Skip the VPS, the Docker,
the cert renewal
Uptime Kuma is excellent open-source software with one of the largest self-hosted communities on GitHub. It is free as in source code, not free as in time. SiteLetter is managed monitoring with Lighthouse on every monitored page, AI visual regression, and white-label client reports. No VPS, no patches, no monitor-the-monitor problem.
Also evaluating UptimeRobot, Hyperping, or Oh Dear? See those comparisons, or browse all alternatives.
Quick verdict
Both are good for the people they were built for. The decision comes down to whether your agency wants to manage monitoring infrastructure or focus on client work.
Self-hosting-comfortable teams
Agencies (or sysadmins) comfortable running Docker on a VPS, with the time and inclination to handle OS patches, version upgrades, backups, and the monitor-the-monitor problem. Or anyone who wants the monitoring database on hardware they control.
Also a strong fit if your check-type needs span many protocols (TCP, DNS, Docker, Steam, Push heartbeats) or your alert routing needs unusual channels (Gotify, niche SMS gateways).
Agencies who want managed + client deliverable
Web design, development, and marketing agencies who want monitoring to disappear into the background while shipping a polished white-label client report every week. No VPS to maintain, no ops time to bill back to overhead, no monitor-the-monitor problem.
SiteLetter at $4.80/site a month with every monitor on covers Lighthouse on every monitored page and AI day-over-day visual regression, each a monitor you switch on per site, fully managed. White-label scheduled client reports, with your logo, colours and sender name, are included with every site, at no separate rate.
What the monthly client report needs beyond self-hosted uptime
Uptime Kuma is genuinely powerful uptime monitoring software. The gaps below are not paid-tier upgrades: as of August 31, 2026, the project's published documentation and repository do not describe these features.
Lighthouse on every page, no headless-Chrome wrapper to deploy
Uptime Kuma's monitor types include HTTP, TCP, Ping, DNS, Keyword, JSON Query, Websocket, Push, Steam-server and Docker-container checks, among others. Google Lighthouse audits with the four-score breakdown are not among them. SiteLetter runs the full Lighthouse on every monitored page across every client site, daily, with DNS, TLS, and CDN cache warmup before each audit, and ships the Performance, Accessibility, SEO, and Best Practices scores in the weekly client report - without you running a headless-Chrome wrapper on your VPS to produce them yourself.
Day-over-day screenshot diffs, not yours to build
Visual regression is not in Uptime Kuma's feature set. To add it you would self-host a separate tool, integrate it into the dashboard yourself, and maintain both. SiteLetter ships the day-over-day flow as a managed service: two screenshots per monitored page (mobile and desktop) every 24 hours, then an AI classifier that dismisses cosmetic noise (rotating banners, cookie variants), summarises intentional content changes in the next scheduled report, and flags real layout breakage in the dashboard and in the report you forward when the daily scan detects it.
White-label reports you forward, not just incident alerts
Email exists in Uptime Kuma as an alert channel - you get a notification when a check goes down. What does not exist is a scheduled, per-client branded weekly or monthly report: no agency logo, no brand colour, no reply-to address on a scheduled artifact. SiteLetter builds that report on the cadence you set and puts it in your inbox ready to forward, and it opens without an account on anyone's dashboard.
Per-client branding included, not "host it on your domain"
Uptime Kuma's branding model is the URL: self-host it at status.youragency.com and that is the branding. Its status pages take a custom domain and badges, but the project's published documentation does not describe a per-client branding profile or a branded report artifact you send to a client (checked August 31, 2026). SiteLetter's reports are white-label per client: your logo, your brand colour, your reply-to address, and your agency name as the sender. An org-level setting keeps the SiteLetter name out of the report body.
Onboarding speed and zero infrastructure
Uptime Kuma runs on your own VPS via Docker or Node, and in practice you also want a reverse proxy with TLS, an external check to watch the watcher, and a patching routine. SiteLetter is a managed signup: add a domain in the dashboard, the product auto-fetches the sitemap and the first scan of up to 15 pages is booked for the next free quarter-hour, usually within minutes, no ops setup on your side.
