FDST-BDS study materials
FDST-BDS study resources

Two Definitions of 'Active Policyholder' and a Friday Deadline

FDST-BDS · Put your knowledge to work in context.

All case studies

Two Definitions of 'Active Policyholder' and a Friday Deadline

Northline Mutual, a specialty insurer formed from a merger 18 months ago, has two groups claiming accountability for the Policyholder data domain. Finance defines an active policyholder as anyone with a policy in force or inside a 31-day grace period; Sales excludes grace-period records until premium is received. The gap is 3.1% of roughly 240,000 records, and it surfaced when a state regulator questioned a quarterly count that Finance had signed off. The data governance council has met monthly for seven months, has no charter clause on tie-breaking, and has deferred the item three times. The CDO, hired four months ago, has one data steward at 0.3 FTE and no authority over either group's headcount. The steering committee wants the CDO to 'just pick a definition' by Friday so the next regulatory filing can proceed, and the Sales VP has said publicly that Finance's number overstates the book.

1. Should the CDO unilaterally choose the definition? Identify the decision right at issue, who should hold it, and how the CDO should respond to the Friday deadline.

Read the worked answer

Framing. The question is not 'which definition is correct' but 'who holds the decision right for this term, in this context.' Two things are being conflated:

  1. The regulatory count. The state's filing instructions define the population (typically in-force plus grace). That is not a governance choice; it is a compliance requirement. If Finance followed the prescribed definition, changing it to satisfy Sales would make the filing wrong.
  2. The operational term 'active policyholder' used in pipeline and commissions. This is a genuine governance choice and needs a single accountable domain owner.

Who decides. Accountability for the Policyholder domain should sit with one named owner - most defensibly Finance or Actuarial, because the count feeds statutory reporting - with Sales as a consulted stakeholder. The CDO does not hold this decision right by default. A CDO who picks unilaterally takes on accountability without authority and will be overruled the first time a VP escalates.

Response to Friday. Meet the deadline without pre-empting governance. File on the regulator's definition, disclose the base and the 3.1% grace-period delta in a supporting schedule, and issue a written recommendation to the steering committee naming the proposed owner, the two term variants, and the tie-break mechanism. Publish both counts with labels in the interim so no downstream consumer blends them. That is a decision the CDO can make - about disclosure - rather than a decision about which number is true.

2. The council has no tie-break clause. Propose the minimum charter amendment and the interim operating procedure until it is ratified.

Read the worked answer

Minimum charter amendment - three clauses.

  • Tie-break default. If the council cannot reach consensus within two meetings, the proposed domain owner decides. The CDO may issue an interim standard valid for 90 days, ratified or overturned at the steering committee's next sitting. Dissents are recorded in the minutes but do not block.
  • Definitional variants. A business term may carry context-specific variants (statutory, operational, analytical), each registered with an owner, a formula, and the processes it binds. Conflict between variants is resolved by scope, not winner-take-all.
  • Escalation clock. Unresolved items auto-escalate after two cycles, accompanied by a written options paper.

Interim operating procedure.

  1. Register both definitions in the business glossary with explicit scope statements and a 'not valid for' field.
  2. Stand up a standing 3.1% reconciliation control on the quarterly count; variance above 1% triggers investigation.
  3. Bind every downstream report and filing to a named variant so a consumer cannot accidentally mix them.
  4. Scope the 0.3 FTE steward to definition custody only - glossary upkeep and issue triage - not remediation, which that capacity cannot absorb.
  5. Log this as precedent #1 in the decision log so the next domain conflict has a template.

The reasoning: definitional conflicts are usually structural (missing decision rights), not analytical. Fix the mechanism once and it applies across all domains.

3. What interim control should be placed on the regulatory filing itself so the same error cannot recur next quarter?

Read the worked answer

Control objective. Ensure the filed number is traceable to a named definition and reproducible, and that the 3.1% question cannot silently reappear.

Controls.

  • Definition provenance. Every filed figure carries a reference to the glossary variant used, its owner, and the version in force at the filing date. If the variant changes, the filing changes with a documented restatement.
  • Dual-count reconciliation. Generate both counts (Finance and Sales logic) from the same snapshot, retain both in the filing workpaper, and document the delta by driver - grace-period policies, cancellations in flight, reinstatements. A delta that cannot be decomposed by driver is a data quality incident, not a definitional one.
  • Independent sign-off. The preparer and the reviewer must not be in the same reporting line; the reviewer confirms the definition applied matches the regulator's instructions, not the internal preference.
  • Change control. No change to the 'active policyholder' logic in the reporting pipeline may be deployed without council or owner approval recorded against the glossary entry.
  • Audit trail retention. Keep the snapshot, the query, and the sign-off for the statutory retention period.

Why this rather than a note in the deck. A footnote explains one filing; a provenance-and-reconciliation control prevents the next one. It also converts the governance argument into an auditable artefact, which is usually what brings the steering committee to a decision faster than a debate about which number looks better.