Importing Data

| Format | Notes | |--------|-------| | CSV / TXT | Standard comma-separated values. Most systems export this. | | XLSX | Microsoft Excel spreadsheets. First sheet is used. | | PDF | Text-based PDFs containing a table. Scanned images are not supported. |

1. Open the Import dialog from Pipeline or Settings.
2. Choose Leads or Sales.
3. Upload your file (CSV, XLSX, or PDF up to 10,000 rows / 5 MB for PDF).
4. OfficeHubb uses AI to normalize column names and data automatically.
5. Preview the results: each row is marked Ready, Invalid, or Duplicate.
6. Optionally toggle Skip duplicates (on by default) and Send to archive (leads only).
7. Use Dry run to validate without saving anything.
8. Click Import to commit the records.

• Leads: matched by email, then phone, then name.
• Sales: matched by policy number + date, or by agent + premium + date.

Duplicates are flagged during preview so you can choose to skip or import them anyway.

In Settings → Recent Imports you can revert any past import (leads or sales). This permanently deletes the records created by that job.

• Column names don't need to match exactly — the AI maps them automatically.
• If the AI can't parse a row, it shows up as "Invalid" in the preview.
• PDF extraction works best with cleanly formatted tables. For complex layouts, export to CSV first.