Supported CRM input
Companies, contacts, deals, tickets, leads, or a qualifying portal custom object.
Blank in the U.S.-only workspace context, or a recognized United States value.
Five ASCII digits, or ZIP+4 that can be reduced to its first five digits.
The HubSpot owner id, from an active owner record—not a HubSpot user ID.
How ZIP values normalize
| HubSpot value | Result | Reason |
|---|---|---|
85001 | Valid: 85001 | Exact ZIP5 |
85001-1234 | Valid: 85001 | ZIP+4 reduces to ZIP5 |
00501 in a text property | Valid: 00501 | Leading zero remains text |
501 in a number property | Valid: 00501 | Whole numeric values are left-padded to ZIP5 |
| Blank | Missing | No routing location |
8501 in a text property | Invalid | Text must remain exact ZIP5 or ZIP+4; TerritoryLogic never guesses malformed text |
| Recognized non-U.S. country | Unsupported country | Current programs are U.S.-only |
Portal preparation checklist
- Create a HubSpot view for the object records you intend to test.
- For every selected object, choose or create an editable single-line text or number property containing ZIP/postal values. Text fields preserve ZIP+4 and leading zeroes. Number fields must contain whole values from 0 through 99999; TerritoryLogic pads them to five digits.
- Resolve duplicate, merged, inactive, or unexpectedly blank owners in HubSpot.
- Confirm the intended territory owners appear as active HubSpot owners.
- Install or reauthorize with a user allowed to approve the selected object's read/schema scope and owner-read scope.
What TerritoryLogic does not copy
The current sync does not retain raw street addresses or malformed ZIP strings. It retains only bounded object identity, current owner, normalized ZIP/country evidence, and normalized city/state fallback evidence needed by the configured program. It does not request sensitive or highly sensitive HubSpot scopes. Activities are outside the current routing release.
After correcting HubSpot
Open Workspace settings and request a full resync, then wait for it to complete before running a new simulation. A previous simulation remains bound to its original snapshot and must not be treated as evidence for newly corrected inputs.