+ integrations. Per-second granularity out of the box. Extremely lightweight agent (~1% CPU). Best for real-time infrastructure visibility without complex setup.
The self-hosted trade-off: Self-hosted monitoring eliminates per-host and per-GB costs but introduces operational burden. You need to run, scale, and maintain the monitoring infrastructure itself — which is ironic when monitoring is supposed to tell you when things break. SigNoz is the easiest all-in-one to self-host (single Docker Compose). Prometheus + Grafana is the most battle-tested but requires assembling multiple components. Elastic is the most feature-complete but resource-hungry. If you're a small team, start with Grafana Cloud free tier and self-host only when you outgrow it.
The Observability Cost Trap
Monitoring is the #1 developer spending complaint. Here's what it actually costs when you outgrow the free tier — at 10, 50, 100, and 500 host scale.
Annual Cost by Scale (Infrastructure + APM + Logs)
Scale
Datadog
Grafana Cloud
New Relic
Better Stack
Self-Hosted (Prometheus/SigNoz)
10 hosts
~$3,720/yr ($31/host/mo)
$0 (free tier)
$0 (within 100GB)
~$348/yr ($29/mo)
$0 (infra only)
50 hosts + APM + logs
~$27,000/yr
~$600-2,400/yr
~$2,400-4,800/yr
~$3,000-6,000/yr
$600-1,800/yr (infra)
100 hosts full stack
~$66,000/yr
~$3,600-9,600/yr
~$6,000-12,000/yr
~$7,200-12,000/yr
$2,400-6,000/yr (infra)
500 hosts enterprise
~$330,000/yr
~$24,000-60,000/yr
~$36,000-72,000/yr
~$36,000-60,000/yr
$12,000-30,000/yr (infra)
100GB/day logs (any scale)
~$61,200/yr (ingest + index)
~$18,000/yr
~$21,600/yr
~$14,400/yr
$1,200-3,600/yr (storage)
Why Datadog costs explode: Datadog uses per-host pricing ($15-31/host/mo for Infrastructure, $31-40/host/mo for APM) that scales linearly. But the real trap is that every feature is a separate add-on: APM, Log Management, Synthetics, SIEM, Database Monitoring, and Continuous Profiler are each priced independently. At 50 hosts with APM + logs, you're paying for 3 separate products. At 500 hosts, that's $330K+/year — more than many companies spend on the infrastructure being monitored. Grafana Cloud, New Relic, and Better Stack include APM and logs in base pricing. Self-hosted eliminates per-host costs entirely.
The cardinality trap: Beyond per-host pricing, Datadog charges $0.05/custom metric for high-cardinality metrics. A single Kubernetes cluster with 200 pods can generate 50,000+ custom metrics = $2,500/mo in cardinality charges alone. New Relic and Grafana Cloud don't charge per-metric on their free tiers. If you use custom tags heavily, cardinality fees can exceed your base monitoring bill.
The 1-day retention trap: Datadog's free tier retains metrics for only 1 day. This means you literally cannot: look at last week's CPU trends, compare this Monday to last Monday, investigate an incident that happened yesterday afternoon, or do any capacity planning. Most serious monitoring requires at least 7-14 days of retention. Grafana Cloud offers 14 days free; New Relic offers 8 days for metrics and 30 days for logs; Axiom offers 30 days.
By Use Case
Recommendations by Use Case
Full-stack observability on a budget → Grafana Cloud
Most generous free tier across metrics + logs + traces. 10K series, 50GB logs, 50GB traces, 14-day retention. Open-source core means your dashboards and queries are portable.
Maximum free data ingest → New Relic
100GB/month across all telemetry types. Full APM, infrastructure, and browser monitoring. Trade-off: only 1 full platform user, so one person owns all configuration.
Simple uptime monitoring → UptimeRobot
50 monitors at 5-min intervals. Dead simple setup. Email + webhook alerts. Free status page. Handles HTTP, keyword, ping, and port checks.
