Loading the site

Skip to content
All work

Case study · No. 04 · Travel · booking · 2024

MECCA Booking: booking near the Haram without guessing the distance

A rental platform for Umrah and Hajj: verified homes within walking distance of the Haram, a three-step booking for pilgrims, and a space for the hosts who welcome them.

  • Web UX/UI
  • Back-office
  • Design system
MECCA Booking home: book your home in Mecca, the search over a photo of the city
Role
Product designer, from research to prototype
Duration
10 weeks
Team
Concept project, solo
Platform
Web, desktop first
Tools
Figma Miro Google Meet Google Docs
Deliverables
Benchmark, Proto-personas, Journey map, User flows, Wireframes, Design system, 28 screens, Prototype

The project in numbers

What was delivered, counted in the Figma file. The product is fictional: no usage results.

  • 28

    screens

    17 for the traveller, 11 for the host and the moderation team

  • 3

    roles served

    The pilgrim, the host, the team that checks listings

  • 23

    components

    On three variable collections: colours, spacing, radii

01 · The project

Staying near the Haram

Every year millions of pilgrims look for a place to stay in Mecca for Umrah or Hajj. In 2024 Saudi Arabia counted 35.7 million Umrah pilgrims, 16.8 million of them from abroad.

For them the criterion is neither style nor price alone: it is the walking distance to the Haram, an arrival possible at any hour after a night flight, and a host they can trust, often with a whole family.

MECCA Booking is a rental platform designed for that stay: search by distance to the Haram, book in three steps, and, for hosts, run listings, calendar and earnings.

Ten weeks, from research to back-office

  1. 01 · Weeks 1–3

    Discover

    The booking journeys of four offers, from generalists to agency packages, and the context of the pilgrimage.

    • Benchmark
    • Desk research
  2. 02 · Weeks 3–4

    Define

    Two proto-personas, traveller and host, and the experience map of a stay.

    • Proto-personas
    • Journey map
    • HMW
  3. 03 · Weeks 4–6

    Ideate

    Architecture of both spaces, booking flow, wireframes.

    • User flow
    • Impact × effort
    • Wireframes
  4. 04 · Weeks 6–9

    Prototype

    Design system, 17 traveller screens, 11 host and moderation screens, a linked prototype.

    • Design system
    • High-fidelity UI
    • Figma prototype
  5. 05 · Weeks 9–10

    Check

    A review against UX laws, empty and error states, a test plan.

    • Heuristic review
    • Test plan

02 · Discovery

A market of pilgrims, not tourists

  • 35.7 M

    Umrah pilgrims in 2024

    A third more than in 2023.

    GASTAT, via Argaam

  • 16.8 M

    from abroad

    47 % of the total: those who need a place to stay.

    GASTAT, via Argaam

  • 13.7 M

    Umrahs performed in Mecca

    73 % of the total, the rest in Medina and elsewhere.

    GASTAT, via Argaam

  • ≈ 1 km

    on foot, at most, with elders

    The distance to the Haram a family accepts, five prayers a day.

    Hypothesis

  • 3 to 6

    travellers per booking

    Parents, spouses, children: people book for a family.

    Hypothesis

  • 2 am – 4 am

    common arrival times

    Night flights land when nobody opens the door.

    Hypothesis

Attendance figures read in October 2026. The others are hypotheses to check in interviews.

Four ways to find a bed

The generalists, home sharing and the indirect alternative: the agency package.

Four ways to find a bed
CriterionMECCA · the projectThe projectBooking.comGeneralistAirbnbHome sharingAgency packageIndirect
Walking distance to the HaramYesOn every listing, and as a filterPartlyDistance to a landmarkPartlyA map, not a distanceNoHotel imposed
Self check-in at any hourYesCode sent the day beforePartlyDepends on the placeYesPartlyGroup welcome
Clear cancellation before payingYesDate and amountYesYesNot documented
Content made for UmrahYesDistricts, gates, suhoorNoNoYesGuidance
Free choice of dates and homeYesYesYesNoThe group's dates

The table scrolls sideways.

A review of public journeys, without an account. Verdicts describe what was visible, not a full audit.

Takeaway

Generalists know booking, not pilgrimage; agencies know pilgrimage but impose everything. The open space: the freedom of a rental, with an agency's knowledge.

03 · Definition

Two sides of the same door

Proto-personas: hypotheses to validate in interviews; their quotes are illustrative.

