Booking.com — UX Redesign
Booking.com is one of the most used travel apps in the world. So why do so many users still abandon mid-search? I audited the existing experience, identified where it was losing people, and redesigned the key flows that were letting them down.
Role: UX Designer — self-initiated project
Type: UX Audit + Redesign · Mobile App
What I did: Heuristic evaluation of the live app → identified pain points across 4 key flows → redesigned screens to address each one
Problem Statement
Booking.com isn’t broken — it’s overwhelmed. Years of feature additions have created an interface where finding and booking a hotel requires more mental effort than it should. Users face cluttered homepages, inconsistent information hierarchy, missing trust signals and a checkout flow that buries the details people actually need. The result is decision fatigue — and users quietly leaving for a competitor.
Goals
Three things I set out to fix:
Reduce cognitive load — cut the noise so users can focus on what actually matters when choosing a hotel
Surface trust signals earlier — reviews, cancellation policy and pricing should never be a surprise at checkout
Improve the search-to-book flow — fewer steps, cleaner hierarchy, faster decisions
What I found — a screen-by-screen audit
I walked through the live app as a real user would — searching, comparing, and attempting to book — and documented every point of friction I hit.
1.Hompage
Homepage lacks a clear entry point
The first screen gives users no direction — no featured destinations, no sense of where to start. For a first-time user especially, it feels flat and uninviting.
Nothing to explore if you’re not ready to search
Users who are in discovery mode — browsing rather than actively booking — have no reason to stay. There’s no inspiration layer, no personalised suggestions, nothing to build intent.
Property cards hide the information users need most
Price, rating and cancellation policy are the three things users look for first. The current cards bury two of those — forcing users to tap through to a detail page just to compare basics.
Redundant cards creating navigation confusion
The Genius card and destination cards often point to the same page — users don’t realise this and tap both, creating a disorienting back-and-forth that breaks the flow.
Offer cards with no context
Cards show discounts without naming what the offer actually is. A 10% badge means nothing if users can’t tell whether it applies to their dates, their loyalty tier, or specific properties only.
Preference warnings in the wrong place
Warning icons on preference settings create unnecessary anxiety. These settings belong quietly inside the profile — not surfaced mid-session as if something is broken.
2.Search page
Search field offers no starting point
An empty search field with no suggestions puts the entire cognitive load on the user. Adding recent searches, suggested destinations, and trending locations would reduce friction and help users in discovery mode find a starting point faster.
Flight booking option disrupts the stays search flow
Surfacing a “Find flights” option inside the hotel search field creates a context switch at exactly the wrong moment — when users are already mid-task. It introduces doubt about whether they’re in the right flow. This option belongs in the navigation, not the search input.
3.Search Listing observations
Inconsistent information hierarchy across listing cards
Hotel name, rating, distance, price and cancellation policy are arranged differently across cards — there’s no fixed visual pattern. Users can’t scan the list efficiently because their eye has to start fresh on every card.
Single photo per listing card — not enough to build confidence
Photos are one of the top factors users use to shortlist hotels. Showing only one image per listing card forces unnecessary taps into the detail page, slowing down comparison and adding friction to the decision-making process.
4.Hotel Summary Page observations
The same information appears in multiple places on the page
Repeating content — especially around pricing and policies — doesn’t reassure users, it confuses them. They start wondering if the two instances mean different things, which creates doubt instead of confidence.
Cancellation policy buried — the single biggest trust issue
This is the detail users most want to see before committing to a booking. Hiding it in small print or making it a separate tap creates anxiety at exactly the moment you need users to feel confident. It should be visible on the listing card itself, not just the summary page.
FAQ section feels like an afterthought
The layout makes it hard to scan and even harder to find a specific answer. Users asking questions at this stage are one answer away from booking — or one frustrating experience away from leaving. The FAQ needed to feel fast, confident, and reassuring — not buried and generic.
Redesigned screens — what changed
Each screen directly addresses a pain point from the audit above. Here’s what I changed.
The homepage redesign shifts from a search-first layout to a discover-then-search experience — giving users a reason to stay even before they know where they want to go.
The search flow now reduces cognitive load at the starting point — the moment users are most likely to abandon if they feel overwhelmed.
The listing page redesign focuses on scanability — users can now compare hotels down the list without tapping into each one just to find the price.
The hotel summary page is now built around the questions users are actually asking — price, cancellation, and reviews — before they ask them.
Final Reflection
Redesigning a product you don’t own — without access to real user data, analytics, or session recordings — is a useful constraint. It forces you to rely entirely on observation and first principles. Every decision I made here came from asking one question: what does a user actually need at this exact moment?
The biggest lesson was about trust. Booking.com doesn’t have a visual design problem — it has a trust problem that shows up in the visual design. Cancellation policies hidden in small print, offer cards with no context, search flows that introduce doubt mid-task — these aren’t aesthetic issues. They’re moments where the interface stops working for the user and starts working against them.
If I were taking this further, I’d validate the redesigned flows with real users through moderated usability testing, instrument the prototype with Maze to capture task completion rates and run an A/B test on the listing card layout specifically — since that’s where most of the comparison friction lives.