
Commercial Property CRM Import Template: Keep the Right IDs
Prepare a commercial property list for CRM import with separate property, entity and contact IDs, a mapping template and a checked synthetic example.

Commercial Property CRM Import Template: Keep the Right IDs
The short answer
Stage a commercial property list before uploading it. Preserve a stable property ID, a separate legal-entity ID and any separately verified person ID. Map those records and their documented relationships to the destination, assign an explicit create, update or hold action, then reconcile the result by identifier.
NextAutomation builds CRE sourcing and owner-resolution workflows around this handoff. The useful result is a reviewable property-to-owner relationship that survives delivery into your team's system. The CSVs below let you inspect the mapping before discussing an implementation.
Define the record you are importing
An export row can contain a property, an owner name, a mailing address and a suggested contact. That does not make it one person record. A legal entity can own several properties. One property can have several documented parties. A person can have different roles for different entities.
Start with your destination's actual object model and permitted use. Where a CRM has no property object, agree how properties and relationships will be represented before the upload. A custom property object, an external reference or an approved association design are choices to validate, not capabilities this template assumes.
First resolve uncertain duplicates using the owner-list deduplication method. Then use the owner-research evidence handoff to preserve what each relationship supports. Import mapping carries those decisions forward; it should not silently redo them.
Use separate keys for separate questions
| Field | Question it answers | Import treatment |
|---|---|---|
| source_row_id | Which original observation produced this row? | Immutable within the source batch. |
| property_id | Which property does the record concern? | Stable internal key; retain jurisdiction and parcel reference separately. |
| entity_id | Which legal entity is named in the evidence? | Separate key; a name or mailing address alone is not a unique identifier. |
| person_id / role | Which verified person and documented role? | Leave blank when unresolved. Do not turn a registered agent into a decision-maker by inference. |
| destination_id / action | Which existing destination record, and what should happen? | Use reviewed mapping: create, update or hold. |
| source_ref / observed_at | Where and when was this relationship observed? | Retain with the handoff so a reviewer can reopen it. |
For a platform-specific example, HubSpot's import documentation, checked November 8, 2026, explains file requirements and unique identifiers. Its rules depend on the imported objects and features. Check your actual provider's current documentation and permissions before adapting the exercise.
Work the synthetic import batch
All names, identifiers, records and relationships in this exercise are synthetic. There are four source observations, three properties and two legal entities. No real contact details are included.
| Source rows | Mapping decision | Expected result |
|---|---|---|
| R-01: P-101 → E-01 | Create a property and its documented owner relationship. | One P-101 property, associated with E-01. |
| R-02: P-102 → E-01 | Reuse E-01; create the second property. | P-102 joins the same entity without becoming P-101. |
| R-03: P-103 → E-02 | Keep E-02 separate despite its shared mailing address with E-01. | A third property and a distinct second entity. |
| R-04: P-101 → E-01, proposed person unresolved | Hold the proposed contact; retain the property observation. | No invented person and no fourth property. |
The staging file retains every row, including the held proposal. A blank person field means unresolved here, not that the owner has no reachable representative. The expected output contains three property records, two entity records and three distinct property-to-entity associations. It creates zero person records.
Keep original source rows.
Map identity separately.
Hold the unresolved person.
Check the import before accepting the handoff
Use a sandbox or an explicitly authorized test destination. Save the original mapping, batch ID and pre-import record snapshot. Check required fields, identifier formats, relationship direction and what the provider does with blank values. A blank can mean leave unchanged or clear a value depending on the operation; do not guess.
Import the reviewed subset, export or inspect its results and compare them with expected-results.csv. Check associations as well as counts: three correctly counted properties linked to the wrong entities still fail. Confirm the held person stayed out. Repeat the same batch in the test environment and verify that stable keys prevent new duplicates under your configured import mode.
Plan correction before a production run. Keep the created and updated record IDs, prior values where available and a reviewed rollback or remediation procedure. A provider may not offer a complete one-click undo. Retain licensing and contact-use restrictions with the delivery; a CRM upload does not expand the right to resell or contact anyone.
Download the mapping exercise
Download staging.csv, expected-results.csv and the field and acceptance notes. These are editable teaching files, not a production connector or a list of prospects.
Bring a sample of your own permitted export, your destination fields and a known exception to the platform walkthrough. The implementation discussion can then focus on the exact records, review rules and export your team needs.
Frequently asked questions
What belongs in a commercial property CRM import template?
Keep source-row, property, legal-entity and person identifiers separate. Include the source reference, observed date, relationship, destination identifier and proposed import action. Your CRM's supported object model determines the final mapping.
Can I match owners using only a mailing address?
A shared mailing address does not establish that two legal entities or people are the same. Preserve distinct identifiers and route uncertain relationships for review before importing them.
Does this CSV work as a direct upload to every CRM?
No. It is a vendor-neutral staging exercise. Adapt and test its fields against the actual destination's supported objects, identifiers, associations and import permissions.
How do I know a repeated import will avoid duplicates?
Test the same approved batch twice in a sandbox or other authorized test environment. Compare record counts and associations by stable identifier; a successful upload notice alone does not establish safe repeat behavior.
