Prepare the data
Work from clean, flat CSV files.- One Account row per organization.
- One Contact row per individual, with a column that identifies the related Account.
- Normalize names and values before importing.
Start with a dry run
A dry run checks column mappings, field types, duplicate or ambiguous match keys, and unresolved links — without writing anything. Read the results before you proceed. For a large or unfamiliar data set, approve a small representative sample before importing the whole file. Bulk creates are quick to make and tedious to unpick.Run the import
1
Import Accounts first
Accounts have to exist before Contacts can be linked to them.
2
Import Contacts
Resolve the Account relationship from the designated source column.
3
Read the final report
It identifies records created, records matched to existing data, records skipped, and rows requiring correction.
If the import is incomplete
Don’t start over with a new batch process. Inspect the log, correct the source carefully, and use the documented recovery path. This is what the match-or-create model buys you: a corrected rerun resolves against what’s already there rather than duplicating it. Starting fresh throws that away. See the batch pattern.Full migrations
For a complete migration — source profiling, field mapping, duplicate handling, sampling, reconciliation, and recovery — use Parley’s migration workflow alongside the CLI importer. See Migration. Growth and Enterprise plans include guided client data import with our team. If you’re coming off a legacy case management system, talk to us before you start: [email protected].Creating records and uploading documents don’t draw credits, at any volume. See How credits work.