One URL to onboard a client, not 15 monitor forms
Onboarding a client on Uptime Kuma means defining 15 monitors per site, by hand in the dashboard or scripted through their REST write endpoints or a backup import, each with its own interval and alert configuration. Onboarding the same client on SiteLetter is pasting one URL. The sitemap is auto-fetched, the 15 most important pages are picked (depth + sitemap priority + recency), and every monitor turns on. Alert rules are written once at agency level and inherited by every client and every site, so the number of alert configurations you maintain does not grow with the portfolio.
Feature-by-feature comparison
The full picture, grouped by category. Some of these are SiteLetter wins (per-page Lighthouse + visual + white-label reports + managed hosting). Many are Uptime Kuma wins (free software, 10+ monitor types, 90+ alert channels, full data control, and no vendor who can switch your instance off). The honest answer is "both, for different operational preferences."
| Feature | SiteLetter | Uptime Kuma |
|---|---|---|
| Cost & hosting model | ||
| Software cost | $4.80/site/mo with every monitor on (managed) | $0 (MIT license) |
| Infrastructure cost | None on your side (SiteLetter hosts it) | $5-20/mo typical VPS (DigitalOcean / Hetzner) |
| Ops time / maintenance burden | Zero (SiteLetter patches, upgrades, monitors the monitor) | Yours: patches, upgrades, backups, monitor-the-monitor, for as long as those take on your setup |
| Out-of-pocket cost at 10 sites | $48.00/mo with every monitor on (managed, no ops time) | ~$10/mo VPS + your own ops time at your own rate |
| Monitoring data retention | 2 years (730 days) for scans, screenshots, Lighthouse runs, asset findings and reports; raw uptime check rows kept 280 days | Operator-controlled (whatever you configure in SQLite; default unbounded with manual pruning) |
| Per-page services on a client site | ||
| Uptime checking | Homepage every 1 min from the EU, confirmed from a second region on failure; the other 14 monitored pages are reached daily by the scan, not per minute | 1 monitor per URL; intervals down to 20 seconds |
| Google Lighthouse audits (Performance / Accessibility / SEO / Best Practices) | Every monitored page, daily, with cache warmup | No |
| AI day-over-day visual regression with classification | Every monitored page, daily; AI classifies every material change | No |
| Broken asset detection across pages | Every monitored page, daily | Not in product as a per-page check |
| Monitoring infrastructure | Managed (SiteLetter runs it, you do not) | Self-hosted on your VPS (you patch, upgrade, watch the watcher) |
| Site-level services | ||
| SSL certificate expiry alerts | Yes | Yes (certificate info + expiry on HTTP monitors) |
| Domain expiry alerts | Yes | No native domain/WHOIS expiry monitor |
| Sitemap auto-discovery | Yes (weekly sync) | No sitemap auto-discovery (monitors defined per URL, by hand or scripted against the unsupported internal API / backup import) |
| Public status pages | Not available | Yes (multiple pages, custom domain, badges) |
| Transaction / multi-step checks | No | Limited (HTTP Keyword, JSON Query against single endpoint; no Playwright) |
| Non-HTTP monitor types (TCP/Ping/DNS/Docker/etc.) | No | Yes (10+ types) |
| Client-facing reports | ||
| Scheduled client-facing email report | Yes (weekly, bi-weekly, monthly, quarterly) | No (email is alert-channel only) |
| Per-client branding (logo, color, reply-to) | Yes | No first-class per-client branding |
| Org-level report branding (hide the SiteLetter name) | Yes: white-label in the report body; sender display name is yours, and with a verified domain the From address is yours too | Self-hosting on your own domain is the branding; no branded report artifact in the published docs (see verification date below) |
| Drag-and-drop report builder | Yes | No |
| Recipients view reports without a login | Yes - the white-label report renders inside the email, and every report carries a public link that opens without an account. | Public status pages are accessible without login; no scheduled email report flow |
| Platform features | ||
| Alert channels | Email, Slack | 90+ (Email, Telegram, Discord, Slack, Teams, Gotify, Pushover, SMS gateways, webhooks, and many more) |
| Public REST API | Not available | Not officially supported - the project wiki says the internal API "is not officially supported for third-party integrations" and that breaking changes may land between versions without notice; the only key-gated surface is the Prometheus /metrics scrape |
| Native mobile app | No | No official mobile app (third-party community apps exist; alerts reach phones via Telegram/Discord/Pushover/SMS) |
| Vendor failure risk | Managed SaaS (vendor-side risk applies) | MIT-licensed code on your own hardware: no vendor who can switch your instance off, and the operating risk is yours |
Sources: github.com/louislam/uptime-kuma, project wiki, and DigitalOcean / Hetzner published VPS pricing. Verified .
