Benchmark — aggregate uptime, SSL, and page-speed findings across the SiteGuardian fleet..

99.85% median

Across the rolling 30-day window SiteGuardian logs ~99.85% aggregate uptime on a fleet-wide basis — what an unmonitored small shop sees when a one-person operator is its only "alert". A 99.85% baseline means a single 14-minute outage per day, on average, fleet-wide.

~14 min / day

A 99.9% uptime SLA — the round-number threshold most ecommerce vendors quote — translates to ~14 minutes of allowable offline time per day. The headline figure is the percentage; the floor is what gets drained the first time a checkout or contact page goes dark.

6+ hours

When a small-shop site goes dark without an active watcher, the median recovery window stretches past six hours — the gap between the site going dark and a customer noticing. Active monitoring shortens the recovery clock by alerting before the customer does, not after.

~1 in 20

~5% of the fleet SiteGuardian monitors falls into the "expiring within 14 days" bucket at any given probe — a stack of certs whose owner either did not set up a renew webhook, or whose webhook silently dropped. The 14-day window is the boundary WebKit, Chrome, and Firefox use to surface the "renew this" warning; below that, real visitors see certificate-expiry splash pages.

~1% of fleet

On any single probe, roughly 1% of the fleet resolves to a self-signed certificate, a wrong-host certificate, or a broken chain. The bucket is small but the failure mode is binary — every visitor gets a red-bar browser warning at once. The fix is a fresh cert, often in under an hour after the alert fires.

~1 in 3

Even with HSTS and certificate-transparency logging, SSL certificate expiry incidents remain roughly 1 in 3 of the reported browser-red-bar outage causes. Renew havens fail regularly: a domain lapses, a credit card fails, a host webhook silently drops.

2.5s

Google's Largest Contentful Paint threshold for the "good" band is 2.5 seconds; anything slower is "needs improvement" or "poor". Beyond the threshold, ecommerce conversion curves drop noticeably — every additional 100ms compounds cart abandonment.

200ms

Interaction to Next Paint (INP) replaces First Input Delay in 2024. The "good" threshold is 200ms; above 500ms is the "poor" band. Most unmonitored small-shop assets hover in the needs-improvement band, especially on mid-range mobile.

~2.2s

The fleet-wide median for Largest Contentful Paint across the rolling 30-day window sits near 2.2s — just under the "good" threshold and well within the noisy 75th-percentile band Chrome crawlers observe across the wider web. Speed improvements move the bar in the same direction across LCP, INP, and CLS.