Works

About Me

Playground

Resume

Contact

Works

About Me

Playground

Resume

Contact

Designing an AI Travel App That Boosted Usability by 65%

Designing an AI Travel App That Boosted Usability by 65%

Designing an AI Travel App That Boosted Usability by 65%

OVERVIEW

OVERVIEW

Ever planned a trip and felt completely overwhelmed?


That was the spark behind this project, a Travel app that simplifies itinerary planning by reducing research overload.

I watched friends juggle tabs, miss out on hidden gems, and feel more stressed than excited. Travel should be about experiences, not logistical headaches.

Ever planned a trip and felt completely overwhelmed?


That was the spark behind this project, a Travel app that simplifies itinerary planning by reducing research overload.

I watched friends juggle tabs, miss out on hidden gems, and feel more stressed than excited. Travel should be about experiences, not logistical headaches.

Ever planned a trip and felt completely overwhelmed?


That was the spark behind this project, a Travel app that simplifies itinerary planning by reducing research overload.

I watched friends juggle tabs, miss out on hidden gems, and feel more stressed than excited. Travel should be about experiences, not logistical headaches.

ROLE

Solo UX Designer


End-to-end, Research to high-fidelity prototype


Passion Project

TOOLS

Figma


Miro


Qualtrics for research


Maze for user testing

TIMELINE

6 Weeks

TARGET USERS

Professional, Frequent, and Occasional Travelers

IMPACTS

Reduced planning time through personalized suggestions.


Increased user confidence by integrating maps and real-time updates.


Usability score increased from 40% to 66%.

UX METHODS

Usability Scores from Maze


Priority Matrix


Affinity Mapping

ROLE

Solo UX Designer


End-to-end, Research to high-fidelity prototype


Passion Project

TIMELINE

6 Weeks

TARGET USERS

Professional, Frequent, and Occasional Travelers

UX METHODS

Usability Scores from Maze


Priority Matrix


Affinity Mapping

TOOLS

Figma


Miro


Qualtrics for research


Maze for user testing

IMPACTS

Reduced planning time through personalized suggestions.


Increased user confidence by integrating maps and real-time updates.


Usability score increased from 40% to 66%.

ROLE

Solo UX Designer


End-to-end, Research to high-fidelity prototype


Passion Project

TIMELINE

6 Weeks

TARGET USERS

Professional, Frequent, and Occasional Travelers

UX METHODS

Usability Scores from Maze


Priority Matrix


Affinity Mapping

TOOLS

Figma


Miro


Qualtrics for research


Maze for user testing

IMPACTS

Reduced planning time through personalized suggestions.


Increased user confidence by integrating maps and real-time updates.


Usability score increased from 40% to 66%.

The Problem

The Problem

The Problem

Planning a trip requires juggling dozens of tabs, comparing countless options, and second-guessing every decision. Most travel apps overwhelm users with information they don’t help you think.


After surveying 16 travelers and evaluating 11 competing apps, three core friction points emerged:

1. Discovery gap

Users couldn’t find activities that matched both their interests and their time constraints together. These two filters rarely existed in the same place.


2. Information overload

Competing apps crammed everything onto one screen. Rather than helping users decide, they created decision paralysis.


3. Trust gap

Without clear distance and timing context, users couldn’t tell if their itinerary was even realistic leading to anxiety and abandoned plans.


The real problem wasn’t a lack of travel content. It was the absence of a thinking layer between the user and that content.

Planning a trip requires juggling dozens of tabs, comparing countless options, and second-guessing every decision. Most travel apps overwhelm users with information they don’t help you think.


After surveying 16 travelers and evaluating 11 competing apps, three core friction points emerged:

1. Discovery gap

Users couldn’t find activities that matched both their interests and their time constraints together. These two filters rarely existed in the same place.


2. Information overload

Competing apps crammed everything onto one screen. Rather than helping users decide, they created decision paralysis.


3. Trust gap

Without clear distance and timing context, users couldn’t tell if their itinerary was even realistic leading to anxiety and abandoned plans.


The real problem wasn’t a lack of travel content. It was the absence of a thinking layer between the user and that content.

Planning a trip requires juggling dozens of tabs, comparing countless options, and second-guessing every decision. Most travel apps overwhelm users with information they don’t help you think.


After surveying 16 travelers and evaluating 11 competing apps, three core friction points emerged:

1. Discovery gap

Users couldn’t find activities that matched both their interests and their time constraints together. These two filters rarely existed in the same place.


2. Information overload

Competing apps crammed everything onto one screen. Rather than helping users decide, they created decision paralysis.


