Goal: understand how Feedier is organized, and set up a first complete scope: sources, teams, Text Analysis, report.
Who it's for, and before you start
Who it's for | The administrator rolling out Feedier in their organization |
Prerequisites | Admin access to your workspace |
To prepare | The list of your feedback sources and your team structure |
Time | Half a day, not counting import times |
Key concepts on one page
Concept | In one sentence | Reference article |
Workspace | Your Feedier environment, isolated from all others | |
Team | A data scope; parent teams see their sub-teams | |
Source | A set of feedback with the same structure (a survey, Google reviews…) | |
Feedback | A response: ratings, texts and attributes | |
Attribute | A piece of context (store, country, customer) | |
Verbatim | A free-text response, analyzed by AI | |
Taxonomy, segment, topic | Your classification plan for verbatims | |
KPI | An indicator: NPS, satisfaction, sentiment, or your own formula | |
Filter segment | A saved, reusable filter |
Watch out for the word "segment": in Text Analysis, a segment is a group of topics. In filters, it's a saved filter. The context always tells you which one is meant.
The steps
1. Organize your teams
Mirror your organization: one team per department, region or entity, with sub-teams if needed. Set the language and time zone for each one. See Users, roles and teams.
2. Connect your sources
Start with the source richest in verbatims: survey, Google reviews, tickets. Place each source in the lowest team that needs it. See Sources and imports.
3. Check your attributes
Make sure the useful attributes (store, country, customer type) are present, mark the most important ones as favorites, and hide or protect personal data. See Attributes and referentials.
4. Create your first taxonomy
Follow the guide Build your topic plan from the customer journey.
5. Invite your users
Give each person the most restricted role that's enough for them, and check the modules enabled per role in Access management.
6. Create your first report
A simple report is enough to start: a KPI chart, a benchmark by entity, the main topics. See Create a recurring report for leadership.
Common mistakes
Mistake | Consequence | What to do |
Putting all sources in the root team | Sub-teams see nothing | Place each source at the lowest useful level |
Forgetting the team language | Topics generated in the wrong language | Set it before the first taxonomy |
Importing the entire history without thinking | A costly first import | Choose a useful history period. See AI analysis budget |
Giving the Admin role by default | Settings changed by mistake | Start with Viewer or Editor |
Going further
Last reviewed: November 11, 2026, version 4.23.5.