Error tracking for small teams → Sentry
5K errors/month on Developer tier. 30-day retention. 30+ platform SDKs. Source maps, breadcrumbs, and session replay. Self-hosted option for unlimited scale.
Unlimited scale, zero per-host or per-GB costs, and full data ownership. Add Loki for logs. Requires infrastructure expertise. The CNCF-blessed observability stack.
Best for startups (cost + features) → Grafana Cloud or Axiom
Grafana Cloud gives metrics + logs + traces free with 14-day retention. Axiom gives 500GB/month ingest with 30-day retention. Both avoid per-host pricing traps. Start with one, migrate to self-hosted when you reach scale.
Best for serverless (Lambda/Edge) → New Relic or Axiom
Serverless functions generate bursty, high-volume telemetry that per-host pricing can't handle. New Relic's 100GB ingest with no host counting, or Axiom's 500GB with unlimited users, fit serverless architectures where "hosts" are ephemeral. Datadog charges per-function invocation for serverless — avoid.
Best for enterprise (compliance, SLA) → Datadog or New Relic
When compliance (SOC 2, HIPAA, FedRAMP) and vendor SLAs matter more than cost, Datadog and New Relic lead. Datadog has the broadest feature set. New Relic has better cost predictability. Both offer enterprise support and dedicated infrastructure.
Best synthetic monitoring → Checkly
Playwright-based browser checks, monitoring-as-code in TypeScript, 20+ global locations, 1-min intervals. Test real user flows, not just endpoint pings. Version your monitoring in git alongside your application code.
Incident management starter → PagerDuty or incident.io
Both offer 5 free users with on-call scheduling. PagerDuty for traditional pager workflows; incident.io for Slack-native incident response.
Cron job monitoring → Healthchecks.io
20 free monitors with configurable grace periods. 20+ notification integrations. BSD-licensed self-hosted option. The de facto standard for dead man's switch monitoring.
Hidden Costs and Gotchas
Datadog's 1-day metric retention
The free tier retains metrics for just 24 hours. This makes it unsuitable for trend analysis, capacity planning, weekly comparisons, or investigating incidents from the day before. You'll be forced to upgrade to Pro ($15/host/mo) almost immediately for any serious use.
New Relic's 1 full user limit
Only 1 team member can create dashboards, configure alerts, or write NRQL queries. Everyone else gets basic (read-only) access. This creates a single point of failure — if that person leaves, you need to transfer full user access. At $99/user/mo for additional full users, adding a second costs more than some entire monitoring solutions.
Per-host pricing that scales linearly
Datadog ($15-31/host/mo) and Sematext charge per host. Auto-scaling infrastructure means your monitoring bill fluctuates with your cloud bill. A Black Friday traffic spike that scales you to 200 hosts also scales your Datadog bill to $6,000+/mo — and that's just infrastructure monitoring, before APM or logs.
Log indexing charges separate from ingest
Datadog charges for log ingestion ($0.10/GB) AND log indexing ($1.70/million events) separately. You can ingest logs for search without indexing, but queries are slower and you lose many features. At 100GB/day of logs, just ingest costs $300/mo — adding indexing can triple it.
Alert channel limitations
Many free tiers limit alert delivery to email only. No Slack, PagerDuty, or phone call alerts without upgrading. UptimeRobot supports webhooks on free, which is a workaround. Grafana Cloud includes all notification channels on free tier — another reason it leads.
Datadog custom metric cardinality charges
Custom metrics cost $0.05/metric/month. With Kubernetes labels, container IDs, and dynamic tags, a modest cluster can generate 50,000+ custom metrics without you noticing. That's $2,500/mo in surprise cardinality charges — often discovered only on the first bill. Use metric aggregation rules and allowlists to prevent runaway cardinality.
Better Stack's 3-day log retention
The free tier retains logs for only 3 days. For incident investigation that spans a week or more (common in production), you'll need to upgrade. The combined uptime + logging is convenient but the retention limit forces upgrades faster than standalone logging tools.
