If your website goes down at 3 a.m. on a Saturday, would you know before your customers do? Most business owners find out from an angry email or a lost sale. The good news: you can check website uptime free, set up automatic alerts in under ten minutes, and do it without writing a single line of code.
This guide is a hands-on walkthrough. We compare six free tools, show you exactly how to set the first one up, and, more importantly, explain which numbers signal a real hosting problem and when it is time to open a support ticket or upgrade your plan.
Uptime vs Response Time: Two Different Health Checks
People mix these up constantly, and it leads to the wrong support ticket being sent to the wrong team.
- Uptime answers a yes/no question: is the site reachable right now? A monitoring service sends a request every few minutes and records whether it got a valid response. The result is a percentage, for example 99.94% over 30 days.
- Response time (often reported as server response time or TTFB, time to first byte) answers a how-fast question: how many milliseconds did the server take to start replying? A site can be up 100% of the time and still be painfully slow.
A site that is up but takes 3 seconds to respond is, in commercial terms, partly down. That is why the tools below are split into two families: continuous monitors that watch your site around the clock, and on-demand testers that give you a deep one-time diagnostic.
The 6 Free Tools Compared
Free tiers change over time, so treat the table as a starting point and confirm limits on each provider’s pricing page before you commit. Figures below reflect what these services advertise as of September 2026.
| Tool | Type | Free monitors | Check interval | Alerts on free plan | Reporting depth | Best for |
|---|---|---|---|---|---|---|
| UptimeRobot | Continuous | Up to 50 | 5 minutes | Email, Slack, Discord, webhooks (SMS on paid credits) | Uptime %, response time graph, event log, public status page | Almost every small business. The default starting point. |
| HetrixTools | Continuous | Up to 15 | 1 minute | Email, Telegram, Slack, webhooks | Uptime %, response time, blacklist monitoring, server resource agent | Owners who want 1-minute detection and email/IP blacklist checks. |
| Site24x7 (free tier) | Continuous | Small fixed number (typically 5) | Configurable, down to 1 minute on higher tiers | Email, mobile app push | Deepest free reports: 130+ global locations, per-region latency breakdown | Sites with international customers who need regional data. |
| Pulsetic | Continuous | Limited free plan | Typically 5 minutes on free | Email, Slack, webhooks | Clean dashboards and very presentable branded status pages | Agencies and SaaS owners who want a public status page clients can see. |
| Pingdom Website Speed Test | On-demand | Unlimited manual tests | Manual only | None | Waterfall chart, page size, request count, performance grade | Pinpointing whether slowness is the server or the page itself. |
| GTmetrix | On-demand + light monitoring | 1 monitored page on free account | Daily or weekly on free | Email alerts on performance thresholds | Core Web Vitals, Lighthouse scores, waterfall, historical graphs, video playback | Tracking page speed trends and SEO performance metrics over time. |
Bonus: instant spot checks with no account
If you just want to know right now whether the site is down for everyone or only for you, the Uptrends Free Website Uptime Test fires a request from dozens of global checkpoints and shows you the result in seconds. No signup, no monitor to configure. Keep it bookmarked for the “is it just me?” moment.
Which one should you actually pick?
- You have one website and zero technical background: UptimeRobot for 24/7 alerts, plus GTmetrix once a month for speed.
- You run a shop or booking site where minutes cost money: HetrixTools for 1-minute checks, so you hear about an outage before your customers do.
- You sell to customers on several continents: Site24x7 free tier, because regional latency is invisible from a single-location check.
- You need to show clients or staff a status page: Pulsetic or UptimeRobot’s public status page.
Walkthrough: Set Up Free Uptime Monitoring in Under 10 Minutes
Here is the exact sequence using UptimeRobot, the simplest of the six. The steps are nearly identical on HetrixTools and Pulsetic. Website Uptime Monitoring covers this in more depth.
- Create the account. Sign up with your email and confirm it. No credit card is required on the free plan.
- Add your first monitor. Choose monitor type HTTP(s), give it a friendly name such as “Main site”, and paste the full URL including https://.
- Set the interval. Free plans usually default to 5 minutes. Leave it there unless the tool offers 1 minute for free.
- Choose alert contacts. Add your email. Then add a second channel that reaches you outside business hours: Slack, Telegram, Discord, or your phone via the mobile app. One channel is not enough, because email is exactly what stops working when your domain has a problem.
- Add a keyword check if available. Instead of only checking that the server replies, tell the monitor to look for a word that must appear on the page, for example “Add to cart” or your company name. This catches the nasty failures where the server returns a blank white page with a 200 status code.
