![107 wooden dominoes standing upright in formation representing Netglare's completed end-to-end tests](/sites/default/files/styles/large_1600x900/public/2026-08/107-test-tubes.webp?itok=Lzvmv4Ip)

 

- [25 March 2026](#)
- [ 2 min read ](#)
 
 

### 107 Ways To Break (We Fixed Them All)

There's a specific kind of anxiety that comes with finishing a product. Not the exciting kind. The "what did we miss?" kind. The kind that keeps you reading through code you've already read three times, running tests you've already run twice, convinced that something somewhere is quietly broken in a way you haven't found yet.

That's where we were before [Netglare's ](https://netglare.com)launch. Feature-complete, sprint-done, theoretically ready—and absolutely certain we'd forgotten something critical.

So we did what any reasonable team does when they're not sure if their product actually works: we tried to break it. Systematically. Thoroughly. 107 times.

## What End-To-End Testing Actually Means

End-to-end testing sounds impressive until you realize it's mostly just trying every possible thing a user could do and seeing what explodes. Create an account. Add a monitor. Set an alert threshold. Trigger the alert. Receive the notification. Upgrade your plan. Downgrade your plan. Cancel your subscription. Try to do things in the wrong order. Try to do things as the wrong type of user. Try to do things that nobody should be doing but someone definitely will.

For Netglare, that meant 107 individual test scenarios across 17 test groups covering everything from basic account registration to edge cases in our alert cooldown logic that we're slightly embarrassed took three attempts to get right.

## What We Found

Here's the honest answer: we found bugs. Not catastrophic ones—nobody's credit card got charged twice and no alerts fired for sites that were perfectly healthy—but bugs nonetheless. The kind that would have confused real users. The kind that would have generated support emails. The kind that would have made us look less professional than we needed to look at launch.

We fixed them. All of them. Then we ran the tests again to make sure fixing one thing didn't break another. (It did. Twice. We fixed those too.)

## What 107 Tests Actually Taught Us

The most useful thing about rigorous testing isn't finding bugs. It's the shift in confidence it creates. Before testing, "it works" means "it works when I use it carefully in a controlled environment." After 107 tests, "it works" means "it works when someone uses it carelessly in ways we didn't anticipate."

That's a meaningful difference when you're about to put your product in front of real users who will absolutely use it carelessly in ways you didn't anticipate.

## What Comes Next

Sprint 7 starts next week. The telemetry layer. The data accumulation that turns Netglare from "another monitoring tool" into something that gets smarter the longer you use it. Then we launch.

But first: 107 tests. 107 passes. Zero shortcuts.

If you want to be one of the first teams to use Netglare when we open the doors, head to netglare.com and sign up for early access. Or reach out directly at getstarted@gatoblan.co if you want to talk through your monitoring setup first.

We promise we tested it.



 

 

 Share: 

- [    ](https://www.facebook.com/sharer/sharer.php?u=https://www.gatoblan.co/markdownify/node/43&title=107%20Ways%20To%20Break%20%28We%20Fixed%20Them%20All%29 "Share to Facebook")
- [    ](https://twitter.com/intent/tweet?text=107%20Ways%20To%20Break%20%28We%20Fixed%20Them%20All%29+https://www.gatoblan.co/markdownify/node/43 "Share to X")
- [    ](https://www.linkedin.com/sharing/share-offsite/?url=https://www.gatoblan.co/markdownify/node/43 "Share to Linkedin")
- [    ](mailto:?subject=107%20Ways%20To%20Break%20%28We%20Fixed%20Them%20All%29&body=https://www.gatoblan.co/markdownify/node/43 "Share to Email")