Menu → Attributes
Who can access ?
Roles | Access |
👑 Admins | Full access |
🛠️ Editors | Read only |
👀 Viewers | Read only |
🔒 Restricted Viewers | Read only |
What are the different visibilities of attributes ?
There is 3 visibility types :
🌐 Public (by default) | Attributes are visible by all logged in users, and can be used as filters or in reports. |
👀 Visible for assigned users | The value will only be shown :
|
🔒 Hidden | The value will not be shown whatever the page (live feed, reports,…). |
The visibility applies to the current team and all its sub teams, except if it was explicitly changed for one of the subteams.
What are Sensitive attributes ?
Settings > Organization > Attribute security
Sensitive attributes are encrypted in our database with the AES-256-CBC algorithm. You cannot filter on their value, but you can retrieve it
through the API
in an email template (in order to insert the attribute value in an email campaign)
when the visibility setting is Visible to assigned users (see above).
In those situations, the value is decrypted only at the moment it’s needed—it is never stored in decrypted form in the database.
How can I mark an attribute that I use often ?
The “Favorite” attribute type makes your attribute usage across the platform more efficient. It will be directly recommended first when using Feedier.
Why assigning owners to Attributes ?
The “Ownership” attribute type allows you to assign email owners to attribute values.
Then in autopilot you can trigger the automation you want and send notification to the owner.
What are NLP attributes ?
NLP attributes are pieces of information automatically extracted from text responses (verbatims) using AI. There are by default activated on the plateforme. They allow users to anonymize sensible informations.
They are replaced by *** inside the Feedbacks items.
Attributes | Meaning |
nlp_personal_name | Names of people mentioned in the text (e.g. "John Smith") |
nlp_localization | Locations mentioned (e.g. "Paris", "New York") |
nlp_product | Products mentioned (e.g. "iPhone", "Model X") |
nlp_brand | Brands mentioned (e.g. "Apple", "Nike") |
nlp_number | Numbers/quantities found in the text (e.g. phone numbers, amounts) |
How to create a Referential on my attribute values ?
The Referential lets you automatically enrich feedback with additional context by mapping one key attribute to several derived attributes. Define the mapping once — Feedier handles the rest, on all past and future feedback.
How it works
You pick a key attribute already present on your feedback (e.g., store_code) and map it to derived attributes (e.g., store_name, region, country):
store_code (key) | store_name | region | country |
MAD-001 | Madrid Atocha | South Europe | Spain |
PAR-042 | Paris Gare du Nord | France | France |
When feedback arrives with store_code = MAD-001, Feedier automatically attaches the store name, region, and country — visible immediately in all reports and dashboards.
Key behaviors
Retroactive — any change to the referential updates all existing feedback instantly
Organization-wide — not scoped to a single team
No time-based validity — a change applies across all history, not from a specific date
Autopilot is not re-triggered — existing feedback won't refire workflows when referential values are updated later
One value per key — multi-value mappings per key are not yet supported