- Add three more monitors while you are there. Monitor your checkout or contact page, your mail server or webmail URL, and your SSL certificate expiry. SSL expiry alerts alone will save you one emergency per year.
- Turn on the public status page (optional). It gives your team and customers a single URL to check instead of calling you.
- Test it. Pause the monitor, resume it, or temporarily point it at a deliberately broken URL to confirm the alert actually lands in your inbox and on your phone.
Wait 7 days before drawing conclusions. One bad afternoon is noise. A week of data is a signal.
How to Read the Numbers Without Being a Sysadmin
1. Uptime percentage: what each number costs you in real time
Percentages look abstract until you convert them to minutes. This is the table to keep on your desk.
| Uptime | Downtime per month | Downtime per year | Verdict |
|---|---|---|---|
| 99.99% | about 4 minutes | about 53 minutes | Excellent |
| 99.9% | about 44 minutes | about 8 hours 46 min | Industry standard SLA |
| 99.5% | about 3 hours 39 min | about 44 hours | Below par, ask questions |
| 99% | about 7 hours 18 min | about 3 days 15 hours | Unacceptable for a business site |
| 98% | about 14 hours 36 min | about 7 days | Time to move |
Rule of thumb: anything consistently under 99.9% for two consecutive months, excluding maintenance you were notified about, deserves a written explanation from your host.
2. Server response time: the thresholds that matter
Your monitoring dashboard shows an average response time in milliseconds. Here is how to interpret it for a standard CMS site such as WordPress on shared or VPS hosting.
| Average response time | What it means | Action |
|---|---|---|
| Under 200 ms | Excellent, well-tuned server or good caching | Nothing to do |
| 200 to 400 ms | Normal and healthy | Monitor the trend |
| 400 to 800 ms | Sluggish. Caching missing, heavy plugins, or a busy server | Optimise first, then ask your host |
| 800 ms to 1.5 s | Visitors notice. Conversions and rankings suffer | Open a ticket with data |
| Over 1.5 s | Serious resource contention or misconfiguration | Escalate or upgrade the plan |
3. Patterns beat single numbers
The average is the least interesting figure on your dashboard. What tells the real story is the shape of the graph.
- Flat line that suddenly steps up and stays up. Something changed: a plugin update, a new theme, a traffic increase, or a change on the server. Compare the date with your own change log.
- Daily spikes at the same hour. Classic backup job, cron task, or a noisy neighbour on shared hosting. Your host can confirm and reschedule.
- Spikes only during business hours. Real traffic is outgrowing your resources. This is an upgrade conversation, not a bug.
- Random short outages of 1 to 5 minutes, several times a week. Often PHP worker or memory limits being hit, or an unstable network segment. Collect the exact timestamps.
- Slow from one country only, fast everywhere else. That is routing or CDN, not your server. Do not blame the host until you have checked from multiple locations.
4. Use the on-demand testers to split server from page
This is where Pingdom and GTmetrix earn their place. Run a test and look at the very first bar of the waterfall chart, the request for the HTML document itself.
- The first bar is long (over 800 ms) but the rest of the page loads quickly: the problem is server side. Hosting resources, database, or PHP configuration.
- The first bar is short but total load time is 6 seconds: the problem is your page. Unoptimised images, too many scripts, external fonts, chat widgets and tracking pixels.
Knowing which of the two it is will save you a pointless support ticket and a lot of frustration.
When to Contact Your Hosting Provider (And What to Send)
Support teams can only act on specifics. “My site is slow” gets a generic reply. Data gets an engineer.
Contact your provider when you see:
- Uptime below 99.9% over a full month, outside announced maintenance windows.
- Repeated 503 or 504 errors, which point to resource limits or a gateway timeout rather than your code.
- Response times that have climbed steadily over several weeks while your site content and traffic have stayed roughly the same.
- Outages clustered at a predictable time each day or night.
- Slowness or downtime confirmed from multiple geographic checkpoints, not just one.
The ticket template that gets a real answer
Copy, fill in and send:
Subject: Uptime below 99.9% and rising response times on [yourdomain.com]
Hello,
I monitor my site with [tool name] using HTTP checks every [5] minutes. Over the last 30 days I recorded:
- Uptime: [99.4%] with [12] separate incidents
- Average response time: [910 ms], up from [340 ms] last month
- Longest outage: [14 minutes] on [date and time, with timezone]
- Error codes observed: [503 Service Unavailable]
- Outages cluster around [03:10 server time] most nights
Could you please check server load, PHP worker and memory limits, and any scheduled jobs for my account during those windows, and tell me whether my current plan’s resources are the limiting factor?
