> ## Documentation Index
> Fetch the complete documentation index at: https://oximy.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Workflows and cohorts

> Find recurring AI work patterns and groups with similar adoption behavior.

Workflows summarize recurring work detected across captured activity. Cohorts group people with similar, license-aware adoption behavior. Together they help turn raw usage into an enablement plan.

<Tabs>
  <Tab title="Workflows">
    Review detected patterns, their activity and participants, and whether a
    pattern is concentrated in one department or shared across departments.
    Open a workflow to inspect its supporting context.
  </Tab>

  <Tab title="Cohorts">
    Review generated groups and their adoption and financial context where
    available. Department-scoped viewers can distinguish cohorts specific to
    their department from groups shared with other departments.
  </Tab>
</Tabs>

## Use workflows to standardize repeatable work

<Steps>
  <Step title="Find repetition">
    Look for a pattern with meaningful recurring activity rather than a single
    isolated session.
  </Step>

  <Step title="Check distribution">
    Determine which people, departments, and tools contribute to the workflow.
  </Step>

  <Step title="Review the evidence">
    Inspect supporting activity and confirm that the grouped sessions represent
    the same business task.
  </Step>

  <Step title="Choose an intervention">
    Document the successful approach, expand access, provide training, or review
    the tool and policy posture before scaling it.
  </Step>
</Steps>

## Use cohorts to target enablement

Cohorts are useful for deciding where a shared action may be more appropriate than person-by-person outreach. Compare a cohort's behavior with department and tool data, then validate the people included before assigning training, licenses, or governance actions.

<Info>
  Workflows and cohorts are derived from available data. Treat them as
  investigation aids, not as permanent employee labels or automatic judgments
  about performance.
</Info>