First decide which spreadsheet is authoritative
One file contains customer phone numbers, another contains quotes, and the last call lives in a note. A CRM may help, but importing every row at once can move the confusion into a new system.
Start with a bounded scope: customers, open quotes and the next task. You do not necessarily need custom software or a new invoicing system. Assess an existing CRM importer when it covers the process.
1. Preserve the source and define the scope
Keep the original file and work on a copy. Identify the current source, its owner and the columns people actually use. Separate active work from historical records kept for reference.
A workshop could start with pending quotes and their customers while leaving old invoices in the existing system. Agree when spreadsheet editing stops, or how later changes will be reconciled. Do not delete history merely to simplify an import.
2. Separate customers, quotes and next actions
Fictional example: customer C-001 has quotes P-101 and P-102. The result should be one customer with two linked quotes, each with its own status and follow-up, rather than two customers with the same name.
Use stable source identifiers and retain the links. Names alone are not reliable keys. A shared email address does not prove two people are the same. The destination CRM determines the final matching rules.
- Customer: source ID, name, contact details and owner.
- Quote: ID, linked customer, document reference, amount and status.
- Next action: related customer or quote, owner and date. Do not invent missing dates.
3. Review possible duplicates before merging
Normalize spacing and formats to find candidates while retaining original values. Similar rows require review, not automatic merging. Agree who resolves conflicting contact details.
Keep identifiers and phone numbers as text where appropriate. Excel can remove leading zeros when it treats a reference as a number. Changing its appearance later does not reliably recover the source value. Check dates, decimal separators and blank fields as well.
4. Test difficult cases, not just clean rows
Include a customer with several quotes, a missing email, a possible duplicate and a pending task. Use a test environment when available and prevent automatic messages during validation.
Read the destination documentation: creation and update behaviour depends on identifiers and import options. Retrying without checking what already imported can create duplicates.
- Does C-001 retain both quotes with the correct amounts and statuses?
- Can the team find the next call and its owner?
- Which rows were created, updated, rejected or left for review?
- Can the imported records be identified and corrected without affecting unrelated data?
5. Verify the outcome with the people using it
Reconcile accepted, rejected and pending records with the source. Check counts, relevant amounts, relationships and selected cases with a business user. Retain the issue report and identifier mapping.
Agree a correction or restoration plan supported by the CRM before the final import. An import is not a live synchronization: if the spreadsheet keeps changing, that connection needs separate rules and scope.
What Adaris needs to assess the work
Initially describe the approximate number of files and rows, your current or preferred CRM, who will use it and the problem to solve. Do not send the customer database through the initial contact form. Column names and fictional examples are usually enough to begin.
Adaris can assess a review, controlled import or integration separately. Sample, acceptance criteria, deliverables, price and schedule are agreed before contracting. We work from Elche with businesses in Alicante and across Spain. Commercial conversations are in Spanish.
Related services
Sources
Are quotes scattered across spreadsheets?
Tell us which files you use and where follow-up gets lost. We define the problem and sample before assessing the work.