If a test email shows variables resolved to data from the wrong lead, the wrong company, or the wrong list, the cause is almost always one of a small number of patterns. Here is how to diagnose and fix each one.
How variables resolve
Every variable in a sequence has three levels of source data:
Lead fields — name, company, jobTitle, industry, etc.
Custom fields — anything else you stored on the lead, including from CSV imports and enrichment.
AI variables — generated by the Sequence Assistant or Artemis using lead and company context.
At send time, BrandJet looks up the variable in that order. If lead.company is empty but custom.company has a value, the custom field wins. Most wrong data issues come from this fallback path.
Common causes
1. Two columns mapped to the same field
If your CSV had both company and company_name, one was mapped to the lead field and one became a custom field. Variables in the sequence using {{company}} might resolve to either depending on which is populated. Pick one source per import and remove the duplicate column.
2. Test sends use the first lead in the list
Test emails default to the first eligible lead in the campaign. If that lead has incomplete data (or weird data from a CSV row you forgot to clean), the test looks broken even though most leads are fine.
Pick a specific lead for testing — the test dialog lets you choose.
Or add a clean test lead to a separate test list.
3. Stale enrichment
Enriched fields can age. If a contact changed jobs, the company on the lead might be the previous employer. Re-enrich the affected leads from Leads → select rows → Enrich.
4. Conflicting AI variables
The Sequence Assistant generates fields like {{ai.openingLine}} by reading the lead. If lead context is thin, the AI fills in plausible sounding but generic guesses. Look at the AI variable preview on the lead before sending. You can regenerate per lead or globally.
5. Wrong campaign assigned to lead
Less common, but if a lead was bulk-moved between campaigns, its remembered variables can be from the previous run. Resetting the step on the lead recalculates variables on the next send.
Diagnostic checklist
Open the lead in Leads and look at the field values directly.
Open the campaign sequence editor and click Preview as lead to see how variables render for that specific row.
If a value is wrong on the lead itself, it is a data problem (CSV mapping, enrichment, manual edit).
If the lead has the right value but the email shows something else, it is a variable resolution problem (custom field winning over lead field, AI variable cached, etc.).
Set fallbacks for missing variables
Inside the sequence editor, every variable accepts a default: {{firstName | there}}. Use fallbacks for any variable that might be empty, especially for AI variables and custom fields. This avoids both blank gaps and embarrassing Hi comma sends.
When to open a ticket
If the lead clearly has the right data but the rendered email is still wrong, send us the campaign name, the lead email, and the timestamp of the test. We can replay the variable resolver against that exact lead and tell you which source returned the bad value.