Uptime monitoring compared — seven tools, one of them ours

We sell one of these. Here is when to buy a different one.

Every figure below comes from the vendor's own pages.

Comparison pages written by vendors are usually worthless, because the competitor's column is filled in from memory and the losing rows are quietly left out. This one records where each number came from, states what each competitor does better than Vigil, and ends with the cases where Vigil is the wrong answer.

Jump to the table Which one should you pick? prices as published, July 2026
The table

Ten axes that actually decide it.

Monitor counts and check intervals are where most comparisons stop. The rows that decide real setups are lower down: who can log in, whether clients can be isolated, and whether anything happens automatically when a check fails.

Uptime monitoring tools compared across licensing, deployment, check types, access control and price
  Vigilcommercial Vigil Corefree Uptime Kuma UptimeRobot Better Stack Pingdom StatusCake
Licence Source-available, commercial AGPL-3.0 MIT Proprietary Proprietary Proprietary Proprietary
Self-hostable Yes Yes Yes — only option No No No No
Runs the checks Your one host Your one host Your one host Vendor, many regions Vendor, many regions Vendor, 100+ probes Vendor, 43 locations
Fastest check 60 s 60 s 1 s 30 s on top tier, else 60 s 30 s paid, 3 min free 60 s 30 s on Business
Check types HTTP(S), keyword, TCP, TLS expiry HTTP(S), keyword ~31 types HTTP, keyword, ping, port HTTP, ping, SSL, domain, ports, Playwright HTTP, transactions, page speed 8 types incl. DNS, SMTP, SSH
Multiple users Yes, 4 roles Yes, 4 roles No — single account Seats billed separately Paid per responder Unlimited 1 on entry tier
Isolated client tenants Yes One org per install No No No No No
Branded status pages Unlimited, included One, included Unlimited, white-label Included from free 1 included; extra $12/mo; white-label $208/page/mo One per organization Separate product, from $12.46/mo
Acts on failure Verified auto-recovery Alerts only Alerts only Alerts only Alerts, on-call, incidents Alerts only Alerts only
Price shape $149 once Free Free Free tier; $7–$124/mo Free tier; $29/seat/mo + add-ons No free tier; from $198/yr Free tier; $20.41–$66.66/mo

Prices are the vendors' own published figures as read in July 2026, annual billing where offered, and they change — check the source before deciding: UptimeRobot · Better Stack · Pingdom · StatusCake · Uptime Kuma. Better Stack's pricing is fully à-la-carte with no named tiers, so the seat price above excludes add-ons you may well need.

Where each one wins

Six honest recommendations against ourselves.

If one of these describes you, buy that product instead. We would rather lose the sale than have it refunded in a fortnight.

Uptime Kumafree · MIT · self-hosted
Choose Kuma if you are one person and you want breadth for nothing. It is excellent, and more permissively licensed than Vigil Core — MIT, so you can even resell it. Around 31 monitor types against our two, roughly 98 notification integrations, one-second checks, and unlimited fully white-labelled status pages. On raw capability per pound it wins outright.

The wall is people, not features: Kuma has no multi-user support at all — one shared account, no roles, no read-only viewer, no per-client isolation. For a team where more than one person needs to log in, or where a client should see their page and nothing else, that is structural rather than a missing setting. The longer comparison →
PingdomSaaS · from $198/yr
Choose Pingdom if you need to prove uptime to someone else. Over 100 probe servers worldwide, automatic second-checking before alerting, real-browser transaction monitoring and unlimited users. When an SLA is contractual and the evidence has to come from independent vantage points, a single self-hosted box is simply the wrong instrument and no amount of our marketing changes that.

Worth knowing before you commit: there is no free tier, only a 30-day trial, and the vendor documents one status page per organization — so it is a poor fit for per-client pages.
Better StackSaaS · $29/responder/mo
Choose Better Stack if you want monitoring, on-call and logs from one vendor. The failure forensics are genuinely better than ours — a screenshot of the error, a second-by-second timeline, traceroute and MTR captured automatically — and Playwright transaction monitoring runs real Chrome, which Vigil does not attempt. Free members who only read telemetry cost nothing.

The cost model is the thing to model carefully: pricing is à-la-carte, so extra monitors, extra status pages, SSO and audit logs are separate line items, and removing Better Stack's branding from a status page is $208 per page per month on annual billing. For an agency wanting ten unbranded client pages, that is the whole decision.
UptimeRobotSaaS · free tier · $7–124/mo
Choose UptimeRobot if you want zero operations and a real free tier. 50 monitors free, and — contrary to what this site used to claim, which we have corrected — its terms explicitly permit commercial and business use. It is the cheapest credible route to 60-second checks from outside your network.

