Skip to content
SHIFTLOOM

People and continuity

How to Read Resignation Risk Without Turning It Into a Score

A quiet week or missed deadline says little on its own. What matters is a sustained departure from someone’s normal working pattern, supported by more than one source and interpreted in context. Even then, the evidence should open a conversation. It should not determine a conclusion.

  • Baseline before anomalyCompare change with the person’s established pattern.
  • Context before interpretationConsider role, leave, workload, and project changes.
  • Conversation before conclusionUse evidence to ask a better question, not assign a label.
1

Look for change, not a profile

There is no universal pattern for resignation risk. The useful question is whether someone’s way of working has changed meaningfully from their own established baseline. A single event is weak evidence; persistence and corroboration make a change worth examining.

  • A change persists beyond an ordinary fluctuation
  • The same shift appears in more than one work context
  • Ownership, participation, workload, or delivery changes materially
  • The pattern remains after known changes in role, leave, or priorities are considered
2

Rule out ordinary explanations

Role changes, time away, difficult projects, shifting priorities, health, and personal circumstances can all alter working patterns. Before interpreting a change, check what changed around the person and distinguish what the evidence shows from what it cannot explain.

The purpose of evidence is to make a conversation more informed, not to make the conversation unnecessary.

3

Open a conversation, not a case

A responsible response starts with an open question. Ask what has changed, listen for workload or ownership problems, and determine whether support or clarification is needed. A pattern should never become a concealed performance label or the sole basis for an employment decision.

4

Where software should stop

ShiftLoom can organize authorized evidence and preserve the path back to each source. It should not infer motive, diagnose burnout, or make an employment decision. Human context and direct conversation remain decisive.

People evidence should remain reviewable, proportionate, and secondary to direct conversation.