All Collections
Digital Asset Management
Asset Migration
How to: Asset Migration (Metadata Sheet)
How to: Asset Migration (Metadata Sheet)

Learn about how to set up/adapt your metadata for migration to Frontify

Updated over a week ago

At this point, you’ve read about the Asset Migration process and have defined the scope of the migration with your Customer Success Manager, and probably already spoken with your former DAM provider. Now, it’s time to dig deeper into the metadata topic. This article tells you how you should, ideally, prepare the metadata sheet for us.

Important note:

We need the metadata in a specific structure to be able to run a successful migration. After you’ve discussed the asset migration with your Customer Success Manager, you’ll receive a link to a template showing the following metadata possibilities. The first four pieces of this metadata information are mandatory, the rest of them are optional.

If you are not yet familiar with Standard and Custom Metadata, please review this help article.


Mandatory Unique ID of the Asset File. If there are no IDs available from the former system, you can use sequential numbers, letters, a checksum, or anything else which is unique to each asset to be imported

Example: 1, 2, 3, a, b, c, d9673f3128fcfbd70d040f7dc18afbd8


Mandatory name of the file. This file name will be stored but it is not shown anywhere in Frontify. The Title (below) is what is displayed in Frontify for users. It is recommended that the file name does not contain special characters or accents (allowed characters are a-zA-Z0-9._-). The filename must have the correct extension appended to it.

Example: my_company_logo.jpg, Sound-file.mp4, 108765393.png


Mandatory location of assets in the AWS S3 bucket folder. If you do a manual export, the easiest way is to have a flat list in the S3 bucket, to make the assets assignable to their location in the S3 bucket, via filename. It is recommended that the location does not contain special characters or accents (allowed characters are a-zA-Z0-9._-) and must be an exact match to the asset's location in the AWS S3 bucket folder.

Example: my_company_logo.jpg, Images/my_company_logo.jpg, 30000/310000/315/my_company_logo


Mandatory title of the Asset. This title is displayed to the end-user (and is also the download name for the assets). If there is no title present in the metadata sheet, it will fall back to the filename without extension.

Example: official logo, sounds of the jungle, packaging


Unique ID(s) of the Asset(s) that the Asset should be related to. If an Asset is related to multiple other Assets, separate the related IDs with a | character. Note that relations are always bidirectional. Hence if, for example, asset ID #1 has a relation with asset ID #4, this relation will be displayed from both asset ID #1 and asset ID #4 (no need to set the relation in both assets). Relations to attachments are not possible.

Relations are being shown in the “relations” tab in the frontend.


Unique ID of the parent Asset for the Attachment. If you have an attachment associated with an asset (e.g., you have a PDF of a publication, but want to have the original InDesign package attached as Attachment), you’d need to add the ID of the parent Asset (the pdf file) in the line of the Attachment (child) asset (InDesign file).

Please note that only the following fields are valid for Attachments:

  • ID

  • Filename

  • Location

  • Title

  • Attached_to_ID


The ID of the parent asset, for which this file should be set as a custom preview (e.g. for zip files). Only gif, jpg or pdf file types are allowed to be used as Previews. Only one value is permitted in the Preview_parent_ID field.

Please note that only the following fields are valid for Preview assets:

  • ID

  • Filename

  • Location

  • Title

  • Preview_parent_ID

All other fields will be ignored.


The folder structure in which the Asset will be located in the backend of your library. This does not influence how the assets are shown to the end-user, in the front end. The folders are separated by a slash (/), so this structure Brand/Assets/Logo Mark would mean a subfolder Logo Mark within a subfolder Assets within the first level folder Brand.

Author (Creator)

The name of the person that created the original Asset (this uploader is shown in Metadata information). This is not mandatory but can be useful for keeping track of the original source of your Assets.


Descriptive text of the asset content. Shown in metadata.


Here, you can add all your Keywords associated with the asset. Tags should be separated by a pipe (|). So, for example, “Brand Assets | Logo | Logo Mark | Whitewould create 4 tags (Brand Assets, Logo, Logo Mark, and White). If you’re choosing to import EXIF file information, the keywords associated with the file will be imported along with any tags defined in the CSV’s tags column.


The Copyright status of the asset. This field only accepts the values “COPYRIGHTED”, “PUBLIC”, and “UNKNOWN”.


The Copyright Notice is a longtext field you can use for specific information about an asset’s copyright attribution.


Names of the licenses to be applied to the asset (separated by pipe character "|"). For example, “External Distribution License|Other license” would apply 2 Licenses (“External Distribution License” and “Other license”). Please make sure to create the Licenses in the destination libraries in advance.


You can define a date when an asset will expire (or has already expired). You can define in your Library if expired assets should automatically be removed from the frontend view of your Library. The valid format for the Expiry Date RFC3339_EXTENDED format. For example, 2026-01-01T23:59:00+01:00 would represent the first of January 2026 at 11:59 PM in the UTC+1 timezone.

Custom Metadata

You can map almost any data to a custom metadata field. Detailed information on custom metadata and different types (eg Multi-select and Single-select) can be found here.

While it may be tempting to add many custom metadata fields to your libraries, remember that it's easy to overwhelm your users and librarians with too much information. Try to strike a good balance. Consider moving some information into tags or splitting your assets into multiple topic-focused libraries.

To add a new custom metadata field, simply add a new column to the CSV sheet, where you specify the name and then the type of the custom metadata field in brackets in the first line. For example: "Department (Single-select)"

Supported custom metadata fields and their formatted examples (please replace the names with your own):

  • Department (Single-select)

    • users can select one value from a predefined list (eg Sales or Finance)

    • any value accepted (text, date, number etc)

  • Product line (Multi-select)

    • users can select one or multiple values from a predefined list (eg Sales and/or Finance)

    • if multiple values are to be applied, separate them with a pipe (|) character. eg: maritime | terrestrial | spacial. See the example in the Migration Template.

    • any values are accepted (text, date, number etc)

  • Notes (Text)

    • text only values are accepted

  • Detailed description (Longtext)

    • text only values are accepted

  • Launch date (Date)

    • use this datetime formatting: 2026-01-01T23:59:00+01:00 (see "expires_at" section above for more info)

  • Weight in kilos (Number)

    • number values only are accepted

More help articles about Asset Migrations:

Did this answer your question?