Essentials
Feedback
Unify feedback then use AI to extract useful quotes, sentiment, and themes
Feedback
TheFeedback page is the unified hub for all customer interactions to be auto-analyzed by AI and reviewed by your team. Every piece of feedback uploaded through the Sources page shows up as new pieces of feedback in the Feedback page.
In the navigation bar, you can see All Feedback to view feedback across all spaces or Feedback that is triaged into a specific space.
Every feedback has additional details that can be viewed by clicking on the Feedback column. This shows the detail panel with the full feedback content, metadata, and most importantly the set of customer quotes extracted by AI and highlighted as important.
Creating New Feedback
The vast majority of new feedback is added through theSources page by uploading DOCs, PDFs, and CSVs, directly integrating apps like Gong and Intercom, or through Zapier.
You can also manaully create new feedback in the Feedback page through the Create New action. Share the customer feedback, set the customer name, date, typing, and space to create a manual feedback.
Analyze Feedback
Every feedback added throughSources automatically gets analyzed after being added. You can also manually trigger the AI analysis by pressing Automations then Analyze Feedback.
Feedback Highlights
When analyzing feedback, Inari specifically looks to extract customer quotes that can be helpful for product development. This specifically means looking forRequests, Defects,Praises, and Learnings mentioned by customers.
Inari extracts all useful quotes, categorizes them into these feedback types, then semantically scores the sentiment of the quote between -10 and 10.
Editing Highlights
You can override the AI suggestions by clicking on each highlight and adjusting the sentiment score or recategorizing its typing. You can alsoApprove or Remove highlights by setting the ✅ or ❌ on each highlight. Approving a highlight lets Inari know moving forward to look for quotes similar to this to extract while Removing a highlight deletes the highlight and reduces the likelihood of similar quotes to be extracted later.
Create Highlights
You can manually create highlights by dragging your cursor across any quote then pressingCreate Highlight. This records the quote as a highlight, uses AI to categorize and score the quote, and makes it ready to use for insight and issue generation later.
Feedback Processing Status
Every feedback has 3 potential processing statuses:- New: The feedback has been recently added and parsed and is pending an analysis by AI.
- Processed: The feedback has been added, parsed, and already analyzed by AI for useful customer quotes.
- Spam: The feedback was filtered out because it was categorized as spam. This is an optional filter when connecting data sources like Intercom or Zendesk where high volumes of emails and support conversations aren’t necessarily product feedback.
Feedback Triage
You can assign orTriage a piece of feedback into a specific Space when adding a new data source or updating the Space value in the feedback’s detail panel.
Feedbacks that are Triaged into a Space will be used as context when generating new insights and issues in that Space. Existing insights and issues in a given space will also be auto-linked to feedback triaged into that space.
This allows for more concentrated and higher-quality insights and issues to be generated since triaging allows for higher-quality clusters to arise.
Personalizing Feedback Analysis
You can customize how the AI feedback analysis runs in Inari in theSettings pages. If you visit Feedback within Settings, you can set custom prompt instructions for how the feedback is analyzed, context around which types of quotes to extract, and instructions on which types of feedback to ignore when filtering spam.
Feedback highlights that are manually approved in each highlight card will also show up in the Feedback Settings page. You can set whether those highlights are used as examples of the types of feedback that you’d like Inari’s AI to extract moving forward.