An Alert Should Tell You What Changed, Not Just That It Did
Most cloud alerting fails two ways: silent on the events that mattered, or so noisy you stop reading it.
It stays silent on the events you actually needed to know about, or it fires a stream of context-free notifications you quickly learn to ignore. Either way, when your cloud changes state, the alert does not tell you what moved, why it matters, or where to go next.
The answer is not more alerts. It is alerts that cover the right events and carry the facts of each one.
ZopNight now widens what raises an alert and, in Microsoft Teams, delivers each one as a card you can read and act on without leaving the channel.
The events that were happening without a signal
A cloud environment changes state constantly, and most of those changes used to pass without a signal. A schedule that started or stopped a resource. A whole resource group acted on at once. An override that expired. An automation rule that changed. A budget crossed, a bill moved, a cost anomaly surfaced, a remediation completed or failed, a report export finished. Each is a moment something in your account became different from how you left it.
Alert coverage now spans a subscribable catalog across five categories: Scheduling, Automation, Budgets and Billing, Cost Anomaly, and Remediation. These are the points where state changes, and where finding out after the fact means reconstructing the sequence from logs. Getting told at the moment turns that reconstruction into a card in a channel.
In Teams, a card instead of a line
Coverage is only half of it. A notification that says something happened, without saying what, sends you off to go find out. In Microsoft Teams, ZopNight now delivers each alert as a color-coded Adaptive Card built for the event it describes.
The card carries the facts that matter for that event type, a severity band so urgency reads at a glance, and cloud-account and resource-group names resolved to what you call them rather than raw identifiers. Provider labels are clean, so a Lambda event reads as AWS, not an internal string. And a one-click link takes you straight to the exact resource in ZopNight when the card is not enough. You triage in the channel and open the console only when you decide to act.
| Raw notification | Adaptive Card in Teams | |
|---|---|---|
| What happened | You go find out | On the card |
| Which account and group | Raw identifiers | Resolved names |
| How urgent | Undifferentiated | Severity band, color-coded |
| Next step | Search the console | One-click deep link |
Subscribable, so coverage does not become noise
Wider coverage is a liability if every event reaches everyone. This catalog is subscribable, so each team receives the categories it owns and nothing else. You pick exactly which alerts you want from Alerts settings, and routing is per organization.
| Category | Example events |
|---|---|
| Scheduling | Manual and scheduled start and stop, resource-group actions, override expired |
| Automation | Automation rule changes, autoscaler and event-readiness lifecycle |
| Budgets and Billing | Budget crossed, billing alerts |
| Cost Anomaly | Anomalous spend surfaced |
| Remediation | Approval, completion, failure, ticket opened |
When broad coverage helps, and when to keep it narrow
If one person watches one account, a narrow alert set is fine, and turning everything on would only add noise. Coverage earns its place the moment more than one team acts on shared infrastructure, where schedules, automation, and remediation all change state and no single person is watching all of it.
Subscriptions are what keep that from becoming the wall of notifications people mute. The goal was never to send more. It was to make sure the events that change your cloud reach the people who own them, in a form they can act on. Broad where it needs to be, quiet everywhere else.