3. Trust gap

Without clear distance and timing context, users couldn’t tell if their itinerary was even realistic leading to anxiety and abandoned plans.


The real problem wasn’t a lack of travel content. It was the absence of a thinking layer between the user and that content.

The Goal

The Goal

The Goal

Design a solution that simplifies trip planning, enabling travelers to create enjoyable and efficient itineraries with ease, by leveraging AI.


Help users discover activities of interest, organize their plans, and balance preferences, while offering destination insights and local recommendations.

Design a solution that simplifies trip planning, enabling travelers to create enjoyable and efficient itineraries with ease, by leveraging AI.


Help users discover activities of interest, organize their plans, and balance preferences, while offering destination insights and local recommendations.

Design a solution that simplifies trip planning, enabling travelers to create enjoyable and efficient itineraries with ease, by leveraging AI.


Help users discover activities of interest, organize their plans, and balance preferences, while offering destination insights and local recommendations.

Research

Research

Research

These findings directly shaped the MVP feature set: a five-input form, AI-generated itinerary, and a clean split between list view and map view.

These findings directly shaped the MVP feature set: a five-input form, AI-generated itinerary, and a clean split between list view and map view.

These findings directly shaped the MVP feature set: a five-input form, AI-generated itinerary, and a clean split between list view and map view.

Survey — 16 participants via Qualtrics

Survey — 16 participants via Qualtrics

  • Users struggled most with finding activities that fit both their interests and time window simultaneously

  • Most found the pre-trip planning process stressful, not exciting

  • Strong demand for personalized, map-integrated suggestions that reduced manual decision-making

  • Users struggled most with finding activities that fit both their interests and time window simultaneously

  • Most found the pre-trip planning process stressful, not exciting

  • Strong demand for personalized, map-integrated suggestions that reduced manual decision-making

Competitor Analysis — 11 apps, heuristic evaluation

Competitor Analysis — 11 apps, heuristic evaluation

  • Most apps prioritized breadth over clarity overwhelming lists with no hierarchy

  • Day-to-day plans were poorly visible; users had to dig to understand their schedule

  • None offered AI-personalized routing based on user preferences combined with time constraints

  • Most apps prioritized breadth over clarity overwhelming lists with no hierarchy

  • Day-to-day plans were poorly visible; users had to dig to understand their schedule

  • None offered AI-personalized routing based on user preferences combined with time constraints

  • Most apps prioritized breadth over clarity overwhelming lists with no hierarchy

  • Day-to-day plans were poorly visible; users had to dig to understand their schedule

  • None offered AI-personalized routing based on user preferences combined with time constraints

Priority matrix

Priority matrix

Priority matrix

Affinity Mapping

Affinity Mapping

Affinity Mapping

Key Design Decisions

Key Design Decisions

Key Design Decisions

These are the three decisions that had the most measurable impact on usability.

These are the three decisions that had the most measurable impact on usability.

These are the three decisions that had the most measurable impact on usability.

Decision 1 — Inline instructions over a walkthrough guide.

Decision 1 — Inline instructions over a walkthrough guide.

Decision 1 — Inline instructions over a walkthrough guide.

THE PROBLEM

THE PROBLEM

First-round testing revealed 67% of users didn’t know where to click to start. My initial fix was a walkthrough guide but after reflection, I realized it added steps before users even began. A guide is friction disguised as help.

First-round testing revealed 67% of users didn’t know where to click to start. My initial fix was a walkthrough guide but after reflection, I realized it added steps before users even began. A guide is friction disguised as help.

First-round testing revealed 67% of users didn’t know where to click to start. My initial fix was a walkthrough guide but after reflection, I realized it added steps before users even began. A guide is friction disguised as help.

THE DECISION

THE DECISION

I placed inline instructional text directly where users would naturally look and interact. No extra screen. No extra step. Just the right nudge at the right moment.

I placed inline instructional text directly where users would naturally look and interact. No extra screen. No extra step. Just the right nudge at the right moment.

I placed inline instructional text directly where users would naturally look and interact. No extra screen. No extra step. Just the right nudge at the right moment.

THE RESULT

THE RESULT

Task completion improved from 67% to 80% in the second round of testing.

Task completion improved from 67% to 80% in the second round of testing.

Task completion improved from 67% to 80% in the second round of testing.

Decision 2 — Split itinerary view from map view

Decision 2 — Split itinerary view from map view

Decision 2 — Split itinerary view from map view

THE PROBLEM

THE PROBLEM

Users consistently felt overwhelmed when both views were combined on one screen. The competing information made it impossible to focus on either the route or the activities.

Users consistently felt overwhelmed when both views were combined on one screen. The competing information made it impossible to focus on either the route or the activities.