Which job are you buying for?
Comparing monitor counts flatters whoever wrote the comparison. These are the two questions that actually decide it, at 10 client sites.
For uptime alone, Uptime Kuma is a complete answer
Uptime Kuma is free software. It does uptime, SSL expiry and status pages very well, down to 20-second intervals, and on uptime alone the whole bill is a small VPS plus the hours you spend keeping it patched and its alerting honest.
SiteLetter can be bought that way too, since uptime is a line item you can switch on without the rest. But uptime alone is not the job this page is about, and pretending otherwise would mean pricing around a comparison you can run yourself in a minute.
Lighthouse, visual change detection, a branded report
Uptime Kuma has no Lighthouse audits, no visual change detection and no scheduled branded per-client report, and none of the three is on its roadmap: it is an uptime tool and does not pretend otherwise. Matching the agency deliverable means building the rest around it.
10 client sites with every monitor on, covering up to 150 monitored pages. Prepaid, nothing recurring.
Self-hosting is a real cost even when the software is free: the VPS, the upgrades, the alert routing, and being the person paged when the monitor itself goes down.
Uptime Kuma figures are the ones that vendor publishes, in the currency they publish them. Verified .
Where Uptime Kuma genuinely wins
Uptime Kuma is a beloved open-source project with one of the largest active self-hosted communities on GitHub. If you are comfortable self-hosting and want full data control and no vendor who can switch your instance off, it is a genuinely powerful tool. The wins below are real and SiteLetter does not match them.
Free open-source software (MIT license)
One of the most-starred self-hosted projects on GitHub, with an active community and frequent releases. Fully MIT licensed. Zero recurring license cost: software itself is genuinely free. The only spend is your hosting infrastructure and time.
10+ monitor types out of the box
HTTP(s), TCP, HTTP Keyword, HTTP JSON Query, Websocket, Ping, DNS Record, Push heartbeat, Steam Game Server, Docker Container, plus more. Down to 20-second intervals on the higher-frequency setups. If your monitoring need spans protocols beyond plain HTTP, Uptime Kuma covers a wider native check-type surface than SiteLetter.
90+ notification channels
Email (SMTP), Telegram, Discord, Slack, Microsoft Teams, Gotify, Pushover, dozens of SMS gateways, generic webhooks, and many more. If your alert routing needs span unusual channels, Uptime Kuma probably ships one for you. SiteLetter has email and Slack today.
Full data control + self-hosted privacy
The SQLite database lives on your hardware. No vendor stores your monitoring data, no third-party processor sees your client URLs, no vendor lock-in. If keeping the monitoring database on infrastructure you control is what you are after, self-hosting answers that outright. SiteLetter answers the same question differently: Monitoring data is stored in the EU. Screenshots go to OpenAI in the US for visual change classification, under EU Standard Contractual Clauses.
Status pages built in
Multiple status pages with custom domain mapping, badges, public or private. SiteLetter does not offer status pages. If hosted public status pages are part of your client deliverable today, Uptime Kuma already ships them.
No vendor decision can reach the copy you are running
A SaaS monitoring vendor can change pricing, sunset features, get acquired, or shut down (see Freshping, March 2026). None of that reaches MIT-licensed software already running on your own hardware: the source is on your machine and the database is on your disk. What self-hosting moves rather than removes is the operating risk, and it moves it to you: the unpatched instance, the VPS that dies at 3am, the project that stops shipping security fixes.
