Multi-Room Recommendation Combos

MAKEMYTRIP

Recommendation EngineFigmaPersonalisationInformation Architecture

Finding the right accommodation for families and groups is challenging when dozens of room combinations, rate plans, and occupancy configurations are available.

This project focused on transforming MakeMyTrip's multi-room hotel booking experience by introducing a recommendation-driven approach that reduced decision complexity and surfaced the most relevant room combinations for different traveler needs. The final solution balanced user goals, business objectives, and inventory discoverability while creating a more intuitive booking journey.

Overview

MY ROLE

Research, UX Strategy, IA, UI Designs, Developer handoff, Stakeholder discussions (Product managers, Developers, Leadership)

DELIVERABLES

Figma high fidelity design (Web & App), Development audits

DURATION

Nov 2025 - Jan 2026

Go Live - Mar 2026

TEAM

Hotels UX

CHALLENGE

Booking accommodation for families and groups is significantly more complex than booking a single room. Users often need multiple rooms, multiple rate plans, and different occupancy configurations, creating hundreds of possible combinations for a single property.

As a result:

Users had to manually evaluate multiple room combinations with little guidance
Premium and alternative inventory remained under-discovered
Recommendation logic was optimised around inventory, not user goals
Decision-making became increasingly difficult for families and group travellers

OUTCOME

Designed and launched a recommendation-driven multi-room booking experience that transformed how families and groups discover accommodation options on MakeMyTrip.

By shifting from inventory-led combinations to intent-based recommendations, the experience reduced decision complexity, improved room discoverability, and enabled users to evaluate combinations based on value, convenience, and stay preferences.

The solution is currently live on MakeMyTrip and drives a 14% attach rate among family and group travelers booking multi-room stays.

The Transformation ✨

Objective

Reduce decision complexity

Make it easier for families and groups to compare multi-room accommodation options

Improve recommendation discoverability

Surface relevant room combinations and hidden inventory through intelligence

Align with traveler intent

Shift recommendations from room-count logic to traveler needs like staying together, best value, or increasing comfort

Build a scalable framework

Create a framework that can adapt to different occupancy, room, and property configurations across platforms

Current UI

Research & Insights

We analyzed booking patterns, selection behavior, occupancy configurations, and existing room-combination journeys to understand how users evaluated multi-room accommodation options.

Users rarely explored beyond the first recommendation

Most users select the first recommendation when multiple shown, with significantly lower engagement on secondary options.

68.2% selected the first combo when two options were shown

Multi-room bookings were concentrated

Users optimized for outcomes, not room structures

Through funnel analysis, we observed that users evaluated options based on goals such as staying together, best value, or getting larger rooms; not on room-count logic alone

Two-room combinations was the most common type amongst family and group travel scenarios.

88.8% of bookings used 2-room combinations

The challenge was helping users discover the most relevant combination faster. This led us to shift from an inventory-led recommendation model to an intent-driven recommendation framework centered around traveler goals.

Competitive Study

Structure of content

Menu Navigation

Core Vision

Design Vision

Context Recognition (traveller goals, occupancy)

Suggest most relevant options for quicker decisions

Transform stay booking into an intent-driven experience that guides decisions, simplifies comparisons, and highlights meaningful trade-offs between room types

Easily evaluate combos for the best experience (price, meal plans, room sizes, inclusions etc)

Clearly communicate why each combo is being recommended

Recommendation Framework

Instead of asking users to manually compare dozens of room combinations, the new experience groups inventory into recommendation buckets based on common traveller intents and booking behaviours.

Cheapest Combo

The most cost-effective room combination for the selected occupancy, helping budget-conscious travellers find the best value quickly.

Stay Together

Prioritizes combinations that keep travellers in the same room or as close together as possible.

Example: Family rooms, interconnected room

Spacious Stay

Highlights larger room configurations that offer more space, comfort, and privacy.

Example: Suites, Larger rooms with lounge area

MMT Recommends

A curated recommendation that balances value, comfort, occupancy fit, and overall stay experience to provide the best all-round option.

Iterations

Design Iterations

Multiple concepts were explored to determine how recommendation categories, room combinations, and pricing information could be presented without overwhelming users

ClarityBalancing DiscoverabilityComparisonInformation Hierarchy

I explored different approaches for filtering, category navigation, and progressive disclosure to help users move between recommendations while maintaining context.

Final design

Introduced intent-driven room recommendations that helped users discover, compare, and select multi-room combinations more confidently.

This is a interactive prototype! Scroll and explore the design!

Let's break this down!

Scaled to Mobile App too!

Learnings

My learnings and takeaway

Design for outcomes, not inventory

Travellers evaluate stays based on value, comfort, and proximity; not room structures and price only.

This project taught me that what matters to users is not only price but the overall value.

Recommendations need context

Users are more likely to engage when they understand why an option is being suggested.

This project taught me to design confidently for undecided users.

Complexity can be guided, not removed

Rather than hiding choices, the experience helped users navigate them through clearer recommendations and hierarchy.

This taught me how to synthesise content, and anchor decisions back to user impact. Data assisted in guiding design decisions was accepted by stakeholders.