Slow Is the New Down: Response Time Threshold Alerts

Traditional uptime monitoring answers one question: is your site returning a 200? If yes, you're "up." But that's not really what uptime means to your users.

A checkout page that takes 8 seconds to load is technically up. Your monitoring tool reports green. Meanwhile, your conversion rate is collapsing and users are abandoning their carts — and you have no idea.

This is why we built response time threshold alerts.

---

How it works

For each monitor, you can now set a response time threshold in milliseconds. When PingBase checks your URL and the response time exceeds that threshold, the monitor is flagged as degraded — a distinct state from "down" or "up."

Degraded means: the site is reachable, but it's not performing within acceptable bounds. You get an alert immediately. When it recovers below the threshold, you get a recovery notification.

The degraded state also shows on your public status page — so your users see "degraded performance" rather than "operational" when you're struggling, which is both more honest and more useful to them.

---

What thresholds should you set?

There's no universal right answer, but here are sensible defaults by endpoint type:

A good rule of thumb: look at your p95 response time in normal conditions, then set the threshold at 3–5× that. You want to be alerted when something is genuinely wrong, not every time there's a brief spike.

---

Real scenarios this catches

Database under load

Your site is up. Your database has hit a connection pool limit. Every request takes 6 seconds to complete. Without response time alerts, this looks like a normal day to your monitoring tool.

Cold starts after a deploy

You deployed new code. The first few requests hit cold function instances and take 4 seconds. Your health check returns 200. Users on your pricing page are getting a loader for four seconds. You find out from a support ticket.

Third-party API degradation

A payment processor or auth provider you depend on slows down. Your endpoints still work — they just take much longer than usual. Response time thresholds catch this class of problem before it becomes a full outage.

---

Setting it up

1. Open any monitor in your PingBase dashboard
2. Under Alert settings, find Response time threshold
3. Enter a value in milliseconds (e.g. 1000 for 1 second)
4. Save — PingBase will start flagging slow responses on the next check

Leave the field empty to disable the threshold for that monitor. You can set different thresholds on different monitors — your health check endpoint and your homepage have very different performance expectations.

---

Available on all paid plans

Response time threshold alerts are available on Pro ($9/mo) and above. All existing Pro monitors can use the feature immediately — just open a monitor and set a threshold.

Free plan monitors still track response times and show them in the dashboard. The threshold alerting (getting notified when you breach it) requires Pro.

---