Importing from Close

Use Business → Migrations to create a Close migration. Connect with a read-only Close API key, review the scan, choose the destination client, and start the import.

Close migrations include user roster entries and custom-field definitions from leads, contacts, and opportunities. Fields are kept separate by their Close record type, so similarly named fields do not overwrite one another.

The API-key field always opens blank and does not display a stored key. After connecting, Myra shows only the connection status and the key's final four characters. Paste a new key only when connecting or replacing the saved key.

API keys are encrypted before storage and are never shown again. Call metadata and available transcripts can be imported, but recorded audio files remain in Close.

Dry runs start automatically

Once you pick a destination client and leave at least one type of data ticked, the dry run starts on its own — you no longer need to click Start dry run. Progress and per-batch counts appear as they land.

If automatic startup is delayed, Run dry run remains available as a manual fallback. While planning is active, the control shows live progress; after planning finishes, approve the plan before Start migration becomes available.

If you change the destination client or the selected data afterwards, the previous plan no longer matches, so it is discarded, any approval is cleared, and a fresh dry run starts. Start dry run stays available as a manual re-run.

The dry run accounts for every type of data you ticked, including the ones that plan nothing. A type with no records reads "0 planned" with a short reason rather than being left out, so a missing line never means a silently skipped import. Users are always listed as reference-only: they are cached for mapping and no team members or logins are created.

Approval never happens automatically: a person must still review the report and click Approve plan before anything is written.

Dry runs, stalled runs and failed batches

A dry run must finish completely before it can be approved. If any batch is still waiting, the run shows as Stalled with a Resume dry run button, and approval is blocked — a partial report is never presented as reviewable.

Each batch shows what actually happened. A batch where records could not be imported reads, for example, "0 imported · 63 failed" in amber or red rather than a green tick, so an empty result is never confused with a failed one.

If a real import fails, expand its progress panel and use Retry all failed (or retry the affected data type). Completed batches are left alone; only failed batches are queued again.

Rolling back an import

Rollback removes only the records the import itself created. Anything you or your team created or changed afterwards — new contacts, replies, bookings — is kept, and imported contacts that have been edited or worked since the import are kept too.

The preview shows three figures before you type ROLLBACK: how many imported records would be removed, how many are being kept because they changed since the import, and anything that cannot be removed automatically because it carries no import marker. Rollback stays limited to the clients this migration wrote to, and is available for 30 days after the run finishes.

Skipped records are always explained

Every run reports three numbers per data type: imported, skipped and failed. A skipped record always carries a reason (for example "duplicate note already imported" or "no matching contact"), so a total that is smaller than the source count is never left unexplained. The users row is shown as "N users catalogued - no accounts created" rather than "0 imported", because it only caches the Close user list for owner mapping.

Calls without a phone number on the activity