Iranian Contact Data Normalization Pipeline

Context

Mar 2025 · Operational data workflow · Maturity: Operational.

Problem

Raw contact exports contained inconsistent formats, invalid values, and duplicates that reduced downstream messaging quality.

Users

Operational messaging workflows that needed clean contact inputs.

My Role

Personal data-product design and implementation.

Product Decisions & Responsibilities

• Defined the clean-data contract and readiness checks for downstream messaging.
• Designed the pipeline around repeatability and operational handoff.

Technical Approach

• Built normalization, validation, deduplication, and export stages for large contact datasets.

Constraints

• Iranian phone-number normalization edge cases.
• Repository snapshots may not contain every behavior from the operational version.

Product Flow

1. Import raw exports
2. Normalize fields
3. Validate contacts
4. Deduplicate
5. Generate clean output

Outcome

• Operational version tested on approximately 1,000,000 contact records.

What I Learned

• Data quality should be treated as a product contract, not an ad-hoc cleanup step.

Evidence Boundary

Next Iteration

Extend the product only after the next decision can be tied to stronger user, operational, or business evidence rather than adding features for their own sake.