self-hosted error monitor Open the panel

Customers

who is suffering, and from what open the screen

What it is for

The same data, sliced by customer. It answers "who is suffering most" and "how many distinct customers does this error hit", which is the difference between a bug affecting one customer with an odd setup and a bug affecting everyone.

How the customer gets into meulog

It comes in with the event, as a bug report field or as an attribute the SDK sent. meulog has no customer registry: what shows up here is what the events brought.

What you can configure

The label. Whoever installs meulog does not always call the customer "company": under Settings you can change it to Agency, Store, Branch, Tenant. The label changes across the whole interface and in the alert messages.

The customer report

Writes, with AI, a summary of one customer's week. The reader is different from the daily summary: the daily one is for whoever operates, this one is for the meeting, and whoever reads it does not want to hear about fingerprints.

It shows on screen to be copied. Nothing is sent: a customer-facing report goes through a person first, always.

The timeline

The clock icon on each row opens that customer's timeline: every usage event (screens, clicks, requests) and every error in the period, in one list, newest first. It answers "what were they doing when it broke" without crossing two screens by the clock. Usage events only exist for applications that send them; see sending errors.

Cohorts

The "Cohorts" button groups customers by the week they first showed up and follows each group for eight weeks. With usage events, showing up is the first usage event and each column is the share still using the system; without them, it is the first error and each column is the share still having errors. The row that fades fast is a group that left, or that stopped breaking.

What you should see

One row per customer with occurrences, distinct errors, machines and the date of the last one. Sortable by any column.