  • P1 · the pilgrim

    Youssef, 46

    Manager in Casablanca, going on Umrah in April with his mother and wife

    Goal
    An apartment under ten minutes' walk from the Haram, booked before the flights
    What gets in the way
    Listings “near the Haram” without a distance, arriving at night after the flight

    « My mother walks slowly: I want minutes, not “close”. »

  • P2 · the host

    Youness, 38

    Has rented three apartments in Ajyad to pilgrims for years

    Goal
    Fill Ramadan and Hajj without spending his nights on the phone
    What gets in the way
    Requests on WhatsApp, cash at key handover, late cancellations

    « I want to see this week's arrivals without opening ten chats. »

Youssef's stay, before

Without a dedicated platform, between a generalist site and messages. Emotions assumed.

Youssef

01Search

02Compare

03Book

04Arrive

05Stay

06Return

Does

Compares listings “near the Haram”
Looks at photos and reviews
Enters dates, guests, pays
Lands in Jeddah, arrives at 3 am
Walks to the Haram and back
Leaves a review, or not

Thinks

Near is how far?
Looks clean
What if the visa is late?
Who will let us in?
We were right to stay close
It would help other families

Feels

Opportunity

Walking distance on every card
Reviews that rate the location
Dated cancellation before paying
Self check-in, code the day before
A review in two minutes

The map scrolls sideways.

The problem to solve

Pilgrims need to know the walking distance to the Haram and the hour they will be able to get in, because they travel as a family and pray five times a day. Today, listings say “close” without proving it.

The questions the design answers

  1. Q1

    How might we make walking distance the first search criterion?

    Why this one · It is the one criterion you cannot fix once there.

  2. Q2

    How might we make arriving possible at any hour?

    Why this one · The peak of anxiety on the experience map.

  3. Q3

    How might we make a host trustworthy before payment?

    Why this one · You hand your family to a stranger.

  4. Q4

    How might we help hosts fill their dates without chats?

    Why this one · Youness does everything by hand.

04 · Ideation

What the first version keeps

Impact ↑

Quick wins · Impact high · Effort low

  • Walking distance to the HaramIn V1

    Q1

  • Dated cancellation, total priceIn V1

  • Self check-in and codeIn V1

    Q2

Big bets · Impact high · Effort high

  • Listings checked by the teamIn V1

    Q3, moderation console

  • Host space: calendar and pricesIn V1

    Q4

  • Host–guest messagingIn V1

Nice to have · Impact low · Effort low

  • Prayer times on the listingLater

Later · Impact low · Effort high

  • Disputes and refundsIn V1

    Needed to take payments online

  • Mobile appLater

    The stay is planned on a computer

Effort →

Impact and effort estimated: a designer's ranking, to review with a team.

Booking, in three steps

  1. Start

    Home

    Destination, dates, guests

  2. Screen

    Results

    List and map, distance to the Haram

  3. Screen

    Listing

    Photos, amenities, reviews, total price

  4. Decision

    Signed in?

    No: sign in or create an account, then back to the listing

  5. Screen

    Your stay

    Guests, arrival time

  6. Screen

    Payment

    Card, PayPal or Apple Pay

  7. System

    3-D Secure check

  8. End

    Confirmation

    Arrival code sent the day before

Low-fidelity wireframes

Four pages before any colour: the search, the results, the listing and the host dashboard.

  1. 01 · Home

  2. 02 · Results

  3. 03 · Listing

  4. 04 · Host dashboard

05 · Design

A warm, quiet design system

Geist, a yellow accent for action, warm neutrals and rounded corners: every colour, spacing and radius on screen is a variable.

Colours

  • Accent

    #FBB040

    Main action

  • Accent hover

    #E89A1F

    Hover state

  • Accent subtle

    #FFF6E5

    Banners, requests

  • Accent text

    #9A5F06

    Readable yellow text (AA)

  • Page

    #FFFFFF

    Background

  • Text

    #0A0A0A

    Headings and text

  • Secondary text

    #525252

    Captions

  • Success

    #15803D

    Confirmed, paid

  • Success subtle

    #ECFDF3

    Status fill

  • Danger

    #DC2626

    Cancellation, dispute

Type

  • Display

    Geist Bold 44

    Book near the Haram

  • Heading 1

    Geist Bold 30

    Popular districts

  • Heading 3

    Geist Bold 20

    What this place offers

  • Body

    Geist Regular 16

    Two air-conditioned bedrooms and a full kitchen.

