Migration & data
Plumbing Customer Import Spreadsheet: Clean Fields Without Losing Data
Prepare a customer import by preserving identity, separating service and billing details, and documenting every field mapping. Keep unknown values blank or flagged rather than inventing information to make an upload look complete.
In this guide
Keep an untouched source copy
Save the original export before editing column names or combining fields. Use a working copy and a mapping note that explains what changed.
A preparation spreadsheet is not automatically the destination's accepted schema. Check the current import tool's required fields, supported formats, and validation behavior before uploading.
GoPlumber currently describes CSV/XLSX customer and job imports. Verify the actual field support for your chosen import path; do not assume this example is a universal direct-upload file.
Use a staging structure
source_system,source_customer_id,customer_name,billing_contact,billing_email,billing_address,service_property_id,service_address,unit,contact_phone,review_status
SOURCE,CUST-EXAMPLE,Example customer,Example contact,,,PROP-EXAMPLE,Example address,Example unit,,needs_review
The row is fictional. Keep sensitive information limited to what the workflow needs and store the files with appropriate access controls.
Map roles instead of guessing from names
A source may contain company name, contact name, customer name, and display name. Those fields can describe different things. Review representative records before combining them.
Likewise, a billing address and service address are not interchangeable. A business customer can receive bills at one location while work occurs at several properties.
| Source field | Review question |
|---|---|
| Company/customer name | Is this the billing entity or a person's display name? |
| Contact name | What role does this person have? |
| Address | Billing, service, mailing, or unknown? |
| Phone | Whose number is it, and is the formatting preserved? |
| Notes | Is important context retained without exposing unnecessary data? |
Preserve identifiers and formatting
Treat customer and property IDs as text. Preserve leading zeros and avoid spreadsheet conversions of long numbers into scientific notation.
For phone numbers, do not guess a missing country code or silently remove meaningful extensions. Keep the source value and a reviewed normalized value when cleanup is needed.
For dates, document the interpretation rather than assuming a value such as 03/04 has the same meaning in every export. Use an unambiguous format in the working file once the source meaning is confirmed.
Review duplicates carefully
Search for likely duplicates using several signals, such as a source ID, verified contact detail, and property context. A shared surname or similar company name is not enough to merge records safely.
Do not split one customer into multiple identities simply because they have several service addresses. Preserve the intended customer-to-property relationship where the destination supports it.
Keep uncertain matches in a review queue. The duplicate-cleanup guide provides a separate process for deciding them.
Test before importing the whole file
Include a normal record, a multi-property customer, a missing optional value, a long note, and a likely duplicate. Inspect the destination records after the test.
Check whether rejected rows remain available with clear reasons. A row that fails validation should not vanish from the reconciliation count.
Compare source rows, imported records, updates, duplicates, and exceptions. Keep unmapped fields in a documented archive rather than dropping them to satisfy a narrow import schema.
The goal is not merely a tidy CSV. It is a customer record that still means the same thing after it enters the new workflow.
Sources and editorial notes
Product and documentation references checked September 29, 2026. GoPlumber publishes this guide about its own product category. Examples and templates are original illustrative material, not customer case studies or market benchmarks. Confirm current vendor terms before acting.