Users consistently felt overwhelmed when both views were combined on one screen. The competing information made it impossible to focus on either the route or the activities.

THE DECISION

THE DECISION

I separated them into two distinct screens with a clear toggle to switch between them. Users could now focus on what mattered to them in the moment without the other view pulling their attention.

I separated them into two distinct screens with a clear toggle to switch between them. Users could now focus on what mattered to them in the moment without the other view pulling their attention.

I separated them into two distinct screens with a clear toggle to switch between them. Users could now focus on what mattered to them in the moment without the other view pulling their attention.

THE RESULT

THE RESULT

Drop-off and misclick rates decreased significantly. First-click accuracy improved, and users reported the interface felt noticeably less overwhelming.

Drop-off and misclick rates decreased significantly. First-click accuracy improved, and users reported the interface felt noticeably less overwhelming.

Drop-off and misclick rates decreased significantly. First-click accuracy improved, and users reported the interface felt noticeably less overwhelming.

Decision 3 — Restructure the sign-up visual hierarchy

Decision 3 — Restructure the sign-up visual hierarchy

Decision 3 — Restructure the sign-up visual hierarchy

THE PROBLEM

THE PROBLEM

Heatmap data showed users were consistently clicking ‘Register with Email’ — but the intended path was Google sign-up, which is lower friction and faster. The layout was creating a conflict between visual prominence and intended behavior.

Heatmap data showed users were consistently clicking ‘Register with Email’ — but the intended path was Google sign-up, which is lower friction and faster. The layout was creating a conflict between visual prominence and intended behavior.

Heatmap data showed users were consistently clicking ‘Register with Email’ — but the intended path was Google sign-up, which is lower friction and faster. The layout was creating a conflict between visual prominence and intended behavior.

THE DECISION

THE DECISION

I restructured the layout so the preferred sign-up method aligned with the user’s natural reading flow — visually dominant, logically placed, and paired with a subtle secondary option below.

I restructured the layout so the preferred sign-up method aligned with the user’s natural reading flow — visually dominant, logically placed, and paired with a subtle secondary option below.

I restructured the layout so the preferred sign-up method aligned with the user’s natural reading flow — visually dominant, logically placed, and paired with a subtle secondary option below.

THE RESULT

THE RESULT

Sign-up friction was eliminated. Drop-off at this step was removed entirely.

Sign-up friction was eliminated. Drop-off at this step was removed entirely.

Sign-up friction was eliminated. Drop-off at this step was removed entirely.

The Results

The Results

The Results

After two rounds of testing with 12 participants via Maze:

  • Task completion rate: 67% → 80%

  • Usability score: 40% → 66%

  • Misclick rate: decreased significantly

  • Time on core task: reduced

  • User satisfaction: notably higher — users specifically cited improved clarity and flow

The biggest shift wasn’t in the numbers alone — it was that users stopped feeling lost. They knew what to do next.

After two rounds of testing with 12 participants via Maze:

  • Task completion rate: 67% → 80%

  • Usability score: 40% → 66%

  • Misclick rate: decreased significantly

  • Time on core task: reduced

  • User satisfaction: notably higher — users specifically cited improved clarity and flow

The biggest shift wasn’t in the numbers alone — it was that users stopped feeling lost. They knew what to do next.

After two rounds of testing with 12 participants via Maze:

  • Task completion rate: 67% → 80%

  • Usability score: 40% → 66%

  • Misclick rate: decreased significantly

  • Time on core task: reduced

  • User satisfaction: notably higher — users specifically cited improved clarity and flow

The biggest shift wasn’t in the numbers alone — it was that users stopped feeling lost. They knew what to do next.

What’s Next?

What’s Next?

What’s Next?

The core flow is validated. The next iterations will introduce collaborative itinerary planning and budget management both features users requested but were intentionally cut from the MVP to keep the first-use experience clean and focused.

These will be validated through a third round of user testing before being built into the high-fidelity prototype.

The core flow is validated. The next iterations will introduce collaborative itinerary planning and budget management both features users requested but were intentionally cut from the MVP to keep the first-use experience clean and focused.

These will be validated through a third round of user testing before being built into the high-fidelity prototype.

The core flow is validated. The next iterations will introduce collaborative itinerary planning and budget management both features users requested but were intentionally cut from the MVP to keep the first-use experience clean and focused.

These will be validated through a third round of user testing before being built into the high-fidelity prototype.

Get in touch.

silumanandhar@gmail.com

Copyright © 2024. Silu Manandhar

Get in touch.

silumanandhar@gmail.com

Copyright © 2024. Silu Manandhar