Where did most of the planning happen?
- WhatsApp group20
- Calls or voice notes13
- In person8
- Shared doc or sheet4
- Instagram or DMs2
Multiple choice. Calls fill in what the chat can't settle.
Conceptual case study · Product design · MakeMyTrip
Trip Squad
Plan a group trip where you book it. Everyone shares a private budget, the group votes, and whatever wins gets booked in the same app.
Independent concept. Not affiliated with or endorsed by MakeMyTrip.
- Role
- Product designer: research, UX, UI, prototype
- Type
- Self-initiated concept, solo
- Method
- Survey, affinity mapping, interactive prototype
- Tools
- Next.js prototype built with ChatGPT and Claude
Problem
→Group trips are planned in WhatsApp, booked by one person across different apps, and paid back later.
Idea
→Move the deciding into MakeMyTrip: private preferences, polls of live options, booking once the group agrees.
Output
A 22-step interactive prototype across five chapters, including payment failure and AI no-match states.
01 · Problem
Every group trip starts in a chat and stalls there.
🏖️
Goa trip Dec 🌴Aarav, Aditi, Rahul, Riya, Vandana, You
Today
RahulGoa in December? 🔥 Who's in7:02 pm
1RiyaI'm in but what's the budget though7:04 pm
AaravAnything works for me 🤷7:05 pm
Chennai to Goa flights from ₹5,150Book cheap flights on IndiGo, Air India, Akasa Air…makemytrip.comhttps://www.makemytrip.com/flight/search?itinerary=MAA-GOI-12/12/2026Found this, 6:10 am non-stop 👇7:11 pm
3AditiI'd rather not fly tbh, train?7:26 pm
142 unread messages2VandanaWait, did we decide the hotel??11:48 pm
4I'll just book everything, send me your share 🙏11:52 pm
Message
- 1
Nobody says a number
The plan is pitched at a guess, and people quietly drop out when the prices land.
“Making sure everyone's budget was on the same level” - 2
Agreement gets buried
“Did we decide the hotel?” gets asked a hundred messages after it was decided.
20 of 23 planned mainly in a WhatsApp group - 3
Limits surface too late
“I'd rather not fly” arrives after the flight has already been shared.
“Everyone having different ideas” - 4
One person carries the booking
They book across apps, pay for everyone, then chase the money afterwards.
12 of 23 had one person book it all
02 · Research
I asked 24 people how their last group trip was planned.
A short online survey (20–26 June 2026) on where planning happened, who booked, what took the most energy, and how money was settled. 23 people answered; 21 had been on a trip with three or more people in the last year.
20/23planned mainly in a WhatsApp group
12/23had one person handle all the bookings
14/23booked through different apps for different things
13/23usually book travel on MakeMyTrip
Survey results
Affinity mapping
I put every open answer about what took the most time or energy, and what people had just accepted, on a note, then grouped them until four clusters stood out. Quotes are verbatim.
Aligning budgets3
Making sure everyone's budget was on the same level
Some friends agree and some don't agree on where to stay, to increase the budget or to decrease the budget
Decision making like this offer is good or not or this stay is worth it to pay
Agreeing on anything4
Considering everyones opinion
Everyone having different ideas
Different people with Different opinion
Making last minute approvals
One person coordinates3
Communicating with all members about the availability of trains and seats. Clearing their doubts and confusions
Fixing on a plan (location), waiting for offers for the flight, gathering details of all the people
One person paying and the rest transferring later
Finding and trusting options4
Comparing with the rest of the apps
Confirming if the options showcased is true or fake
Finding the suitable property
Comparing pakage
Four themes, four design responses
- 01
Aligning budgets is the hardest part
Nobody wants to say their number out loud, so plans are pitched at a guess.
Design responseCollect budgets privately and show the group one number everyone can afford. - 02
Deciding takes longer than booking
Different opinions and last-minute approvals stretch planning over days.
Design responseGive every decision a visible state: who voted, what leads, and when it's agreed. - 03
One person coordinates everything
12 of 23 had one person book it all and answer everyone's doubts.
Design responseLet anyone add options and book once the group agrees, so the work is shared. - 04
Options are compared across apps
14 of 23 booked through different apps and worried if listings were real.
Design responseKeep every option inside MakeMyTrip, priced for the group's dates and size.
What surprised me
I assumed
Settling money afterwards would be the biggest pain.
The survey showed4 of 23
mentioned anything unresolved about money. Budgets before the trip were the real friction.
I assumed
Most people plan, so everyone feels the pain.
The survey showed13 of 21
were along for the ride, not planning. The product has to work for them in 30 seconds.
I assumed
Groups are small, three or four friends.
The survey showed11 of 23
trips had five or more people, where a chat breaks down fastest.
So what did I do with it?