Self-hosted operational burden
Running Prometheus + Grafana + Loki at scale requires dedicated infrastructure: persistent storage, backup procedures, high-availability setup, and regular upgrades. The irony of self-hosted monitoring: when your monitoring infrastructure itself breaks, you have no monitoring to tell you. Budget 0.5-1 FTE for a production self-hosted monitoring stack above 50 hosts.
Data egress and API rate limits
Exporting data from monitoring platforms is often rate-limited or restricted on free tiers. New Relic limits NRQL query rate. Datadog's free API has lower rate limits. This makes it hard to migrate away or build custom integrations — classic vendor lock-in by design.
Pricing Change Timeline
Recent monitoring and observability pricing changes tracked in our index.
Date
Provider
Change
Impact
discovered Sep 27, 2026 · effective date unknown
Svix
Svix cut its Free plan from 200 to 50 messages a second and from 30-day to 7-day payload retention, and moved the 99.9% uptime SLA to a new Basic plan from $20 a month. Free's 50,000 messages a month, beyond which messages were billed, now reads "Up to 50k messages per month". The pricing page changed between the Internet Archive's 2026-09-14 capture and our read on 2026-09-27. Source ↗
MEDIUM
effective between Sep 9, 2026 and Sep 26, 2026
AppSignal
The free plan now offers 30,000 requests/month instead of 50,000. Source ↗
MEDIUM
discovered Sep 7, 2026 · effective date unknown
Better Stack Logs
The free tier now offers 3 GB logs retained for 3 days, 3 GB traces retained for 3 days, 30 GB metrics, and 3 GB web events retained for 3 days. It also includes 10 monitors & heartbeats, 1 status page, Slack & e-mail alerts, 100,000 exceptions per month, and 5,000 session replays. Unconfirmed. We could not check this change against an archived copy of Better Stack Logs's page, so it does not count toward the rating.Source ↗
LOW
discovered Sep 3, 2026 · effective date unknown
LastPass
The free tier now has dark web monitoring and is limited to 1 device type. Unconfirmed. We could not check this change against an archived copy of LastPass's page, so it does not count toward the rating.Source ↗
LOW
discovered Aug 28, 2026 · effective date unknown
Tomorrow.io Weather API
The free tier now includes access to 60+ data layers, historical data, and various APIs (Weather Forecast, Historical Weather, Weather Maps, Weather Monitoring). Unconfirmed. We could not check this change against an archived copy of Tomorrow.io Weather API's page, so it does not count toward the rating.Source ↗
HIGH
discovered Aug 28, 2026 · effective date unknown
Hotjar
The free plan now offers 200k sessions per month, a significant increase from 20,000 sessions. It also includes features like error monitoring, funnels, and 10+ integrations. The free plan also includes a 15-day trial of the Growth plan. Unconfirmed. We could not check this change against an archived copy of Hotjar's page, so it does not count toward the rating.Source ↗
HIGH
discovered Aug 28, 2026 · effective date unknown
Grafana Loki
The page details a free tier of Grafana Cloud Logs offering up to 50GB of logs. The page now highlights Grafana Cloud as the easiest way to get started, with a free tier. Unconfirmed. We could not check this change against an archived copy of Grafana Loki's page, so it does not count toward the rating.Source ↗
MEDIUM
effective Jun 11, 2026
UptimeRobot
UptimeRobot's terms stopped limiting the Free plan to personal, non-commercial use between 2026-06-08 and 2026-06-11. The terms now allow commercial and business use. Source ↗
MEDIUM
effective May 3, 2026
Middleware.io
No longer in force (2026-09-08). Middleware replaced its Free Forever plan (100 GB of data, 1,000 RUM sessions and 20,000 synthetic checks a month, 14-day retention) with a 14-day free trial. The change falls between the 2026-04-20 and 2026-05-03 Internet Archive captures of middleware.io, and Middleware's April 2026 product update announced a pricing update to its free trial limits and free usage tiers. No longer in force from 2026-09-08 at the latest: Middleware's site shows the Free Forever card again in a Common Crawl capture of 2026-09-08, after the 14-day trial in the Internet Archive capture of 2026-09-01, and middleware.io/pricing/ opens with it today: "Free Forever $0 Free access to all features with monthly limits." Source ↗
