Skills CX operations

Attrition early warning

Most "attrition prediction" projects fail in one of two ways: they surveil what they should not (private life, health proxies, social graph), or they produce a list nobody acts on — a risk score with no owner, no intervention, and no feedback loop. Both erode trust and do not reduce leavers.

CX operationsWorkforceAnalysisAny helpdeskRead-only
Installnpx rulebase-skills install cx-attrition-early-warning

When to use it

Reach for this when someone says any of these — they are the phrases the skill itself triggers on:

  • attrition risk
  • who might leave
  • early warning on leavers
  • retention signals
  • schedule driving quit rates

How it works

The method, in the order the skill runs it. The full procedure — tables, worked examples and the edge cases — is in the skill itself.

  1. Signals that are in bounds

    These reflect working conditions, not private life.

  2. Step 1: frame the question as population health

    Ask: "Where is the organisation creating unnecessary leave risk?" not "Which agents should we worry about?"

  3. Step 2: build a simple signal panel

    No black-box model required for most ops teams. For a defined window (e.g. rolling 8 weeks): Voluntary attrition rate by team, tenure band, channel, Mean and variance of occupancy vs agreed target band, Schedule change count per FTE, QA volatility, Optional.

  4. Step 3: test for confounds before blaming managers

    Attrition clusters often trace to: Pay or comp change in the same quarter, Site or vendor contract uncertainty, Product incident raising handle time and QA markdowns together, Roster policy change (e.g. new weekend mandate), Selection.

  5. Step 4: turn signals into retention design actions

    Every flagged pattern needs an owner and intervention type.

  6. Step 5: governance

    Document: Data sources and retention period, No use for performance rating, termination, or comp reduction, Review cycle, Disparate impact check if any scoring is used.

Related skills

Free and open source, and vendor-neutral — it reads the conversations from whichever helpdesk you already run. Browse all 149 skills · connect your helpdesk over MCP · source on GitHub

Review every conversation. Act on what it finds.

AI for customer operations, built for financial services. Specialist agents chase every issue to resolution and every stalled customer to activation.

Rulebase dashboard