Incidents
These cases look related.
Similarity plus timing is the signal. Three customers describing the same failure within an hour is worth surfacing even if the wording is completely different.
Rate, not just similarity
Twelve invoice questions this week against a normal two is an emerging issue. The same twelve spread over a quarter is ordinary demand.
A suggestion, never automatic
The product proposes an incident; a person creates it. Automatically declaring an incident on a false pattern is its own kind of outage.
Cases can be added and removed
Grouping is rarely perfect first time. Adding a fifteenth case or removing one that turned out to be different is a normal, cheap operation.
Incidents
The rest of this area.
Group related cases, communicate once, close them all together.
Who is affected
Everyone hit, including the ones who have not written in.
Status updates
One message, to every affected customer, at once.
Public communication
A status page for people who have not written in.
Telling everyone it is fixed
One action closes the loop with every affected customer.
When it happens again
Repeated incidents pointing at one root cause.
Take better care of every customer.
Give your team the context, knowledge and AI they need to resolve problems properly — and know who needs attention before they ask.
Keep the mailbox you already use · The AI is never metered · [email protected]