HIGH
discovered Apr 19, 2026 · effective date unknown
Hasura
Hasura DDN Free plan replaced the legacy Hasura Cloud free tier (60 req/min, 1 GB passthrough). Free tier now offers unlimited API requests, unlimited concurrent users, and unlimited models at any scale. Restrictions: 1 supergraph developer and 15-minute observability retention. Source ↗
HIGH
effective Apr 8, 2026
Hyperping
Between 2026-03-08 and 2026-04-08, Hyperping's free plan went from 5 monitors checked every 3 minutes to 20 monitors checked every 5 minutes, with 1 basic status page and 1 seat. Paid plans did not change. Source ↗
MEDIUM
effective Aug 15, 2025
Sentry
Switched from performance units to spans-based metrics for error and performance monitoring Source ↗
MEDIUM
Data Source
How we track this data: the tables on this page were compiled by hand from vendor pricing pages. They do not come from the catalogue behind our search and MCP tools. Paid tier cost estimates use publicly available list prices as of 2026-04-03. Actual costs vary with usage, commitment discounts, and enterprise negotiations. Check data freshness.
A source check reads each cited page for that service's name and a price. The figures in the tables above are from our own records, except for the services whose entry below dates them to a read of the vendor's own page.
AppSignal — We read that on 2026-09-26 from appsignal.com/plans, and our check recorded: the page names AppSignal as "appsignal" and states "279 Logging 1GB /month". Our last read of it, on 2026-09-26, is where the terms above come from.
Axiom — We read that on 2026-09-30 from axiom.co/pricing, and our check recorded: the page names Axiom as "axiom" and states "500 GB/mo" and "500 GB / mo" and "10 GB-hours / mo". Our stored terms were last confirmed on 2026-09-13.
Better Stack Logs — We read that on 2026-09-22 from betterstack.com/logs/pricing, and our check recorded: the page writes "betterstack", the domain we cite Better Stack Logs from, and states "$0" and "100,000 exceptions per month" and "3 GB per month" and "100k exceptions / month". The terms above come from our read of 2026-09-07, and we have read the page since, on 2026-09-22, without confirming them.
BetterStack — We read that on 2026-10-06 from betterstack.com/pricing, and our check recorded: the page names BetterStack as "betterstack" and states "100,000 exceptions per month" and "3 GB per month" and "100k exceptions / month". Our stored terms were last confirmed on 2026-10-06.
Bugsnag — We read that on 2026-10-06 from bugsnag.com/pricing, and our check recorded: the page names Bugsnag as "bugsnag" and states "1M spans per month". The terms above come from our read of 2026-09-13, and we have read the page since, on 2026-10-06, without confirming them.
Checkly — We read that on 2026-09-20 from checklyhq.com/pricing, and our check recorded: the page names Checkly as "checkly" and states "$ 0". Our stored terms were last confirmed on 2026-09-20.
Cronitor — We read that on 2026-09-19 from cronitor.io/pricing, and our check recorded: the page names Cronitor as "cronitor" and states "$0"
Datadog — We read that on 2026-10-01 from datadoghq.com/pricing, and our check recorded: the page names Datadog as "datadog" and states "$ 0"
Elastic — the page we cite for this offer states that it is free, in words, and states no allowance we can read
Grafana Cloud — We read that on 2026-10-06 from grafana.com/pricing, and our check recorded: the page names Grafana Cloud as "grafana cloud" and states "$19" and "10k active series per month" and "50 GB ingested per month". The terms above come from our read of 2026-09-02, and we have read the page since, on 2026-10-06, without confirming them.
