Keyboard shortcuts

Press or to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Sentiment Analysis

Feature Overview

  • Provide two distinct interfaces:
    • Admin Dashboard – Full view of sentiment data for all restaurants in the platform.
    • Partner Portal – Scoped view of sentiment for one restaurant or specific branches.
  • Displays sentiment breakdowns (Positive, Neutral, Negative), trends, and sentiment distribution filters by dynamic time ranges.
  • Enables internal and partner stakeholders to monitor sentiment performance, detect early warning signs, and celebrate positive momentum with data-driven insights. Design Link : https://www.figma.com/design/ymdLN2ZKzfI0oyX4kps2GQ/Partner-Portal-Desktop?node-id=16748-70&t=wysc60gqnZV98raF-1

Problem Statement

  • Admins need centralized visibility into customer sentiment trends across all restaurants for timely operational decisions.
  • Restaurant partners need focused sentiment analysis on their restaurants to manage service quality and marketing opportunities.
  • Without structured analytics and granular filters (especially for real-time or weekly insights), sentiment performance is difficult to monitor and react to.

Goals & Objectives

Business Goals:

  • Improve sentiment transparency across HungryHub’s ecosystem.
  • Enable data-informed operational improvements at both platform and restaurant level.

User Stories / Use Cases

Persona: Admin (super-user)

  • As an Admin, I want to view real-time (hourly) sentiment trends for today to identify operational issues instantly.
  • As an Admin, I want to filter sentiment data by Last 7 Days or This Month to compare performance across restaurants.
  • As an Admin, I want to see a Last 3 Months view to compare sentiment trends across quarters.

Persona: Restaurant Partner

  • As a Restaurant Partner, I want to track daily sentiment changes over the Last 30 Days to spot shifts in service perception.
  • As a Restaurant Partner, I want to monitor hourly sentiment for today to detect and act on customer dissatisfaction fast.

Feature Scope

Must‑have

  • Platform → Partner Portal and Admin Dashboard
  • Sentiment classification display (Positive / Neutral / Negative) → All Sentimens by default
  • Dynamic time range filter with presets: same with existing Private (https://app.clickup.com/9003122396/docs/8ca1fpw-30456/8ca1fpw-53296)
    • Today (hourly) → Full day, not depends on the open and close time
    • Last 7 Days (daily)
    • Last 30 Days (daily)
    • This Month (daily)
    • Last Month (daily)
    • Last 3 Months (monthly granularity) → Default
  • Empty reviews excluded from sentiment classification.
  • Trend line chart that adjusts its granularity automatically based on the selected time range.
  • Ability to switch between branches in Partner view.
  • Tooltip display on charts (e.g., counts, % sentiment per bar).

Should‑have

  • Export sentiment report data as Excel or PDF.

Design Considerations

  • Admin Dashboard UI:
    • Grid of restaurants with visual stacked sentiment bars and labels → Top 5 restaurant with the most review
      • Filter by Last 3 Months by default
    • Sort/filter by sentiment volume or trend change.
      • Filter by : Last 3 Months by default
      • Sentiment : All Sentiments by default
    • Sentiment Trend review → need to select a restaurant first
      • Select Restaurant by default
    • Sentiment table of all restaurants
    • Can export by Excel or PDF
  • Partner Portal UI:
    • Same with admin dashboard, but only display restaurant or branches
    • Add subpage inside Analytics page
    • Sort/filter by sentiment volume or trend change.
      • Filter by : Last 3 Months by default
      • Sentiment : All Sentiments by default
    • Individual Login
    • Group Login
      • Sentiment Trend review only can select specific branch (Restaurant A, Restaurant B and so on). can't all branches at the same time
      • View all sentiment reviews. can see All branches or specific branch
  • Accessibility:
    • ARIA support, keyboard navigation, and WCAG-compliant color contrasts.
    • Use iconography and patterns for sentiment (not color only).
  • Responsive Design:
    • Support mobile web view and responsive resizing for smaller viewports.
  • Performance:
    • Ensure load time < 2s for 3 months of data across hundreds of restaurants.