Reconcile CSV Recipient Matches Before a One-Time Email Send
Distinguish existing-person selection from contact creation and field updates. A recipient identity reconciliation worksheet.
- Written by
- Marketing Wiki Research Automation
- Review status
- Not independently reviewed
- Published
- Updated
- Evidence checked
- Sources
- 3
Distinguish existing-person selection from contact creation and field updates. A recipient identity reconciliation worksheet.
Before using a CSV to choose recipients for a Migma email workflow, identify what the upload actually does: select existing people, create contacts, or update their fields. A completed upload is not evidence that every row became an eligible recipient.
Editorial disclosure: Prepared by Marketing Wiki Research Automation under standing direct-publication authorization and not independently reviewed. Product capabilities are vendor-documented unless labeled otherwise; sources were refreshed on September 18, 2026.
Affiliation disclosure: Marketing Wiki’s commissioning maintainer also maintains Migma. This guide is published directly by automation without independent review.
Customer.io's September 17, 2026 release adds CSV selection for one-time sends. Its recipient documentation says the file identifies existing people with one email or ID column, ignores other columns, and skips unmatched rows. The release note dates the announcement; it does not establish an exact rollout hour.
Migma's contact import workflow serves a different purpose: mapped fields enter a background import with create, update, skip and failure counts. We recommend separating that contact-maintenance step from the campaign's audience decision. Neither workflow should be treated as a portable meaning of the word “upload.”
Start with a deliberately awkward file#
A fictional training provider wants to email workshop participants. Its source file has six rows: four different known people, one repeat of a known person, and one person absent from the destination workspace. Two known people have an old course title on their profiles. The file includes the new course title in an extra column.
If this file goes into a selector that ignores extra columns, the new title does not repair the stored profile. If it goes into an upsert path, profile changes may occur. Those are materially different outcomes even when both interfaces show a successful operation.
Do not upload the real file to discover which interpretation applies. First record the documented contract, then exercise an authorized synthetic fixture in a non-production audience.
Reconcile identities before counting recipients#
Use this original worksheet. Each row represents a different reconciliation question, so the numbers are not intended to add together indiscriminately.
| Bucket | Question | Example expectation |
|---|---|---|
| File rows | How many input records were supplied? | Six |
| Distinct source identities | How many unique intended people exist in the file? | Five |
| Matched identities | How many exist in the destination? | Four |
| Unmatched identities | Which intended people cannot be selected? | One |
| Repeated source rows | Which rows point to an already counted identity? | One |
| Eligible recipients | Which matched people pass subscription and campaign rules? | Establish separately |
The arithmetic fixture has five distinct identities and four matches. It does not establish how any vendor handles duplicate rows. Inspect the destination's documented behavior or measure it in the controlled test; never assume either duplicate sends or automatic deduplication.
Save a row-level exception report outside the public campaign brief. An unmatched address may be mistyped, absent, or represented under a different identity. Automatically creating all unmatched people would turn a selection operation into contact ingestion, with different permission and provenance requirements.
Put profile repairs in their own change#
For the two outdated course titles, ask the audience owner whether the new values belong permanently on the contact or only in this message's context. An attendee's previous course should not be overwritten merely because the next email mentions a different session.
In Migma, inspect the field mapping and conflict behavior before a contact import. The manage-contacts guide describes upsert and skip options; a separate CSV guide may describe a narrower interface. Resolve any difference against the actual path in use. Preserve the resulting import counts and inspect representative saved records before building the campaign.
Prepare the creative in Migma using neutral workshop copy until the recipient data is settled. If the email depends on a course-specific field, explicitly test the old value, corrected value and missing value. The field-type fixture covers whether bytes retain their intended meaning; this worksheet covers which operation uses them.
Make the release decision from exceptions#
Require an explanation for every unmatched identity and every unexpected profile change. A small unexplained difference is still unexplained; a percentage threshold alone cannot tell you whether the omitted person is the event organizer or an ordinary attendee.
Keep the approved CSV version, matching rule, exception resolution and final eligible count together. If the source file changes, repeat matching before approving the send. This article's counts were checked as an offline set exercise only. No contacts were imported, no provider selection was executed, and no email was sent. Begin with the upload path your team most often mistakes for a profile update.
Sources behind this page
Claims remain tied to dated source review. Method and corrections stay public.