End-to-End Case Study · 2018

HOOQ Progressive Web App

A progressive web app to augment the top-of-funnel experiences.

Three phones showing the HOOQ progressive web app at hooq.tv: the Discover page with a featured rental, the Channels page with live TV channels, and the Movies page with action and drama titles.

My role

Product Strategy & Discovery

  • Facilitated Design Sprints

    Led collaborative design sprints to explore opportunities, align stakeholders, and shape solutions.

  • Shaped Product Direction

    Mapped customer journeys to identify pain points and opportunities that informed product strategy.

Design Systems & Product Experience

  • Established the Design Language System

    Partnered with engineering to build the mobile web design system from the ground up, establishing reusable components and interaction patterns.

  • Owned End-to-End Product Design

    Designed end-to-end product experiences, translating customer needs and requirements into intuitive interfaces.

Prototyping & Validation

  • Brought Ideas to Life

    Created high-fidelity prototypes to explore concepts, test interactions, and gather customer feedback.

  • Led UX Research Collaborations

    Partnered with Customer Insights and external research vendors to conduct research and validate design decisions.

About

HOOQ was a video-on-demand streaming platform operating primarily across Southeast Asia and India. In 2018, one of its key differentiators was its diverse content library, offering a mix of popular Hollywood titles and locally produced content tailored to regional audiences.

Backed by one of Southeast Asia's largest telecommunications groups, HOOQ leveraged its telco partnerships as a key customer acquisition channel. One of its primary strategies was to bundle HOOQ subscriptions with prepaid mobile data purchases, giving customers an added incentive to try the platform while expanding its reach through the telco's existing subscriber base.

Problem

People problem

“I want an easy and convenient way to discover and watch great video content, from popular Hollywood titles to stories produced locally.”

Business problem

Fewer than 1% of pre-provisioned customers were converting into active subscribers. Our activation journey relied primarily on a mobile app, supported by a static desktop website, limiting the ways customers could discover, access, and activate their subscriptions.

Conventional funnel compared with HOOQ's top-of-funnel problem. On the left, a wide Acquisition stage narrows into Activation and flows into Retention, Referral, and Revenue. On the right, Acquisition narrows to a single point, leaving only a thin line through Activation, Retention, Referral, and Revenue.

Why aren’t more customers activating their free trial?

How might we measure ground truth of actual page landing to conversions?

Customer profile

  • Emerging market customers

    The majority of our customer base was from Indonesia. They consumed a lot of video content on their mobile devices.

  • Mobile data usage sensitive

    These users were very cautious with their mobile data. A key value proposition for them was data bundles with a separate video data limit.

  • Best bang for buck

    Subscriptions were bundled with prepaid data packs, so HOOQ was seen as a value-added service when purchasing data bundles.

  • Cost-effective devices

    Our current data showed most of our customers were on more affordable Xiaomi phones on Android.

Design sprint

I facilitated a cross-functional Design Sprint to build a shared understanding of our business ecosystem and identify key opportunities. The workshop brought together stakeholders from Engineering, Marketing, Product Management, Data Science, and executive leadership.

Through Lightning Talks, problem mapping, How Might We questions, solution sketching, and dot voting, we aligned on key challenges and explored potential solutions.

Activation journey across Telco, Customer, and HOOQ. Customers discover HOOQ through an SMS or an ad, then must go to the App Store or Play Store, download the app, sign up, and activate before they can browse, read info, play videos, or play trailers. The Download HOOQ app step is highlighted in red as the bottleneck.
App size problemThe conversations we had in the Design Sprint made us realize our app size was a big issue. Customers wouldn’t activate because they wouldn’t want to spend a large portion of the mobile data bundle that they’ve just purchased on downloading a large app.View full size

Hypothesis

We believed that introducing a web-based experience would reduce friction in the activation journey and improve conversion across both telco partnerships and organic acquisition channels.

To validate this, we defined two key success metrics:

  • Pre-provisioned Activations

    Track the end-to-end conversion from mobile web landing to sign-up and subscription activation.

  • Streaming Minutes

    Measure engagement by tracking the total time users spend streaming content on the platform.

Design foundations

I worked closely with software engineers to shape HOOQ's mobile web Design Language System. While we already had an established desktop website, its design foundations weren't built to scale to mobile. This meant creating many of the core elements from the ground up, including foundational styles, reusable components, and interaction patterns.

