
Role: Contracted Founding Product Designer for the startup, responsible for driving the entire design of both the consumer facing product and the hosts' Property Management System.
Experiences: Mobile-first room search and booking experiences. Hosts' CMS and property management platform.
Tools: Figma, UserTesting, ChatGPT, Otter.ai, Balsamiq, Google Analytics, SurveyMonkey
Timeframe: September 2024 - July 2025
Similar to AirBnB and Booking.com, bnbme is an online booking service for short-stay and holiday homes.

The focus is high-end luxurious travel-experiences for people traveling in the Middle East. The rooms are offered by vetted hoteliers with experience in the field coupled with a range of extra add-ons to makes someone's stay more comfortable.

To re-design everything from scratch, including all traveler experiences, the hosts' landing page and the hosts' management tools. All the consumer experiences should be designed with a mobile first approach and support an Arabic version.

Image generated with ChatGPT.
"Can I improve the conversion rates in all parts of the website?"
"Can I design the Property Management System in-house on top of the rest?"
"Can I design a mobile-first approach and support our Arabic users too?"
At first I started running workshops to extract what the owners' vision was, in order to understand how to prioritise my process. 
The Chief Commercial Officer, the Lead Product Manager and the Tech Lead also joined the workshops, so we could drive towards a vision.

I looked at Google Analytics, customer reviews and run a survey with SurveyMonkey on the consumers' website. The survey was completed by 26 people.
I ran research workshops with hosts that have their properties listed with bnbme and some that were on other platforms. I used Otter.AI as a notetaker, and connected it with ChatGPT to help me summarise the findings.

This was the number of total views of the landing page recorded in a period of 3 months.
There was no booking completed from the website, meaning that we failed to convert any traffic.
This was the cost that bnbme paid to use an external platform for the hosts' room management.
"If we design the website to look familiar, visitors will feel more comfortable to book their holidays through bnbme."
"If we bring the room management in-house, we will save the HostHub fees, and more hosts will list their rooms with us."
After talking to users and running the surveys I created 4 personas which helped me understand the target audience from Europe/USA and from Arabic countries. The personas also helped put into perspective the hosts I wanted to design for.

Together with the owner and the Lead Product manager, we had a product vision workshop. I used ChatGPT for story mapping and to generate some visual insperation for the UI.

Image generated with ChatGPT.
From Google Analytics, it seemed like the website had very low conversion rates throughout. This meant that we would need to prioritise our releases.
With the owners' approval, I decided to prioritise the users' first touch points, so focus on the Landing page, Search results and Room details pages. This also made sense, because these were the pages that we already had ownership of, and were not hosted by Beds24.

At first I mapped out the user flow for the consumers to understand their experience when searching for a room they want to rent for their trip.

After the hosts told me that they mainly rely on Airbnb and Booking.com, I decided research and analyse these two websites in depth.

I generated some concepts that look closer to Airbnb with more prominent views of the map and a focus on a "discovery" experience. I also generated some concepts closer to Booking.com with a bigger hero section on the landing page and more detailed filtering system.

Image generated with ChatGPT.
As competitors stated, most users book either through Airbnb or Booking.com, so I decided to design a version that is similar to Airbnb, a second version similar to Booking.com and a third version that is a mix of both.
I prototyped and experimented on these three versions with 50 participants on UserTesting.

Results from UserTesing experimentation.
Users liked the Hero section, the "What is around" section and the "Trending destinations".
The "Explore" experience, "Search on the map" section "Room details" felt cleaner and were preferred.
As anticipated, users liked the combined version the most, because it felt clean with enough details.
Total number of testers: 50
Due to the lack of trust to the website I decided to run rebranding workshops with the owner, CCO and Lead Product Manager. The main focus was to keep the gold theme as a common luxury colour and a desirable colour in the Middle East. We concluded in a colour palette inspired by Nescafe gold, and I created a Typescale using <Inter> because it would be compatible with the Arabic version.

I created a design system from scratch with the idea that it will make the Arabic support very efficient. I designed everything with the philosophy to be aligned either to the left or to the right, and little content being centrally aligned. This made it easy to mirror everything and the components helped to efficiently reflect the content.

After the first release, I recruited people using UserTesting and run research sessions. I also run a new survey on the consumers' platform with SurveyMonkey.
I researched and found that airlines have the best upselling booking flows in tourism. I analysed airlines that are famous for their luxurious experience and are also close to the Arabic world. I found out that the best example to follow would be Ryanair as the best upselling flow and most familiar with Europeans, with easy skippable steps and accessible progress tracker.

After I saw how airlines structure their booking experiences, I revised the booking flow, to improve the conversion and increase the upselling of the add-ons. I decided to design the upselling steps to be skippable.
In the past, when I designed high-converting onboarding flows, I changed steps to be optional, and observed that users were more likely to complete a step because they want to, and less likely to complete a step if they are forced to.

The major elements that we needed were the "Skip" buttons on each step, to not force anything to users, and an accessible progress tracker to reduce the stress from users on lengthier tasks.
The UI needed to be more "eye-catching" like Ryanair's in contrast to boring UI like Emirate Airlines'. More "white space" to make the page look less crowded and more visual content.
I decided to prototype and test a version booking process with the "Next" and the "Skip" buttons on the top right corner of the PC version and anchored above the menu bar on mobile. The intention behind this was to allow users to flow through these steps quickly and not interrupt their momentum when booking.
I mapped a user flow to understand how we would drive more hosts to register and list their rooms with bnbme.

The main thing was to convince hosts that this would be a profitable choice for them and keeping the sales contacting spots as accessible as possible.

New hosts said to our sales team, that they would prefer to list their rooms with a platform that they don't need to pay extra. Would we have a return of investment if we built our own PMS with a host Dashboard?
*Property Management System
After doing more workshops with some of bnbme's existing hosts, and exploring other PMS tools, I collected all the insights and details I needed. I used them to create prompts and form an idea generating chat in ChaGPT to help me design the PMS.

Images generated with ChatGPT.
I stripped the dashboard down to the tools that would be used more than 90% of the users, and excluded tools used less than 1% of a hosts monthly usage, to keep the production cost down, but build usefull product.
The platform features a dashboard with a calendar and latest bookings widgets, a messaging app for hosts to chat to visitors, incoming requests widget and performance highlights. 
The rest of the platform is locked behind a disclosure agreement and I am unable to share more.

This is what we ended up achieving by the end of my contract.
Recorded in 3 months since the release of the new booking journey.
More hosts registered listing more rooms for Dubai, but also expanded in Riyadh.
I captured these metrics with Google Analytics in a timeframe of three months. Starting from the day that the new booking flow was released and ending 2 weeks after the PMS was released. This is where my contract ended, but a growth in those numbers is anticipated for the next year, as more people find out about bnbme.
©Evangelos Angelis ️