Export your data models and segments

Sync your model profiles and attributes into a SFTP server to enable personalization and rapid data retrieval in your apps and websites.

Overview

This destination service allows DinMo to insert new files or update existing ones in a SFTP server, based on your DinMo models or segments.

To use this service, follow these three steps:

1. Create a SFTP destination. Follow our step-by-step guide to establish this connection.

2. Create your DinMo model or segment representing the data you want to send to your SFTP server.

3. Activate your model or segment with the SFTP destination to start synchronization.

Every time the activation runs:

• If the file does not yet exist: DinMo will create it with all the rows in the query results

• If the file already exists: DinMo will overwrite it, with only the lines added since the last sync.

If you don't want to overwrite existing files, we recommend using the timestamp in your file names.

Activation Setup

Once the SFTP destination is configured, create an activation to begin syncing your data. To do so, go to the Activations tab and click on "New activation". You will be asked to choose the model or segment you want to send to your SFTP server and select your SFTP destination you've just created

Then, you can configure your activation:

• Indicate the folder where you wish to store your file. By default, the file will be sent to the home folder of the SFTP server.

Learn more about absolute or relative path here

The user needs to have access to the target directory and file with write privileges. A permission denied error message during a sync indicates the user may not have write permissions for both the directory or the file.

• Indicate the name you wish to give to your file.

If you don't want to override existing files, we recommend including timestamp variables in the filename. To do so, you just need to surround each variable with {}. DinMo supports these timestamp variables:

• {YYYY}: Full year (e.g., 2025)

• {YY}: Last two digits of the year (e.g., 25)

• {MM}: Month (01-12)

• {DD}: Day of the month (01-31)

• {HH}: Hour (00-23)

• {mm}: Minute (00-59)

• {ss}: Second (00-59)

• {ms}: Millisecond (000-999)

• {X}: Unix timestamp in seconds

• {x}: Unix timestamp in milliseconds

For example: {YY}-{MM}-{DD}_export will be 25-04-14_export.csv for the upload of April 14th 2025.

• Select file format. DinMo supports CSV, JSON (and new delimited JSON), XML and Apache Parquet.

  • For the CSV option, you'll be asked to choose the CSV delimiter and if you want to include the CSV headers.

• Indicate the type of run you would like to do, based on the result you would like to see in your file.

Check this section if you want to learn more about the Run Types and the Sync Modes

• Map all the DinMo attributes that you want to include in your file. You can of course rename any column you're syncing by choosing your "custom attribute name".

The example above shows how to export the age, name, phone_number and boolean is_active. These columns are mapped to new fields in the destination file as age, last name , phone and is_active. DinMo exports these fields to the new fields in the file and ignores all other columns from your model/segment.

Run Types and Sync Modes

When configuring your sync, you will be asked to choose the Run Type and the Sync Mode.

For SFTP activations, here are the available options:

Run Type	Description	Use Case	Behavior
FULL ONLY	Every sync processes all records from the source and exports a complete file (or several, if size limit is reached).	When the exported file must always contain a full extract, and incremental updates are not required.	No delta logic Every sync rebuilds the full export Recommended for Snapshot mode
FULL THEN DELTA	The first sync exports all records. All following syncs export only changed records, based on DinMo’s delta detection logic.	When exporting large datasets frequently and wanting to reduce file size or processing time.	Sync 1 → Full export Next syncs → Only changed records (new, updated). Compatible with INSERT, UPSERT, and DIFFERENCE modes

Sync modes determine how data is synchronized between your data source and destination. They control whether to insert new records, update existing ones, or both, and how to handle the synchronization process.

For the SFTP destination, here are the available options:

Sync Mode	Description	Use Case	Behavior
SNAPSHOT	Exports a complete snapshot of all records at the time of sync. Each sync can generate a new file with a timestamped name.	Useful for backups or systems expecting full “point-in-time” extracts.	Always exports the full segment / model Creates a new file with timestamp No incremental logic
UPSERT	Writes all updated or new records into the exported file.	Keeps SFTP files up-to-date with the latest source data.	New records → included in the exported file Existing records → included if, and only if, there are updated values Deleted records → simply not present in the file
INSERT	Adds only new records to the exported file. Existing records are never modified.	Useful for append-only files, such as historical logs or event tracking.	New records → added to file Existing records → ignored Missing records → no action
DIFFERENCE	Generates separate files for added, updated, and removed records between syncs.	For audit trails, incremental processing, or systems needing change-specific files.	Creates _added.csv, _updated.csv, _removed.csv (or other extensions) Each file contains only the relevant change type

Scheduling

Define how frequently your data is sent in your SFTP server.

Each scheduled operation performs a delta operation, inserting the rows that were added since the last sync. This helps optimize bandwidth and storage costs while keeping your exports fresh and relevant.

Troubleshootings

• How do I know where to send my file?

You can either input an absolute path (starting with /) or a relative one. Relative path will start from the home directory of the user connected. To target directly the home directory, input .

For example, consider the following directory:

/
├── foo
├── home
│   └── alex
│       └── baz

We are logged as user alex.

• To target the foo folder, use an absolute path : /foo.

• To target the baz folder, an absolute path is: /home/alex/baz.

You can also use a relative path. Relative paths start from the home directory of the alex user, and do not start with /. The home directory is the alex folder, so just input baz.

Keep in mind that relative notation depends on the user logged in, so if you change the SFTP user used in the platform, the files will end up in a different location on the server.

To target directly the alex folder in relative notation, you can input ., which means "here" in Unix pathing. To go back in the upper folder, use ..

folder            absolute path               relative path
─────────────────────────────────────────────────────────────────
/                 /                           ../..
├── foo           /foo                        ../foo
├── home          /home                       ..
│   └── alex      /home/alex                  .
│       └── baz   /home/alex/baz              baz