BetishOps
Observation platform

Metrics that lead to answers.

Units, periods, trends and context. Every indicator should explain what it measures and lead back to its evidence.

Illustrative scenario · No customer data1 hour

2026-09-24 09:00 → 2026-09-24 10:00 UTC

How to read this scenario

Queue: 12 → 184 records in the illustrated observation window. CPU: 38% → 87% (+49 percentage points). Coverage: 11 of 12 planned observations.

Open the period’s data, method and evidence →
Definition & purpose

Every measure has a question.

IndicatorDefinitionOperational question
Queue backlogNumber of pending records in the latest observation for the context.Is the backlog growing or shrinking?
Oldest record ageWaiting time of the oldest observed record.Is work remaining stalled?
Interrupted jobsCount of jobs in the defined state within the period.Is there a recurring pattern?
CPUPercentage reported by the source; distinguish a sample from an average or peak.Which resource change coincides with the event?
CoverageReceived observations relative to those expected for the interval.Do we have enough information to compare?

Zero does not mean no data.

Observed zero

A valid reading returned zero. It is not the same as a missing reading.

Partial coverage

Available data is useful, but it does not establish that the whole period was observed.

Older data

The latest valid observation remains identified by its time and revision.

A contextual threshold

A color guides attention; definition, scope and evidence explain the result.

From your question to a demonstration

Let’s explore your operational context.

Let’s discuss your sources, your questions and the outcome you need. No credentials or sensitive data required.

Request a demonstration