It stops fitting when the job becomes per-client: the free plan is five-minute checks with one status page, and the entry paid tier is explicitly single-login. The longer comparison →
StatusCakeSaaS · free tier · $20.41+/mo
Choose StatusCake if you want one vendor for uptime, SSL, domain and page speed. Eight uptime test types including DNS, SMTP and SSH, 43 locations across 30 countries with published IP lists for firewall allowlisting, and the ability to pick which continents run your tests — a real capability if you need regional failure detection.

The trap is status pages: they are a separate product, not included in any monitoring plan, and an unbranded page starts at $41.63/month on annual billing on top of your monitoring bill.
Freshpingdiscontinued
Not an option any more. Freshworks closed Freshping to new signups in June 2023 and disabled the remaining free accounts in January 2026, deleting the data. If you are here searching for a replacement, the closest like-for-like on generosity is Uptime Kuma if you can host it, or UptimeRobot's free tier if you cannot.

It is worth naming the risk it illustrates: the product was perfectly good and was retired because it stopped fitting a large vendor's portfolio. That is an argument for self-hosting something you can keep running yourself — which is the argument Vigil is built on, and the reason Vigil Core is published under a licence that cannot be withdrawn.
Against ourselves

Four reasons not to buy Vigil.

The same list is on the home page. It has not been softened for this one.

You need independent vantage points
Buy Pingdom or Better Stack

Vigil checks from the one host you deploy it on. A failure means that host could not reach the target — which is not the same statement as "the target is down", and it is not evidence you can put in front of a customer disputing an SLA.

You want ping, DNS or cron heartbeats
Buy Uptime Kuma or StatusCake

Vigil does HTTP(S) with keyword assertions, TCP ports and TLS expiry. No ping, no DNS, no push/heartbeat for cron jobs, and no scripted browser journey that logs in and buys something.

You do not want to run a server
Buy any of the hosted four

Self-hosting is the trade Vigil asks of you. Two processes and one Postgres is deliberately boring, and setup is included with a licence — but it is still your box, your backups, your upgrades. If that is not a trade you want, a hosted product is the honest answer.

You need someone to answer at 3 a.m.
Buy a vendor with a support org

Sikur is one engineer. Pre-sales questions get answered quickly and setup is included, but there is no 24/7 support rota and no SLA, and pretending otherwise would be the fastest way to lose your trust. That constraint is part of why the price is what it is.

Where Vigil does win

Two cases, stated narrowly.

The agency case. Many clients, each needing an isolated organization and their own branded status page, with your team on separate accounts and real roles. Every hosted product here either bills per page, bills per seat, or caps you at one status page; Kuma cannot do multi-user at all. One install, one payment, unlimited tenants is a genuinely different shape.

The recovery case. If the human woken at 3 a.m. is going to run a restart command that a machine could have run, Vigil closes that loop: re-verify the failure, call your signed hook, wait, probe again, and page only if it did not come back. The alert you would have received is dropped. Nothing else on this page does that as a built-in.

Every competitor figure here is from the vendor's own page, dated July 2026
Each competitor has a section saying what it does better than Vigil
We corrected a false claim we previously made about UptimeRobot's free tier
The free edition is public source, so you can check the engineering yourself
Questions

About this comparison.

Yes — we make one of the products and we want you to buy it. What we can offer instead of a claim of neutrality is method: every competitor figure came from that vendor's own published pages with the date recorded, each competitor has a section naming what it does better than Vigil, and the page includes four reasons not to buy from us.

Vigil, Vigil Core and Uptime Kuma. Better Stack, UptimeRobot, Pingdom and StatusCake are hosted services with no self-hosted or on-premise option — which also means none of them can reach an endpoint that is not publicly routable, so internal services behind a VPN are out of scope for them.

Uptime Kuma — MIT-licensed, free, no commercial restriction — if one shared login is acceptable. UptimeRobot's free plan also permits commercial use, at five-minute checks with one status page. Both stop fitting when each client needs an isolated, branded page and your team needs separate accounts with roles.

Vigil's commercial edition does, as a built-in: re-verify the failure, call a restart endpoint you own with an HMAC-signed payload, wait, probe again, and page a human only if the service did not come back. The others alert, and alerting is where they stop — several can trigger an external automation tool, but the verified detect-act-verify loop is not a feature they ship.

No. Freshworks closed it to new signups in June 2023 and disabled the remaining free accounts in January 2026, deleting the data. If you are searching for a Freshping alternative, that is why — and it is a reasonable argument for running something you can keep alive yourself.

Something here wrong or out of date? Tell us and it gets fixed — @s8kur or s8kur3@gmail.com. Corrections about our own claims are welcome; one is already recorded above.