I moved the weight of the product from paying back to deciding: private budgets, a group summary, and polls with a clear majority. Split payment stays, but as a light option at checkout, not a feature of its own. 23 answers is a small sample, so I treated the data as design constraints, not proof of demand.
03 · Reframing
The problem wasn't finding options. It was agreeing on one.
How might we help a group agree on a trip, fairly and quickly, inside the app where they'll book it?
- 01
Private inputs, shared outcome
Budgets stay private. The group sees one number everyone can afford.
In the product: preferences and the group summary - 02
Every decision has a state
Votes with faces, a clear majority, and a discussion attached to each option.
In the product: polls and per-option threads - 03
Decide where you book
Options are live inventory, and anything agreed can be booked in one checkout.
In the product: Add to group trip and Book now
04 · Solution
Five chapters, 22 steps.
The key screens below are the live prototype. Each line points to the exact element it describes; choose Try this screen to use it.
- 1DiscoverAdd any result to a group trip
- 2StartCreate, invite, add preferences
- 3DecideGroup summary and destination poll
- 4Plan and votePolls, discussion, Myra
- 5BookCheckout, split, recovery
01Discover
Start from where people already search.
- Problem
- 13 of 23 already book on MakeMyTrip, but a group trip starts with someone pasting a fare into WhatsApp.
- Decision
- Add to group trip on every result, as the secondary action next to Book.
- Why
- It turns a pasted link into an option the whole group can vote on, without leaving the app.
- Trade-off
- A second button adds weight to a dense listing, so Book stays primary.
02Start
Thirty seconds of input from each person.
- Problem
- “Making sure everyone's budget was on the same level” was the most-cited time sink, and 13 of 21 were participants, not planners.
- Decision
- One short form for everyone: budget, dates, mood and anything to avoid, with the budget kept private.
- Why
- People give an honest number when only a combined figure is shown to the group.
- Trade-off
- A form is a step before anything fun. It's short, and moods and limits accept free text.
03Decide
A poll that knows the group.
- Problem
- “Everyone having different ideas” turns choosing a place into days of back-and-forth.
- Decision
- Myra sums up everyone's replies in one line, ranks places with plain reasons, and the group votes. A majority locks it in.
- Why
- Reasons like “Fits everyone's budget” make trade-offs visible without exposing anyone's number.
- Trade-off
- A ranking can feel like the AI deciding, so it only orders the list. The vote decides.
04Plan and vote
Every option carries its own conversation.
- Problem
- In chat, the reason someone preferred the train is lost by the time anyone books.
- Decision
- Transport, stay and activities are polls of live options. People can vote for several; the most votes leads. Each option has its own thread.
- Why
- Discussion stays attached to the thing being discussed, so anyone catching up sees why.
- Trade-off
- Separate threads spread the conversation out. General chat still lives in Notes.
05Book
Book what's agreed, in one checkout.
- Problem
- 12 of 23 had one person book everything and 12 had one person pay and collect later.
- Decision
- Anyone can book once an option has a majority. Checkout reviews everything agreed together, with a light option to split the bill.
- Why
- Booking stops being one person's job, and the money conversation happens at the moment of payment.
- Trade-off
- Split payments can stall, so a failure always says no money was deducted and how long seats are held.
05 · Build and QA
Designed by me, built with AI, tested by hand.
I did the research and product decisions, then used ChatGPT and Claude to build the prototype in MakeMyTrip's design language, taken from their live mobile site. I used every flow, logged what was wrong, described it precisely and had it fixed.
Issue found while testingWhereStatus
Screens contradicted each other: 1 of 6 replied, then 4 of 6 agreedFlowFixed
A bottom sheet stayed open after moving to another screenNavigationFixed
Scroll position leaked from one screen into the nextNavigationFixed
Two destinations used the same photoContentFixed
Myra suggested a 16-hour bus to someone avoiding long road tripsAI logicFixed
The phone keyboard left the page pushed up after typingMobileFixed
Back from a ticket returned to the very first screenNavigationFixed
- All 22 steps and 10 quick states open on one consistent trip
- Clicked through in the browser with no console errors
- 54 automated flow tests passing
- Type-check and lint clean
- Fares from real distances between cities
- Checked on phone, laptop and large monitor
06 · Reflection
What I learned
Takeaways
I went in designing a way to split the bill. The survey said the hard part happens weeks earlier, when nobody wants to say their budget out loud. Listening to that changed what the product is.
A group product also lives or dies on consistency. One screen that disagrees with the last breaks trust faster than a missing feature, so the prototype is built around a single shared trip state.
Limitations and next steps
- Small sample23 answers.
- Not usability testedBuilt from survey insight alone.
- Single deviceFriends' replies and votes are simulated.
- TiesMultiple votes are allowed.
- Simulated bookingPrices, payments and tickets are illustrative.
Try the prototype
Start from Home, or use the step list to jump to any part of the journey.
Open the prototype