Data cleaning
AuraScore 79/100

Omnichannel Loyalty Profile Identity Resolution Verification

Cleanse and verify customer identity merges across physical point-of-sale, e-commerce, and loyalty club databases.

Use this checklist when executing a customer master data consolidation or CDP identity resolution job. It validates contact formatting, deduplication confidence, and consent state preservation.

Template

Role: Principal Customer Data Platform (CDP) Architect with specialization in retail identity resolution and consumer privacy compliance.

Context

  • Merchant Brand: {{merchant_brand}}
  • Upstream Identity Silos: {{crm_data_sources}}
  • Identifier Precedence Hierarchy: {{matching_key_hierarchy}}
  • Data Cleansing Standard: {{pii_sanitization_standard}}
  • Regulatory Policy: {{consent_policy_version}}
  • Resolution Match Threshold: {{merge_confidence_cutoff}}

Task

Develop a pre-merge and post-merge data cleaning checklist to eliminate duplicate shopper profiles, format consumer identifiers, and preserve customer state for {{merchant_brand}}.

Method

  1. Validate standardized normalization across all email addresses, physical postal codes, and phone numbers in accordance with {{pii_sanitization_standard}}.
  2. Detect and isolate invalid placeholder phone numbers (e.g., 000-000-0000) and throwaway email domains across {{crm_data_sources}}.
  3. Execute deterministic match checks across primary keys specified in {{matching_key_hierarchy}} before running probabilistic models.
  4. Evaluate probabilistic merge candidates against {{merge_confidence_cutoff}}, flagging edge cases for manual review.
  5. Separate shared household accounts from unique individual identities based on first name and birth date discrepancy rules.
  6. Audit consent flags and opt-out statuses under {{consent_policy_version}} to verify that the most restrictive opt-out survives the merge.
  7. Check points balance, tier status, and transaction history continuity across merged survivor master records.

Constraints

  • Merge criteria MUST NOT consolidate records with active consent contradictions without defaulting to the opt-out state.
  • Verification items MUST include explicit roll-back and survivor record assignment rules.
  • Exclude all unmasked live PII examples from documentation output.
  • Format checklist items into clear inspection gates with defined pass metrics.

Output format

Generate a four-phase Markdown verification checklist: Phase 1: Contact Field Standardization, Phase 2: Deduplication & Merge Evaluation, Phase 3: Privacy & Preference Reconciliation, and Phase 4: Master Record Integrity Sign-off. Use checkbox format (- [ ]) with columns or inline labels for 'Verification Objective', 'Validation Logic', and 'Threshold'.

Self-review

  • Ensure all variables ({{merchant_brand}}, {{crm_data_sources}}, {{matching_key_hierarchy}}, {{pii_sanitization_standard}}, {{consent_policy_version}}, {{merge_confidence_cutoff}}) are referenced.
  • Confirm clear guidance is given for surviving master record selection.
  • Check that consent preservation rules strictly adhere to the constraints.
AuraScore breakdown
79/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering8/12 · Adequate

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness5/5 · Strong

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

data-analytics
data-cleaning
retail-consumer-goods
cdp
identity resolution
data cleaning