We have reviewed this again and determined that we do not plan on increasing the limit based on the risks that it would present to the majority of our customers.

We cannot support this out of the box for two reasons: first, too long values are not ideal because they become unreadable in filters, and second, if we were to split them ourselves, we would not know the correct logic for how to break the value apart, since it varies depending on what the data represents and how it is formatted.

There are other connected systems which also have the same limitation (one example would be Google VertexAI Search for Commerce). If we were to increase the limit this would then cause additional issues with any connected system such as this.

Instead it would be better to split up the data. For example with the use case previously raised regarding related model numbers (sometimes up to 400 characters in single attribute). A better approach is to split them across multiple attribute values, so they stay within the limit of characters. They can assign the same attribute type multiple times on the product, each with a different value and stay within the character limit.

We recommend resolving this at the project level and staying within the defined character limitation for the attribute value field.

As a CSM, I am submitting this enhancement request on behalf of Stellar Industrial and their systems integration partner, Apexon, who are integrating an external AD-based product data feed with Optimizely Commerce Connect.

Stellar Industrial's AD system supports attribute values up to 500 characters, but Commerce Connect currently limits the attributeValue table to nvarchar(255). When their data feed refresh job runs, any attribute value exceeding 255 characters causes the job to throw an error, resulting in incomplete and inaccurate product data in their catalog.

What we're trying to solve: Stellar Industrial needs their AD data feed refresh jobs to complete without errors, and for all product attribute values to be saved accurately in Commerce Connect regardless of length.

Desired outcome: Increasing the column size to nvarchar(500) would align Commerce Connect with the capacity of their source AD system, allowing data feed jobs to run reliably and ensuring product attribute data is fully and accurately reflected in Stellar Industrial's catalog.

This is currently a blocker for reliable product data synchronization and is impacting catalog accuracy at scale.

Thank you for submitting this request! While we do allow configuration of max character limits in the Application Dictionary for some fields; we do have a database nvarchar limit of 255 characters for Attribute field which cannot be easily increased at this time and could cause additional risks.

Do you frequently have attributes greater than this limit or is this an edge case for the "Suitable For Use With" use case?

There are current alternatives for this particular use case that could be approached by reaching out to an implementation partner customize it using the extensions table.

We do encourage facetable attributes to be as small and 'normalized' as possible. If these model numbers/data were normalized in ERP or integration layer it could be brought over as individual attributes. As individual attributes the users on the front end can interact with them with greater flexibility.