Skip to content

Protect numeric-looking source identifiers from coercion

Keep long, leading-zero and exponent-like identifiers intact through tables and exports.

Guides

Protect numeric-looking source identifiers from coercion — what should the operator record?

A spreadsheet can reinterpret an identifier as a number, remove leading zeroes or display rounded digits. The table may look clean while its joins are broken. Treat identifiers as opaque strings from ingestion through export and test the specific representation before sharing it with another operator.

Practical operations check
Keep long, leading-zero and exponent-like identifiers intact through tables and exports.
Mark native_id and local crosswalk fields as strings in the projection contract; do not infer numeric type from their appearance. Scientific notation, thousands separators and a numeric display format are all warning signs for key coercion; none should be repaired by arithmetic or visual approximation.
Record the checked import/export path and block unrecoverable identifiers from joins until their actual values can be revalidated legitimately. A checked JSON representation may preserve string type more clearly than a spreadsheet view, but still inspect any conversion step used before the destination application receives it.
01 / Guides

Identify vulnerable fields

Mark native_id and local crosswalk fields as strings in the projection contract; do not infer numeric type from their appearance. Scientific notation, thousands separators and a numeric display format are all warning signs for key coercion; none should be repaired by arithmetic or visual approximation.

02 / Guides

Preserve raw values

Retain an authorized raw copy and use import controls that prevent automatic numeric conversion in the chosen table application. Importing a CSV by double-clicking may differ from an explicit text-column import, so test the actual path recipients are expected to use rather than a safer unused path.

03 / Guides

Test coercion cases

Use labeled fixture strings 00042, 12345678901234567890 and 1E12, verifying that each survives an export and reimport unchanged. Compare recovered strings character for character, including length and leading zeroes; a visually rounded cell can still hide a different underlying value or a damaged one.

04 / Guides

Reconcile existing damage

Compare suspicious cells against the original receipt rather than guessing lost digits or restoring leading zeroes from a neighboring row. If the original permitted receipt is unavailable, leave the damaged identifier unresolved; neighboring records or assumed numeric sequences cannot reconstruct an opaque source key reliably.

05 / Guides

Approve the safe format

Record the checked import/export path and block unrecoverable identifiers from joins until their actual values can be revalidated legitimately. A checked JSON representation may preserve string type more clearly than a spreadsheet view, but still inspect any conversion step used before the destination application receives it.

Keep the boundaries clear.

  • Keep long, leading-zero and exponent-like identifiers intact through tables and exports.
  • Operational record: Identify vulnerable fields. Context badges do not certify access.

Start with a read-only check: ask your agent to report actual source coverage before collecting records or exporting an audience.

Primary references

Documentation reviewed 2026-11-15. Product capabilities, client availability and plans can change. Third-party marks identify their products, not an affiliation.

Know what social media shows about you.
Start with a question.

Get started
montage