Alerts

Get told when it matters, not for everything.

A brand new issue or a sudden spike reaches you where you already work. A known, quiet error stays quiet. You tune the threshold instead of muting the whole feed.

This week, by alert reasonSample data
New issue 9
Spike over baseline 5
Regression after deploy 3
Muted and known 41

How it works

01

Watch for what is new

A first-seen issue triggers an alert, because a new failure mode is the thing worth interrupting for.

02

Catch sudden spikes

When an existing issue jumps well above its baseline, that counts as news even if it is not new.

03

Reach your channel

Alerts land where your team already looks, so nobody has to camp on a dashboard.

New-issue alerts
Interrupt for a genuinely new failure, not a familiar one.
Spike detection
A jump above baseline is treated as news.
Your channel
Delivered where the team already works.

Ship with a net under you.