Data quality

Addresses, phone numbers, and names benefit from different handling rules. These guides focus on useful representations, preserved components, explicit context, and readable uncertainty. Start with the purpose of the record and avoid treating a more uniform string as proof that the underlying information has been independently verified.

Review transformations across the lifecycle, including exports and support tools. A suggested change should remain distinguishable from the original value when the workflow needs that distinction. Treat unfamiliar or unresolved inputs as a design case with a real next action rather than forcing them into a convenient but incorrect format.

Each guide includes a primary reference and links to related documentation. Use the articles as a review starting point, then align the recommendations with the actual methods, data, and policies of your implementation.