The Quick Answer
- Most websites: Every 5 minutes
- Revenue-critical pages: Every 1-3 minutes
- SSL certificates: Daily
- Domain expiry: Daily
- Background jobs: Per schedule (heartbeat)
If you're unsure, start with 5-minute intervals. It's the standard for a reason—fast enough to catch problems, slow enough to avoid false positives.
Check Intervals Explained
A check interval is how often your monitoring service sends a request to your site. A 5-minute interval means your site is tested every 5 minutes, 24/7.
What the Interval Really Means
Your check interval is also your maximum detection time. With 5-minute checks, worst case is discovering downtime 4 minutes 59 seconds after it starts. On average, detection takes half the interval.
| Interval | Max Detection Time | Avg Detection Time | Checks per Day |
|---|---|---|---|
| 30 seconds | 29 seconds | ~15 seconds | 2,880 |
| 1 minute | 59 seconds | ~30 seconds | 1,440 |
| 3 minutes | 2 min 59 sec | ~1.5 minutes | 480 |
| 5 minutes | 4 min 59 sec | ~2.5 minutes | 288 |
| 15 minutes | 14 min 59 sec | ~7.5 minutes | 96 |
The question is: does the difference between knowing in 30 seconds vs 2.5 minutes change the outcome? For most sites, no.
Recommended Frequency by Use Case
E-commerce Checkout
Every minute of checkout downtime is lost revenue. Fast detection lets you act before significant money is lost.
SaaS Application
Users notice quickly when a SaaS app is down. Fast detection matters for customer trust and SLA compliance.
API Endpoints
API consumers (mobile apps, integrations) are affected immediately. Fast detection prevents cascading failures.
Marketing Website
A few extra minutes of detection time rarely changes the outcome. 5-minute checks are efficient and reliable.
Blog / Content Site
Low urgency. Users will come back. SEO impact only kicks in with extended downtime.
Internal Tools
Your team will survive learning about an outage a few minutes later. Save faster intervals for customer-facing services.
SSL Certificates
Certificates don't expire suddenly. Daily checks with 30/14/7-day alerts give you plenty of time to act.
Domain Registration
Domain expiry dates don't change. Daily WHOIS checks catch any unexpected issues well in advance.
Faster Isn't Always Better
More frequent checks seem obviously better. But there are real trade-offs:
More False Positives
With 30-second checks, a brief network hiccup triggers an alert. With 5-minute checks, that same hiccup resolves before the next check and you never hear about it. Without confirmation checks, faster intervals mean more noise.
Load on Your Server
Each check is an HTTP request to your server. At 30-second intervals from 12 regions, that's 1,440 extra requests per hour. For most servers this is negligible, but for small instances or rate-limited APIs, it matters.
Cost
Most monitoring tools charge more for faster intervals. If 5-minute checks are free but 1-minute checks require a paid plan, ask yourself: is the extra ~3 minutes of detection time worth the cost?
More on this decision: Free vs paid monitoring.
Diminishing Returns
Going from 15-minute to 5-minute checks saves you up to 10 minutes of detection time. Going from 5-minute to 1-minute saves only 4 minutes. Going from 1-minute to 30-seconds saves just 30 seconds. Each jump costs more but delivers less improvement.
Confirmation Checks: The Missing Piece
Check frequency and confirmation checks work together. Here's how:
A confirmation check means: don't alert on the first failure. Wait for 2-3 consecutive failures before notifying you.
Without Confirmation
1-minute interval, alert on first failure
Real alert time: 1 minute
False positive risk: High
With Confirmation (2 failures)
1-minute interval, alert after 2 consecutive failures
Real alert time: 2 minutes
False positive risk: Low
Best Balance
3-minute interval, alert after 2 consecutive failures
Real alert time: 6 minutes
False positive risk: Very low
For most sites, 3-5 minute intervals with 2-failure confirmation is the sweet spot: real problems detected within 10 minutes, virtually no false positives.
More on alerting: How to set up alerts that work.
How Regions Affect Frequency
If you check from 12 regions at 5-minute intervals, your site is actually being checked much more frequently in aggregate—roughly every 25 seconds from somewhere in the world.
This means multi-region monitoring gives you faster effective detection even at longer individual intervals. A 5-minute multi-region check often detects issues faster than a 1-minute single-region check.
Key insight: Multi-region + 5-minute intervals often beats single-region + 1-minute intervals for both detection speed and reliability.
Learn more: Multi-region monitoring and checking uptime from users' perspective.
Frequently Asked Questions
Is 5-minute monitoring good enough?
For most websites, yes. 5-minute intervals detect problems within 5 minutes maximum, 2.5 minutes on average. Combined with multi-region checks and confirmation, this catches real problems quickly while avoiding false positives. Only e-commerce checkouts and critical SaaS apps typically benefit from 1-minute intervals.
Do more frequent checks affect my website performance?
Minimally. Each check is a single HTTP request—equivalent to one user visit. Even at 1-minute intervals from 12 regions, that's only 720 extra requests per hour. For any production server, this is negligible. The only exception is extremely rate-limited APIs where every request counts.
Should different pages have different check intervals?
Yes, ideally. Your checkout or API should be checked more frequently than your blog or about page. Most monitoring tools let you set intervals per monitor. Prioritize fast detection on revenue-critical paths and use standard intervals for everything else.
What if I can only afford 5-minute checks?
Use them with multi-region monitoring and confirmation checks. 5-minute multi-region checks detect most real outages within a few minutes. The extra 3-4 minutes compared to 1-minute checks rarely changes the outcome for most businesses.
Start With 5 Minutes, Optimize Later
Don't overthink frequency. 5-minute intervals with multi-region checks and confirmation is the right starting point for 90% of websites. Upgrade to faster intervals for specific critical endpoints when the data tells you it matters.