Testing Whether Disposition Codes Survive a CRM Export

Research question and evidence scope

When a calling team exports its CRM records, do the disposition codes remain usable for reporting? An export can move every row successfully while collapsing several codes into a blank or a generic value. The result is a report that looks complete but cannot explain what happened on the calls. The test asks whether the exported codes match the source and the written definitions.

This test does not decide whether the disposition taxonomy is good or whether a caller used the right code. Those questions require the call record and the quality review. The narrow purpose is to verify that the export preserves the code values and the labels the team relies on. Keep the source reference with the observation rather than relying on a later summary.

The unit is one call record with a disposition code. A record with several attempts should still have one final disposition, and the earlier attempts should be kept separately if the CRM stores them. If a record was edited after the export, keep the export version and the edited version linked. State the observation window in the note so the reader can compare the next review.

Operating scenario and evidence

Start with the CRM export and the code list used during the review window. Record the export date, the filters, the row count, the person who ran it, and the code definitions. Keep the files unchanged for the comparison. Exclude test records and internal calls, and record the exclusions.

A code survives when the export contains the same value as the source record and the value maps to the approved definition. A code is transformed when the export changes it in a way the team did not document. A code is lost when the field is empty or replaced by a default. Separate unavailable evidence from a confirmed negative before drawing a conclusion.

Method and review criteria

For each sampled record, compare the exported code with the source record and the code list. 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 code label matches the value.

Check for codes that were retired during the period. A retired code may appear in the export with an old value and no definition. Record the retirement date and the replacement code so the reviewer can judge whether the record is usable. Name the owner of the next action so the finding does not become orphaned.

Use a census for small exports. For larger files, select a documented sample that includes each queue, each shift, and each disposition family. Freeze the sample before comparing the files. Have a second reviewer check a subsample and record disagreements before resolving them.

Authority and practical implications

Confirm that the customer identifier and the call timestamp match across the export and the source. A code can survive while the record is linked to the wrong call. Check the record count and the duplicate count as well, because a duplicated row can inflate the apparent use of a code. Preserve the original record when a definition changes and label the transition.

The result can support a change to the export template, the code list, or the reporting layer. It cannot prove that the calls were handled correctly or that the customer was satisfied. Pair the code test with a sample of call quality reviews before drawing a conclusion about performance. Report the limitation in the same note as the result rather than a separate memo.

A high match rate does not mean the codes are meaningful. The team may be applying a small set of codes to very different situations. Pair the export test with a review of definition clarity and caller calibration. That combination is more informative than either measure alone.

Limitations, conclusion, and sources

Report missing and transformed codes separately, and keep the limitations beside the finding. The output should name one process change, one owner, and one recheck date. A code list that is reviewed and tested is more credible than a dashboard that counts records without checking what the values mean. Use the same unit and window in the next review so the results remain comparable.

Run the test after any change to the CRM, the code list, or the export template. Repeat it at the next quarterly reporting review. Archive the export, the sample, and the coding notes so the next reviewer can reproduce the result without reconstructing the method. Record the definition and the sample size beside the result so the review can be repeated.

Sources: FCC consumer guidance on unwanted calls; FTC guidance on the Telemarketing Sales Rule; NIST Cybersecurity Framework; U.S. Small Business Administration guidance on managing business finances; U.S. Government Accountability Office internal-control guidance. Each source was checked on November 4, 2026 and is used for general context, not as a local benchmark.