Overview
Iris is your customer research assistant that reads every single customer review, picks out the important insights, and organizes the details into categories.
The core focus of the analysis is to dive deep into what the customers are really saying:
What specific problems are they trying to solve?
What made them finally make the purchase decision?
What are they absolutely loving about the product and why?
What features do they wish for?
How are they actually using the product in real life?
How does the product make them feel?
What physical and psychological benefits do they gain after using the product?
Generating Insights
To create a Brand Insight, go to the insights page and click "Generate Brand Insight."
This will display a popup where you can select a brand that you want to analyze.
Once a new Brand Insight is triggered, Iris will carry out the following process:
Step 1: Quality Score Check
Iris will read each review individually and determine the overall quality score. The score is based on the following factors:
Specificity of Feedback
Does the review mention specific features, use cases, or experiences?
Are there concrete examples rather than vague statements?
Understanding the 'Why'
Does the review explain motivations and reasoning?
Is there context that helps explain their experience?
Depth of Information
Are multiple aspects of the experience covered?
Does the review go beyond surface-level reactions?
Short reviews that don't contain detailed insights are filtered out or de-prioritized, while high quality reviews with detailed and useful feedback are identified and selected for deeper analysis in Step 2.
Step 2: Key themes identification
Once high quality reviews have been identified, Iris proceeds to extract key insights grouped by specific type of analysis.
This analysis and the pre-defined themes have been developed with top performance creative teams, based on the exact frameworks and workflows that they use.
Each theme serves a unique purpose in understanding your customers:
Theme | Type of Insight | Core Focus |
Praises | Specific positive feedback about concrete details of the product, brand, or service | Identifies what the customers explicitly love and appreciate |
Complaints | Specific negative feedback regarding the product, brand or service experienced after purchase | Identifies frustrations with product functionality or service |
Feature Requests | New features or improvements customers explicitly suggest | Identifies missing product functionality that would enhance their experience |
Pain Points | Frustrations or problems customers had before the purchase | Identifies the specific issues that motivated their search for a solution (the "jobs" the customers trying to get done) |
Goals | Functional aims customers hope to achieve by using the product or service | Identifies the desired outcome (independent of the product) |
Benefits | Explicit positive outcomes achieved by the customer after using the product or service | Identifies the real-world impact on the customers' lives (the transformational results they experienced) |
Anxieties | Pre-purchase worries or hesitations about buying the product (experienced during the consideration period) | Identifies the objections that almost prevented the purchase |
Discovery Channels | How customers discovered the product | Identifies discovery paths, marketing touch-points and the referral sources that worked |
Purchase Triggers | Specific events that prompted the search for a solution | Identifies the situational, biological, emotional, or social triggers that made the customer search for a product or service |
Other Solutions | Alternative products or services that the customer tried or considered | Presents the competitive landscape from the customer's point of view |
Step 3: Customer Quotes
Each extraction is accompanied by customers' exact words and phrases, creating a direct link to the "voice of the customer." Iris will preserve the authentic language your customers use.
For example if a review says "I love the apple flavor, but I wish the bottle had a handle to make it easier to carry," Iris would identify and extract the following insights:
Praise: "I love the apple flavor"
Feature Request: "...wish the bottle had a handle to make it easier to carry"
Step 4: Categorization
At the end of the analysis, Iris groups the key insights into meaningful categories based on patterns in the actual feedback.
These are not pre-defined categories — they are organically created based on what the customers are saying.
Pattern Recognition
Iris identifies when multiple customers mention similar themes
Groups form naturally around common experiences
Categories emerge from the data, not from assumptions
Dynamic Evolution
Categories adapt as new more reviews are analyzed
Emerging trends create new categories
Outdated issues fade away naturally
Your categorization stays current with customer sentiment