Output Data File Details

Data feed file format: feed files delivered as CSV, JSON or Parquet under a dated path

The following table describes the general details for the output files generated by Forager:

Feature	Supported	Notes
Location of files	Files in Amazon S3	Files can be unloaded directly to any user-supplied bucket in S3, then can be downloaded locally using AWS utilities.
	Files in Google Cloud Storage	Files can be unloaded directly to any user-supplied container in Cloud Storage, then can be downloaded locally using Cloud Storage utilities.
	Files in Microsoft Azure	Coming soon.
File formats	Delimited files (CSV, TSV, etc.)	Any valid delimiter is supported; default is comma (i.e. CSV).
	JSON
	Parquet
File encoding	UTF-8	Output files are always encoded using UTF-8, regardless of the file format; no other character sets are supported.

Example file

Person NDJSON file: data_0_0_0.ndjson Organization NDJSON file: data_0_0_0.ndjson

Row format

Each record in an NDJSON file is one JSON object with these top-level fields (see the example files above):

Field	Description
`id`	Forager record ID (integer).
`date_updated`	When the record was last crawled, for example `2026-07-24 22:42:18.804 Z` (UTC).
`schema_version`	Schema version of `serialized_data`, for example `2.0.4`.
`serialized_data`	The record itself, as described in the schema for the feed type and schema version.

File paths & names

File paths are constructed as so:

/forager_data_feed/<data_feed_id>/<date_created>/<data_feed_type>_<schema_version>/data_<partition_id>.<file_type>.<compression_type>

# Example:
/forager_data_feed/123/2025-10-15/person_2.0.4/data_0_0_1.json.gzip

Where the above variables represent:

data_feed_id: A Forager provided data feed ID, provide by our support team.
data_feed_type: Possible options are; person, organization, or job.
schema_version: Forager schema version that represents the versioning of serialized data for data feed type, this will be provided by your customer support rep.
date_created: Date when the export was created, Ex: 2025-10-16.
partition_id: Partition column values generated when exporting your files, Ex: 0_0_0, 0_0_1, etc.
file_type: This will be one of the following options; json, csv, tsv, parquet.
compression_type: This will be one of the following options; gzip, bz2, brotli, zstd.

Change the feed, format or compression to see where a file lands:

Serialized Data Schema

Schema located here.

Data Feed

Data Delivery

Legacy (v1)

Example file

Row format

File paths & names

Serialized Data Schema

​Example file

​Row format

​File paths & names

​Serialized Data Schema

Example file

Row format

File paths & names

Serialized Data Schema