HOOQ color palette board with named swatches grouped into primary, secondary, detailing, dividers, and indicators, each with its hex value.
Color palette: primary, secondary, detailing, divider, and indicator colors.
HOOQ typography styles board listing title, body, small, and tiny styles with their font size, weight, line height, and color, alongside single, double, and triple variants.
Typography styles: six default styles optimized for mobile, with their variants.
HOOQ icon library board showing the 24 by 24 pixel icon grid with a 2 pixel stroke, and a grid of line-style icons.
Icon library: line-style icons drawn on a 24px grid with a 2px stroke.
1 / 3

One important consideration was supporting both light and dark themes. I developed corresponding colour swatch pairs to ensure visual consistency, accessibility, and a cohesive experience across both themes, regardless of user preference.

The HOOQ web app home screen in light and dark themes side by side, with arrows mapping each light-theme color to its dark-theme pair: white to thunder, eggplant to gold, coal text to white text, and white to bastille.

UI patterns

Spotlights

Spotlights are the full width banners that feature the latest/popular content.

Option 1, a massive spotlight: an animated full-screen banner for the featured movie Bad Genius, with a Watch now button, pushing the category filters and Popular row to the bottom of the screen.
Option 1: Massive
Option 2, a conservative spotlight: a shorter banner for Avengers: Endgame, leaving room for category filters and a row of popular titles below it.
Option 2: Conservative

Customers felt the massive spotlights occupied too much space. We had no infrastructure to implement personalisation yet and spotlights were curated manually. If a massive spotlight doesn’t interest customers, it would make the rest of the catalog more inaccessible.

We implemented a conservative spotlight for MVP instead so we could highlight the breadth of the catalog better at the initial view.

Collections

Used to curate a group of titles into one genre or theme.

Option 1, an expandable collection: an Action grid of six titles with a Show more button that loads more titles into the grid.
Option 1: Expandable
Option 2, rail collections: rows of titles for New, Action, and Popular, each scrolling horizontally.
Option 2: Rail

We decided to implement expandable rails since we got more conclusive signals from tracking explicit “Show more” clicks than horizontal scroll events.

Title Details

We explored quick view window components that allow customers to see details without having to leave the main “Discover” feed.

Option 1, a quick view: tapping a title on the Discover feed opens its details, including a Watch now button, runtime, languages, synopsis, director, and actors, in a window over the blurred feed.
Option 1: Quick view
Option 2, direct to details: a full title details page for Spider-Man: Homecoming with a Watch now button, runtime, languages, synopsis, director, and actors.
Option 2: Direct to details

We decided not to pursue the quick view, as research did not provide conclusive evidence of its value relative to the implementation effort. Any potential benefits were expected to be marginal and did not justify the added complexity.

States

We also designed microinteractions to provide clear feedback and set user expectations during data-loading moments throughout the web app.

Animated loading states for a primary and a secondary button.
Button loading
Animated page loading state for the Discover page, with placeholder shapes standing in for the spotlight banner and title posters while content loads.
Page loading, created with Illustrator, After Effects, and Lottie

Screen flows

We focused on three core principles to make the PWA experience accessible, intuitive, and flexible. Users could explore the content catalog without signing in, encounter sign-up prompts at relevant moments without interrupting their browsing, and activate their subscriptions through a streamlined flow regardless of whether they came through a telco partnership or organic acquisition.

Open Catalog

Let users explore the content catalog without requiring an account or sign-in upfront.

Screen flow for browsing without signing in: the Discover, Channels, Movies, and TV Shows tabs, then a title details page for The Flash, leading straight into video playback.

Contextual Calls to Action

Introduce non-intrusive sign-up prompts at relevant moments, allowing users to engage with content before committing.

Screen flow for contextual sign-up prompts: while browsing, a bar at the bottom of the Discover page offers the first month free with a Sign up now button; after watching an episode, a screen asks the user to sign up to continue watching the next episode, with options to log in or skip.

Scalable Activation

Support a consistent activation experience for both pre-provisioned telco subscribers and organic users signing up directly.

Screen flow for activation: from the sign-up prompt, users sign up with a mobile number and verify it with a one-time code. A decision then checks whether the customer is pre-provisioned: if yes, they go to pre-provisioned activation; if no, organic activation continues to choosing a plan and a payment method.

Outcomes

What went well?

  • Pre-provisioned activations on web surpassed iOS by ~7%
  • Organic activations were now happening on mobile web
  • Users were actually watching videos on PWA

What didn’t go so well?

  • There were some missing partnership integrations that prevented us from redirecting traffic to PWA for activation
  • Identifying which content was free to watch as an anonymous user was challenging due to licensing

Retrospective

What I’d change if I could do it again?

  • Enable experimentation

    We could have included better frameworks for A/B testing. That would have been the next step after launching 0 to 1.

  • Cross-platform influence of Design System

    Building the Design System could have been scaled to native mobile platforms and TV applications. The Design team refactored components visually but they were still fragmented across platforms in code.