When to pick which
Three honest scenarios.
Agencies who want managed
Web design, development, marketing agencies that bill clients for their time. Managed monitoring at $4.80/site a month with every monitor on, with white-label reports, Lighthouse and AI visual regression, saves the ops time you would otherwise spend on Uptime Kuma maintenance, and that saved time bills back at agency rates.
Self-hosting teams
DevOps-comfortable teams already running infrastructure, or anyone who wants the monitoring database on hardware they control. That is what Uptime Kuma's MIT license and self-hosted model are for.
Agency hosting + agency client work
You host client infrastructure AND deliver client monitoring reports. Uptime Kuma on the infrastructure side (low-cost uptime + protocol checks for ops alerting), SiteLetter for the per-page Lighthouse + visual + branded client reports. The overlap is narrow: both do HTTP uptime and SSL expiry. Everything else does not: per-page Lighthouse, visual regression and branded client reports on one side, protocol checks and status pages on the other.
SiteLetter vs Uptime Kuma: FAQ
The software is free (MIT). Running it is not: a small DigitalOcean or Hetzner droplet is typically $5-$20/mo, or nothing beyond electricity if you already run a home server or a Raspberry Pi. Beyond hosting, you pay in ops time: patching the OS, upgrading Docker / Node, applying Kuma version upgrades, maintaining the SQLite database and disk space, renewing the dashboard TLS certificate, and running an external "watch the watcher" check so you know if the host itself goes down. How long that list takes is not a number anyone can quote for you: it depends on your stack, your patching discipline, and how much of it you have already automated. Price it at your own billable rate and that is the honest cost comparison. Free as in software, not free as in time.
No. Uptime Kuma ships HTTP(s), TCP, Ping, DNS, Keyword, JSON Query, Websocket, Push, Steam game server, and Docker container monitors, among others. It does not run Google Lighthouse audits with the four-score breakdown (Performance / Accessibility / SEO / Best Practices) on monitored URLs. SiteLetter runs full Lighthouse on every monitored page on every site daily with cache warmup, and the four-score breakdown ships in the weekly client report.
No. Visual regression is not part of the Uptime Kuma feature set. SiteLetter takes a daily screenshot of every monitored page and diffs it against the last good capture, normally yesterday's, with AI classifying each diff big enough to matter as noise (rotating banners auto-dismissed), an intentional update (logged in the weekly report), real breakage (missing images, layout collapse, flagged in the dashboard and called out in the report you forward), or couldn't-verify.
No. Email is one of Uptime Kuma's many alert channels for incident notifications (you get an email when a check goes down), but there is no scheduled per-client weekly or monthly report builder, no per-client logo, no branded report layout. SiteLetter is built around that flow.
Status pages and scheduled branded client emails are different deliverables. Uptime Kuma's status pages are excellent: multiple pages, custom domain mapping, badges. Your client visits a URL to see current status. SiteLetter's scheduled email report lands in your inbox on your cadence (weekly/bi-weekly/monthly/quarterly), already carrying your agency logo, brand color, and a write-up of what changed - including Lighthouse trends and AI-classified visual changes - so forwarding it is the whole job, and once you verify your own sending domain you can put your client on the report instead and skip even that. Status pages answer "is it up right now?"; scheduled reports answer "what happened this month and is everything healthy?"
A lot, for the right team. 90+ notification channels (Telegram, Discord, Gotify, Pushover, Slack, SMS gateways, and more). Multiple status pages with custom domains. A wide range of monitor types including Docker containers, Steam game servers, push heartbeats, and Websocket monitors. Full data control: the database lives on your hardware. Active OSS development on one of the most-starred self-hosted projects on GitHub, with frequent releases. Zero recurring license cost. If you are comfortable self-hosting and want full control over your monitoring infrastructure, Uptime Kuma is genuinely powerful.