Healthchecks.io — We read that on 2026-09-22 from healthchecks.io/pricing, and our check recorded: the page names Healthchecks.io as "healthchecks io" and states "$0". Our stored terms were last confirmed on 2026-09-22.
Hyperping — We read that on 2026-10-03 from hyperping.com/pricing, and our check recorded: the page names Hyperping as "hyperping" and states "$0"
incident.io — We read that on 2026-09-26 from incident.io/pricing, and our check recorded: the page names incident.io as "incident io" and states amounts, none of which we matched to the terms we publish. Our stored terms were last confirmed on 2026-09-26.
Jaeger — the page we cite for this offer states no amount, tier or rate we can read
Middleware.io — We read that on 2026-10-03 from middleware.io/pricing, and our check recorded: the page names Middleware.io as "middleware" and states "$0". Our stored terms were last confirmed on 2026-10-03.
netdata.cloud — We read that on 2026-10-01 from netdata.cloud, and our check recorded: the page names netdata.cloud as "netdata cloud" and states amounts, none of which we matched to the terms we publish
New Relic — We read that on 2026-10-01 from newrelic.com/pricing, and our check recorded: the page names New Relic as "new relic" and states "100 GB/month" and "100GB/month". Our stored terms were last confirmed on 2026-10-01.
OnlineOrNot — We read that on 2026-10-01 from onlineornot.com/pricing, and our check recorded: the page names OnlineOrNot as "onlineornot" and states "$0"
PagerDuty — We read that on 2026-09-24 from pagerduty.com/pricing, and our check recorded: the page names PagerDuty as "pagerduty" and states "0 up to 5 users /month". Our stored terms were last confirmed on 2026-09-24.
pingbreak.com — the page we cite for this offer states no amount, tier or rate we can read
Prometheus — the page we cite for this offer states no amount, tier or rate we can read
Pulsetic — We read that on 2026-09-22 from pulsetic.com/pricing, and our check recorded: the page names Pulsetic as "pulsetic" and states "$0". Our stored terms were last confirmed on 2026-09-22.
Rollbar — We read that on 2026-09-25 from rollbar.com/pricing, and our check recorded: the page names Rollbar as "rollbar" and states "1K replays free per month". Our stored terms were last confirmed on 2026-09-25.
Sematext — We read that on 2026-10-04 from sematext.com/pricing, and our check recorded: the page names Sematext as "sematext" and states "500 MB/day"
Sentry — We read that on 2026-10-04 from sentry.io/pricing, and our check recorded: the page names Sentry as "sentry" and states "$0"
SigNoz — We read that on 2026-09-28 from signoz.io/pricing, and our check recorded: the page names SigNoz as "signoz" and states "$49" and "$ 0.3" and "$ 0.1" and "$0.30"
StatusCake — We read that on 2026-10-06 from statuscake.com/pricing, and our check recorded: the page names StatusCake as "statuscake" and states "$0"
UptimeRobot — We read that on 2026-10-02 from uptimerobot.com/pricing, and our check recorded: the page names UptimeRobot as "uptimerobot" and states "$ 0". Our last read of it, on 2026-10-02, is where the terms above come from.
Frequently Asked Questions
What is the best free monitoring service in 2026?
We publish no ranking of these. Nothing we record measures popularity or generosity, so naming a best or a top few would be our preference rather than our data. We hold 59 Monitoring services in our catalogue, each with the free tier terms. This page compares them on free tier limits, what each charges past them, and lock-in risk; the table above carries the figures side by side. Our ranking rule is published at /criteria.
Which monitoring free tier is most generous?
We name no most generous free tier here, because generosity is not one quantity: the columns above measure different things and a tier that leads one trails another.
How do monitoring free tiers compare on limits?
Each provider structures free tier limits differently — some cap storage, others cap requests or compute hours. Our comparison table provides exact numbers for each provider. Check the growth cost analysis section to understand what you'll pay when you exceed free tier limits.