Click, not code:
monitor client sites without writing TypeScript

Not as a first-class native product. Lighthouse access from inside Checkly browser checks is a long-standing user request tracked on their public feedback portal at feedback.checklyhq.com, and users currently roll their own Lighthouse runs inside a Playwright script. SiteLetter runs full Google Lighthouse on every monitored page on every site daily with DNS/TLS/CDN cache warmup, and the four-score breakdown (Performance / Accessibility / SEO / Best Practices) ships in the weekly client report.

Checkly ships Visual Regression Testing as a labeled feature on Team and Enterprise tiers - automatic screenshot capture and comparison against a stored baseline inside Playwright browser checks. That is CI/test-style visual regression: the screenshot is compared against the baseline you set, the check fails if they differ, you update the baseline manually. SiteLetter is continuous: every monitored page is screenshotted every 24 hours and diffed against the previous day, and AI classifies each diff big enough to matter as noise (rotating banners auto-dismissed), intentional update (logged in the weekly report), or real breakage (missing images, layout collapse, emailed as soon as the daily scan spots it). Different shapes of the same idea - Checkly is "did the page differ from my baseline this CI run," SiteLetter is "what changed on every monitored page since yesterday, and is any of it real breakage."

No. Checkly's email is alert-only: notifications when a check fails. There is no scheduled, per-client branded weekly or monthly report builder. SiteLetter is built around that flow.

You can consume Checkly as an end-user for your own client monitoring work, but read the Use Restrictions in their terms carefully (Section 2.7 of checklyhq.com/terms as of May 2026): it prohibits "sell, resell, encumber, rent, lease, time-share, distribute, transfer or otherwise use or make available any of the Service to or for the benefit of any third party." Agencies running resale arrangements (selling Checkly access to clients under your agency brand) would need explicit written agreement from Checkly. Agencies using Checkly as a back-end monitoring tool to inform their own work for clients are normal end-users. SiteLetter's terms also require written permission to resell or redistribute the Service, but using SiteLetter as an agency to monitor client sites is the intended use and is not restricted. This is not legal advice; read both sets of terms.

Monitoring as Code: Checkly's flagship workflow. The checkly-cli lets you define checks in TypeScript/JavaScript, version them in Git, run them in CI (GitHub Actions has a first-class integration), and deploy via checkly deploy. They have a Terraform provider, a public REST API, and explicit AI-agent positioning in 2026 ("The active reliability layer for developers & agents." on their homepage as of August 2026). If your team is developer-first and wants monitoring defined as code alongside the application, Checkly is the better tool today. SiteLetter does not ship monitoring-as-code, CLI, or Terraform provider.

Per-page Google Lighthouse on every monitored page daily with DNS/TLS/CDN cache warmup (Checkly has no first-class Lighthouse product). AI day-over-day visual regression that classifies each diff big enough to matter as noise / intentional / breakage (Checkly has snapshot/golden-image comparison inside Playwright scripts, not continuous classified diffing). Branded scheduled per-client email reports with per-client logo, brand color, reply-to (Checkly's email is alert-only). Sitemap auto-discovery: add a domain, SiteLetter picks the 15 most important pages and every monitoring service turns on for them (Checkly is per-monitor manual definition). Report branding included for every client: your logo, your brand color, your agency name as the sender display name, and your reply-to address. $5/site flat for up to 15 pages, every service.

It depends on what you need. Checkly's Hobby Free tier covers 10 uptime monitors at $0/mo - cheaper than SiteLetter at small homepage-only scale. For per-page coverage at agency portfolios (15 monitors per site), Checkly's Team tier at $64/mo covers up to 75 monitors (5 sites worth), and above that you need Enterprise (custom pricing). At 6+ sites of matching coverage, SiteLetter's $30/mo flat is meaningfully cheaper than Checkly's Enterprise tier (6 sites is 90 monitors, past the 75-monitor Team cap).

SiteLetter keeps 2 years (730 days) of Lighthouse scans, screenshots, asset checks and reports, uniform across every subscriber. Raw per-minute uptime samples roll off after 7 days; uptime incidents are kept with the rest of the history. Checkly's raw retention varies by plan (around 30 days on Starter/Team, longer on higher tiers per checklyhq.com/docs/monitoring/how-we-store-data); aggregated rollups are retained long-term. For agencies showing month-over-month or year-over-year client trends from raw monitoring data, SiteLetter's 2-year retention is meaningfully longer.

Yes, with a caveat: if your existing setup depends on monitoring-as-code workflows (checks defined in TypeScript, deployed via CI, version-controlled in Git), SiteLetter does not replicate that workflow. You add sites in the SiteLetter dashboard, the product auto-discovers up to 15 pages of each site from your sitemap, falling back to the homepage's links. Lighthouse, screenshots, uptime and asset checks start on the first scan; SSL and domain expiry fill in within 24 hours. Alert rules are the one thing you write yourself: you do it once at agency level, and every client and every site inherits them. For agencies whose value is in the client deliverable (Lighthouse, visual regression, branded reports), the SiteLetter setup is shorter and the trial includes one site so you can validate the experience before extending.