Published November 4, 2026.
Question and scope
When lead records are imported into the CRM, do the source, campaign, and channel values remain usable for reporting? An import can move thousands of contacts successfully while quietly collapsing several source values into a default or dropping the campaign field entirely. The result is a CRM that looks complete but cannot explain where the pipeline came from. This test asks whether the imported values match the values in the source export and whether the CRM applied any transformations. It does not judge whether the original source taxonomy was correct. The narrow purpose is to verify continuity from the source file to the destination record.
Population and unit
Start with a dated export of the import file and the CRM records created or updated during the import. Record the import date, the file name, the row count, the person who ran it, and the matching rule used. Keep both files unchanged for the comparison. The unit is one imported lead record. A record that was updated rather than created should be labeled separately because the overwrite behavior may differ. If a contact was merged with an existing record, keep the merge reference so the original source value can be traced. Exclude test records and records that were deleted immediately after import, but record those exclusions. A deleted record may still reveal a mapping problem and should not disappear from the review without a note.
Definitions and measures
A source value survives when the CRM record contains the same normalized value as the source export, or a documented equivalent that the reporting layer maps correctly. A value is transformed when the CRM changes it in a way the team did not document. A value is lost when the field is empty or replaced by a default. Test the source, campaign, channel, and any custom field used for attribution. For each field, record exact match, documented transformation, undocumented transformation, missing, or not comparable. Report counts by state rather than a single pass rate. Also record whether the import created duplicate contacts or attached the lead to the wrong company. A source value can survive while the record itself is linked incorrectly, and that problem will distort reporting just as much as a missing field.
Test design
Use a census for small imports. For larger files, select a documented random sample and add a targeted sample of records from each source and campaign. Freeze the sample before comparing the files, and report any later expansion separately. Have one reviewer compare the records and a second reviewer check a subsample. Record disagreements before resolving them, because a difference in what counts as a documented transformation can change the result. Keep the source export and the CRM extract in separate folders. Where the CRM displays a label rather than the stored value, export the underlying field or open the record in a view that shows the raw value. A friendly label can hide a broken mapping.
Interpretation limits
A passing test does not prove that the attribution model is correct. It shows that the imported values reached the CRM in a usable form. The team may still need to fix the source taxonomy or the reporting definitions above the field level. A failing test does not prove that the import vendor was careless. The mapping may have been ambiguous, the source file may have used a value the CRM did not recognize, or the team may have changed the field configuration after the import. Report the observed state and the evidence. Avoid using the result to judge the person who ran the import without reviewing the mapping instructions. The useful question is whether the import checklist includes a source-value verification step.
Review cadence
Run the test after every import that affects lead source or campaign reporting, and run a broader review each quarter. Keep the comparison sheet with the import log so a later reviewer can repeat the check. End each review with a decision record: add a value map, require a sample export before import, change the matching rule, or leave the process unchanged. Name the owner and the recheck date. Archive the files and the notes so the test can be repeated.
Sources
- Federal Trade Commission: Privacy and security
- National Institute of Standards and Technology: Privacy Framework
- Cybersecurity and Infrastructure Security Agency: Small and medium businesses
- U.S. Small Business Administration: Grow your business
- USAGov
Sources
- https://www.ftc.gov/business-guidance/privacy-security
- https://www.nist.gov/privacy-framework
- https://www.cisa.gov/audiences/small-and-medium-businesses
- https://www.sba.gov/business-guide/grow-your-business
- https://www.usa.gov/