Thank you.
A good host will come back with load graphs, error log extracts and a clear recommendation. At Ameritech Hosting, that is exactly the level of detail our team is expected to provide, and our clients are welcome to send us their monitoring exports so we can match them against server-side logs.
When It Is Time to Upgrade Your Hosting Plan
Not every performance problem is your provider’s fault. Some are simply a sign of growth. Consider moving up a tier or to a VPS when:
- Your site is fast at night and slow at peak hours, every single day.
- You hit CPU, RAM, entry process or PHP worker limits regularly, and the host’s logs confirm it.
- Response times stay high even after you have installed caching, compressed images and removed unused plugins.
- You have added a shop, a membership area, a booking engine or a large media library since you first signed up.
- Traffic has grown by more than 50% since your last plan change.
- You now need staging, dedicated resources or a strict SLA for commercial reasons.
Conversely, do not pay for an upgrade if your monitoring shows a healthy sub-400 ms server response and the slowness comes from a 5 MB hero image and eleven third-party scripts. More server power will not fix a heavy page.
A Simple Monitoring Routine for Non-Technical Owners
- Once, today: set up UptimeRobot or HetrixTools with 2 alert channels and an SSL expiry monitor.
- Weekly, 2 minutes: open the dashboard, glance at the response time graph, note any spikes.
- Monthly, 10 minutes: export or screenshot the uptime percentage and average response time. Keep them in a folder so you can prove a trend later.
- Quarterly, 15 minutes: run GTmetrix and Pingdom on your homepage and your most important money page, and compare with the previous quarter.
- Whenever something changes: after a redesign, a plugin batch update or a migration, run a test the same day so you know your new baseline.
FAQ
How do I check the uptime of a website?
Sign up for a free monitoring service such as UptimeRobot, HetrixTools or Pulsetic, add your URL as an HTTP(s) monitor, and choose a check interval of 1 to 5 minutes. The service then requests your site continuously and reports the percentage of successful checks, plus a log of every outage. For a single instant check with no account, use a free one-off tester such as the Uptrends uptime test.
What is the best free tool for monitoring website uptime?
For most small businesses, UptimeRobot offers the best balance: up to 50 monitors, 5-minute checks, multiple alert channels and a public status page at no cost. If you need faster detection, HetrixTools gives 1-minute checks on its free tier. If you need global coverage and richer reports, look at the Site24x7 free plan. UptimeRadar: Free Uptime Monitoring Tool is a useful companion to this.
Is there a free way to test my website speed?
Yes. The Pingdom Website Speed Test and GTmetrix are both free to run and show a full waterfall chart, page size, request count and performance grades. Google PageSpeed Insights is another free option focused on Core Web Vitals. Run the same page on the same tool each time so your comparisons are meaningful.
Is 100% uptime possible?
Over a short window such as a month, yes, many sites record 100%. Sustained 100% over a year is effectively impossible because every stack needs occasional maintenance, kernel patches and hardware work. That is why serious providers commit to 99.9% or 99.99% rather than 100%, and why redundancy and failover, not promises, are what reduce real-world downtime.
How often should uptime checks run?
Every 5 minutes is enough for a brochure site. For e-commerce, booking systems or anything transactional, use 1-minute checks so a short outage is caught and confirmed quickly. Remember that with a 5-minute interval, a 4-minute outage can go completely unrecorded.
Does downtime hurt SEO?
Short, rare outages generally do not. Repeated or prolonged downtime does: crawlers reduce crawl frequency, and if a page returns errors for an extended period it can be dropped from the index. Slow server response also affects Core Web Vitals, which feed into page experience signals. Keeping response times under 400 ms protects both users and rankings.
Can I monitor more than just my homepage?
You should. Add monitors for your checkout or contact form page, your login or client portal, your API endpoint if you have one, your mail service, and your SSL certificate expiry. Homepages are often cached and stay up while the dynamic parts of a site fail silently.
The Takeaway
You do not need a technical background, and you do not need a budget, to check website uptime free and keep a permanent record of how your hosting is performing. Ten minutes of setup gives you alerts, evidence and leverage: three things that turn a vague complaint into a fixed problem.
Set up your monitor today, let it collect a month of data, then compare it against the thresholds in this article. If the numbers are not where they should be, talk to your provider. If your provider cannot explain them, that is your answer too. The Ameritech Hosting team is always happy to review monitoring data with you and recommend the right plan for your actual traffic, not a bigger one. Anyone digging further should read Uptime Monitoring by Better Stack.