  • Price

    Geist Bold 20

    €858

23 components

  • Logo
  • Button
  • Badge
  • Input
  • Header
  • Footer
  • Search
  • Stepper
  • Amenity
  • Rating
  • Listing card
  • District card
  • Host card
  • Review
  • Price summary
  • Option
  • Booking row
  • Avatar
  • Price pin
  • Message bubble
  • Stat
  • Host menu
  • Host bar
MECCA Booking components board: buttons and their states, fields, controls, badges, cards and the icon set
The components as they are in Figma: every variant, then the icons.
The MECCA Booking design system board: colours, type, spacing, radii and shadows
The design system board, in Figma.

Home: four fields, three guarantees

First iteration

First iteration of home: six search fields and district cards

Final version

Final home: four fields, the promise and three guarantees

What changes

  • Search goes from six fields to four: destination, dates, guests, then the call to action.

    UX law · Hick’s LawKept

  • Three guarantees under the search: verified hosts, distance to the Haram, free cancellation.

    UX law · Serial Position EffectKept

  • One yellow button on screen: the one that runs the search.

    UX law · Von Restorff EffectKept

The listing: decide without scrolling

First iteration

First iteration of the listing page

Final version

Final listing: title, rating, distance to the Haram and photo gallery

What changes

  • Title, rating, reviews and distance to the Haram fit on one line above the photos.

    UX law · Law of ProximityKept

  • The whole stay's price and the cancellation stay on the right while reading.

    UX law · Anchoring EffectKept

The traveller's journey

  • Results with a map and distance to the Haram

    Results: distance first

  • Your stay: guests and arrival time

    Your stay

  • Payment with summary and cancellation terms

    Payment

  • Booking confirmation

    Confirmation

  • Booking details

    Booking details

  • Messages with the host

    Messages

Host and moderation side

The back-office answers Q4: see the week, set prices, get paid, without chats.

  • Host dashboard: earnings, occupancy, arrivals

    Dashboard

  • Calendar and nightly prices

    Calendar and prices

  • Earnings and payouts

    Earnings

  • Add a home

    Add a home

  • Moderation console: disputes and refunds

    Disputes, team side

06 · Validation

UX laws at work

  • Broken

    Hick’s Law

    More choices, slower decisions: cut and group the options.

    Six fields before the first answer.

    Home, first iteration

  • Kept

    Hick’s Law

    More choices, slower decisions: cut and group the options.

    Four fields, sensible defaults.

    Home

  • Kept

    Von Restorff Effect

    The one thing that differs is the one thing remembered.

    Yellow is kept for the main action only.

    Everywhere

  • Kept

    Law of Proximity

    What sits together reads together.

    Rating and distance read at a glance.

    Listing

  • Kept

    Anchoring Effect

    The first number sets the scale for every next one.

    The whole stay's price, fees included, before paying.

    Summary

  • Kept

    Peak-End Rule

    People judge the peak and the end, not the average.

    The most dreaded moment becomes a code sent the day before.

    Arrival

Expected results

What a test and a launch should show. Nothing has been measured.

Expected resultIndicatorIndicative target
A safer choiceTesters who mention the distance to the Haram before the price4 out of 5 in testing
A booking without detoursBookings completed, from listing to paymentAbove 60 % (hypothesis)
Arrivals without a callNight arrivals with no contact with the host9 out of 10 (hypothesis)

The test that would validate the journey

ParticipantsFive people who went on Umrah with family in the last three years, in 40-minute remote sessions on the prototype, screen shared over Google Meet.

  1. 01

    TaskFind a place for three adults within a kilometre of the Haram, 20 to 26 April.

    Success criterionFound in under 3 minutes, distance filter used.

  2. 02

    TaskCheck until when cancellation is free.

    Success criterionDate given without leaving the listing.

  3. 03

    TaskBook and state a 3 am arrival.

    Success criterionPayment made, arrival mode understood.

  4. 04

    TaskAs a host: accept a request and block two nights.

    Success criterionDone from the dashboard, without help.

Interview questions

  • Tell me how you found your last place to stay in Mecca.
  • What worried you most before leaving?
  • At what distance from the Haram would you turn a home down, and why?

What I take from it

  1. 01

    One criterion can carry it all

    Walking distance to the Haram ended up organising the search, the cards, the listing and the reviews.

  2. 02

    Draw the other side

    The traveller journey only held with the host space: without it, self check-in stayed a promise.

  3. 03

    An iteration is a study

    Going over the first version screen by screen showed what fresh eyes would catch: too many fields, no guarantees.

Got a project in mind?

Describe the screen or the workflow that gets in the way: I reply with a first reading of the problem.