Bala logobala
← Back to work
Consumer / Mobile

Smart Remote App for TV Set-top Box

Designing a second-screen mobile experience to revitalise Set-top box engagement against streaming competition in India

Smart Remote App for TV Set-top Box
RoleUX Designer
TimelineUser-Centered Design — 5 phases
TeamBala M, Maryada Palaskar, Teja
ToolsFigma, Flinto, User Testing

Overview

India's satellite television market was losing ground to streaming services. Poor advertisement strategy and limited user privileges had diminished the Set-top box experience — despite it remaining a cost-effective entertainment option for millions. The challenge: design a mobile companion app that made the Set-top box experience modern, interactive, and compelling again.

The Market Context

Streaming services were eroding Set-top box viewership by offering on-demand content, personalisation, and second-screen interactivity. The traditional remote control offered none of this. Yet Set-top boxes remained the primary entertainment source for a large segment of Indian households — cost-effective and already installed. The opportunity was to meet users where they already were (on their phones) and extend the TV experience.

User Research & Discovery

We used a 5-step User-Centered Design methodology. The discovery phase centred on user shadowing at natural locations — observing how people actually used their TV setups at home, what friction they encountered, and what they wished they could do. The target audience was "Highly Desperate Users" (HDU) aged 24–35: people who wanted more from their television experience but were constrained by the hardware they already owned.

Key issues identified from user shadowing
User shadowing findings — key friction points observed while watching people use their Set-top boxes at home
User persona for target audience aged 24-35
Target user persona — the "Highly Desperate User" aged 24–35, already on their phone while watching TV

Ideation & Feature Design

The app was designed around features that genuinely extended the TV experience rather than just replicating the physical remote on a touchscreen. Device pairing via Wi-Fi, program guides with reminders, live polling and comments, ratings and reviews, and multi-TV support were all prioritised based on what users most wanted from a companion experience.

Mobile to Set-Top box Wi-Fi connectivity diagram
Device connection model — Wi-Fi pairing that works without internet dependency for basic TV control
Application navigation user flow
User flow — mapping the key journeys through the app, from pairing to program browsing to live interaction

Design & Prototyping

Low-fidelity wireframes mapped core flows before moving to high-fidelity interactive mockups built in Flinto — allowing realistic gesture-based testing that static mockups couldn't provide.

Low fidelity wireframes for initial testing
Low-fidelity wireframes — testing the information architecture and core flows before investing in visual design
High fidelity mockup — program guide and home screen
High-fidelity mockup — Program Guide, live content controls, and personalised recommendations
High fidelity mockup — interactive features and TV controls
High-fidelity mockup — Live Polling, Comments, and TV Controls designed for phone ergonomics

Usability Testing

Participants evaluated 10 core tasks: device connection, program navigation, episode playback, setting reminders, managing subscriptions, account management, and volume control. Users found core functions largely self-evident — a strong signal that the interaction model matched existing mental models.

User testing session
Usability testing — validating all 10 core task flows with real users before final iteration

Outcomes & Impact

10
Core tasks evaluated

Device connection, program navigation, playback, reminders, subscriptions, and TV controls — all tested with real users.

5
UCD phases completed

Discovery, ideation, design, prototyping, and testing — a complete User-Centered Design process from research to validated prototype.

High
Task discoverability

Majority of users found core functions self-evident — a strong signal that the interaction model matched existing mental models.