Managed infrastructure: SiteLetter runs the host, the patches, the upgrades, and the TLS renewals (Uptime Kuma is self-hosted; all of that is your operational responsibility, for however long it takes on your setup). Per-page Google Lighthouse on every monitored page daily with DNS/TLS/CDN cache warmup (Uptime Kuma does not run Lighthouse audits). AI day-over-day visual regression that classifies each diff big enough to matter as noise / intentional / breakage (Uptime Kuma does not screenshot pages). White-label scheduled per-client email reports with per-client logo, brand color, reply-to (Uptime Kuma's email is incident notifications, not scheduled per-client deliverables). Sitemap auto-discovery: add a domain, SiteLetter picks the 15 most important pages and books their first scan for the next free quarter-hour, usually within minutes, with Lighthouse, screenshots, broken-asset checks and the SSL and domain-expiry readings all on that scan (Uptime Kuma is per-monitor manual definition through the dashboard). Uptime probes the homepage every minute.
It depends on how you count the time. Software cost: Uptime Kuma is $0; SiteLetter is $4.80/site a month with every monitor on. VPS hosting cost: Uptime Kuma needs $5-20/mo for a small VPS regardless of monitor count; SiteLetter has zero infrastructure cost on your side. Ops time: Uptime Kuma adds ongoing patching, upgrades, backups and monitor-the-monitor configuration; SiteLetter adds zero. That is the one line nobody can fill in for you, and it is where the comparison is actually decided: put your own hours on it, at your own billing rate, and set the total against the per-site figure above. The honest comparison is "do you want to manage monitoring infrastructure, or focus on client work."
Uptime Kuma runs on one host. If that host crashes, the monitoring also stops, and the alerts that would tell you it is down originate from the same host that is down, so they never fire. Self-hosted users solve this by running a second external check (commonly a cron-based ping from another VPS, or a third-party uptime monitor) that watches the Uptime Kuma instance. That is "the monitor monitoring the monitor" - extra infrastructure to maintain. SiteLetter is managed: SiteLetter runs the infrastructure, monitors it, and the alert pipeline is on SiteLetter rather than your team at 3am. Every down result is confirmed by a second AWS region before alerting (EU primary, US verifier), and a third region in Singapore breaks the tie when the two disagree. The limit is worth stating plainly: the dispatcher itself runs in eu-north-1, so a full eu-north-1 outage pauses checks until the region recovers - same trade-off any managed monitoring service makes, but at least the operational response is SiteLetter's problem to solve, not yours.
Yes. The migration is conceptual rather than technical: Uptime Kuma stores monitor configs in a SQLite database on your VPS; SiteLetter sites are added via the dashboard with auto-discovery of up to 15 pages per site (sitemap + link graph). A 100-monitor Uptime Kuma setup packs into 7 SiteLetter sites when every site is filled to the 15-page cap, and more than that as the same pages spread across more domains. That arithmetic assumes those monitors are extra pages of domains you already cover: the collapse is per domain, not per monitor. A hundred monitors sitting on a hundred separate client homepages stays a hundred SiteLetter sites. New accounts get 14 days of full monitoring on one site with no card, so you can validate the managed experience on one client site before bringing the rest over.
Still have questions? Get in touch.
Comparison based on publicly available Uptime Kuma documentation, project repository, and community discussion, last verified . Uptime Kuma is open-source software released under the MIT license, maintained by Louis Lam and a community of contributors. SiteLetter is independent and not affiliated with the Uptime Kuma project. For the most current Uptime Kuma information, see github.com/louislam/uptime-kuma.
You set up care-plan reporting once.
After that it runs without you.
SiteLetter is the kind of tool you forget you're using, until your client thanks you for the report you didn't write.
- ✓ Just add a domain. SiteLetter auto-discovers pages, starts monitoring, schedules reports. You pick your alert rules and who gets each report.
- ✓ No scripts to install. SiteLetter monitors from the outside, like a real visitor.
- ✓ Nothing recurring. No subscription, no contract. You buy credits when you want them and each site spends its own.