the data reports
Monitoring, measured.
Every "best monitoring tool" page on the internet is written by a company selling one. So the one thing none of them publish is the thing you actually want to know: how do these tools behave when nobody's watching the marketing site? We collect that ourselves and put the numbers here — no vendor wrote this, and nobody paid to look good.
Three signals, gathered first-hand on a schedule and committed as dated snapshots so every figure is auditable: status-page incident history (how often a SaaS tool's own status page lit up), GitHub maintenance velocity (is the open-source project actually alive?), and pricing drift (did the page quietly change?).
Data as of Jun 24, 2026 · 16 vendor status feeds polled · 25 vendors in the panel.
Reliability snapshot — who had a quiet 90 days
Significant incidents (minor, major or critical — informational notices and planned maintenance excluded) logged on each vendor's own Atlassian Statuspage over the trailing 90 days. Over the panel that's 130 significant incidents across 16 vendors. The quietest: Axiom, ServiceNow Cloud Observability (Lightstep), Mezmo. The busiest: Grafana Cloud / LGTM (43), Elastic Stack (ELK) (22), Sumo Logic (17).
| Vendor | Significant · 90d | Incident-minutes | Worst impact | Last incident |
|---|---|---|---|---|
| Grafana Cloud / LGTM | 43 | 27.4 days | critical | Jun 22, 2026 |
| Elastic Stack (ELK) | 22 | 6.4 days | major | Jun 17, 2026 |
| Sumo Logic | 17 | 17 h | major | Jun 18, 2026 |
| Datadog Network Monitoring | 14 | 35 h | critical | Jun 22, 2026 |
| Coralogix | 12 | 8 h | critical | Jun 24, 2026 |
| Logz.io | 7 | 21 h | minor | Jun 8, 2026 |
| Catchpoint | 5 | 4.0 days | major | Jun 22, 2026 |
| Cisco ThousandEyes | 5 | 3.1 days | major | Jun 24, 2026 |
| New Relic | 3 | 15 h | major | May 18, 2026 |
| Sentry | 1 | 2 h | minor | Jun 24, 2026 |
| Atlassian Statuspage | 1 | 29 min | minor | Apr 9, 2026 |
| Axiom | 0 | 0 min | none | — |
| ServiceNow Cloud Observability (Lightstep) | 0 | 0 min | none | — |
| Mezmo | 0 | 0 min | none | — |
| Papertrail | 0 | 0 min | none | — |
| StatusCake | 0 | 0 min | none | — |
Method & honest caveats. Pulled from each vendor's public
/api/v2/incidents.json Statuspage feed. "Significant" counts
minor/major/critical and drops the informational notices vendors post for tooling — which is
why a tool with a busy-looking status page can still rank clean here. Incident-minutes
sum per-incident durations; on a multi-component platform those overlap, so it is a
relative busyness signal, not a downtime figure. Vendors on a bespoke
(non-Statuspage) status page aren't counted — absence here isn't a perfect score. This data
also rides along on each tool profile
and the picker, where it actively
demotes flaky tools.
The reports
The Monitoring Maintenance Index
CollectingWhich open-source monitoring projects are actually maintained — latest release, last commit and open-issue load across every tool with a public repo.
The Incident Leaderboard & The Monitoring Tax
In progressDeeper cuts on reliability over time and on pricing drift — published once we have enough dated snapshots to make the trend honest. The reliability snapshot above is the live preview.
Everything here is collected by a scheduled pre-build job and committed as a dated JSON snapshot, so every number is reproducible and every refresh is a reviewable diff. Found something stale or wrong? It might be — tell me.