Who is Kiwi Collection?

Kiwi Collection is a premier online travel agency for luxury hotel needs. Since 2003, Kiwi Collection has provided expert guidance on the best hotels–from boutique gems to celebrated luxury brands. Kiwi also has a 10 year+ and ongoing partnership with Visa in handling their high-end clients and hotel partnerships.

My challenge

At the end of 2018, I joined Kiwi Collection as a UX Designer to assist the Senior UX Designer in improving our companies’ conversion rate optimization initiative and introducing a brand new UX culture. Working closely with the Senior UX Designer and Product Managers, we’ve created a plan to achieve this business goal through data, design and technology.

Problem space

In order to improve conversion, first we needed to understand our customers. That’s when we realized the company’s user personas have not been updated in years and they were originally created without true customer insights–data and user research.

Users would spend less than 5 minutes browsing a hotel page. Though they made their way to checkout, they weren’t converting. Why?

Part 1: Research and Personas

Working with the analytics team, we sent out a questionnaire to customers who have travelled with us in the past year. We’ve also collected customer data through various channels (Hotjar polls, Hotjar user recordings, Snapgengage customers conversations, weekly user conversations through our Concierge team, Survey Monkey post stay surveys and tools like Google Analytics) to collect as much quantitative and qualitative data as possible.

We were also able to get in touch with 11 customers to in order to gain further insights through phone interviews.

Results

In the past, Kiwi Collection has identified their core customer as Sophia. Within 2018, the company has spent over $120,000 marketing to female millennials. In fact, 70% of all email subscribers were females, despite male subscribers accounted for 63% of all travel bookings in 2018.

Therefore, we pivoted from our target user Sophia (aspirational user) to John and Ryan as our core customers.

Core Users

Though males accounted for higher bookings, our customers rarely travelled alone. In the last year, 84% of all our bookings included more than 1 traveller and 48% included at least 1 child.

In reality, our users found luxury hotel collections too expensive for business or regular stays. It was fitted for special occasions–birthdays, anniversaries, baby moon etc. Our questionnaire results showed, 53% of all our customers travel for family vacations and 46% travelled for romantic getaways.

What are their values & frustrations?

We asked customers to rate how disappointed they would be if Kiwi Collection no longer existed, and their answers were:

46% answered ‘very disappointed’:

• they would lose a curated collection of luxury hotels

• they would lose specific details as to why we love a specific hotel

• they would also lose exclusive perks & offers we offered through through Kiwi in general or Visa card specific perks

53% answered ‘somewhat disappointed’

• these customers identified they could book through with our competitors

• they enjoyed perks & offers elsewhere

• they were fine with not booking with Kiwi

1% answered ‘not disappointed’

• they simply didn’t care for the existence of Kiwi Collection

We asked, why 54% of our customers would only be ‘somewhat disappointed’ and 1% ‘not disappointed’ at all:

1. We lacked a loyalty program our competitors had

2. Our site speed was much slower than our competitors

3. Though we provided customers details as to why we loved a hotel, we weren’t providing them enough hotel details. Customers had to search elsewhere (leaving Kiwi Collection) for more details.

Those who valued us:

1. Our curated hotel collection–personally selected, rated and reviewed by our industry experts.

2. Customers not only wanted to see the best, but also wanted insider knowledge on unique experiences as to why we love a specific hotel or destination. In fact, customers themselves said they had no interest in guest reviews because they could be bias and untrustworthy.

3. Our exclusive Kiwi and Visa perks. These added extras were the justification to why our customers were willing to spend extra money on luxury travel.

Storyboarding

In conclusion, we determined the product team would focus on 2 of the 3 top user frustrations: slow site speed and hotel details. In the interim, the UX team would assist the development team with site-speed improvements and further collaborate with the content team with adding hotel details.

Part 2: The Design

Header

1. We wanted to move away from the original tab layout and design a page that encourages natural scrolling (anchor layout)

   • Our Baymard UX audit identified tab layouts were found to perform the worst compared to other layouts (accordion, anchor, vertical and tab). It also kept content hidden from our users.

2. We introduced an edge to edge full screen gallery to showcase our hotel imagery.

   • Our users have emphasized on the love of the beautiful imagery on our site.

   • In the past we used a 1024px autorotating carousel, which did not perform well. Additionally, the carousel skewed out click data in Hotjar.

3. We emphasized the hotel title–seen always within first glance at any size.

Image gallery

1. ‘See all photos’ from the previous image brings a users into a photo gallery that is opened into it’s own page.

   • Uses are now able to scroll through hotel imagery rather than clicking an arrow multiple times.

   • Images would be placed in the order of how a customer would explore the hotel in person. They would scroll through the images, creating the same visual as if they were walking through the hotel.

   • Hotel images are also categorized and users will be able to skim through photos by area. Previously users would click through a carousel and guess which area they’re looking at.

2. This new user experience would also perform better on mobile devices. We’ve learned from our data, carousels do not perform well on mobile screens.

3. This would also increase the speed for page load. Our previous design required the loading of 15+ retina display images at once, which had a major impact on our site speed.

A quote from our customer interview:

“I think the new photo feature is better cause you only have to click once and scroll. I know before you had to click click click, so removing clicks is big. Especially on mobile… the images are bigger and categorized. The scrolling is much more natural, it’s much better”

- Amy

Kiwi Rating & ‘Why we love it’

1. From our user interviews, we found our curated collection to be a big conversion driver. Therefore, each hotel now contains a rating from our industry expert. We know our users filter through hotels using this rating, and in order to justify the rating given, we’re also providing valuable content and experiences they’ll find unique to each property.

2. The content curation is a continuous collaboration between the UX and Content team.

3. When a user starts scrolling, a sticky bar appears at the bottom of the page. The design feature allows the primary CTA to always be visible to the user.

Exclusive Perks & Benefits

1. Exclusive perks and Visa Premium Benefits is also another conversion driver for Kiwi Collection.

   • Here we highlight all the extras our users get when booking this hotel with us

   • Also, the icons next to the titles are to help users select the proper rate associated with the perks & benefits

Location, Amenities, Things to know

1. To help our customers decide on the right hotel, we also needed to sell services and experiences around the location.

   • Working with the Content team and referring to our survey results, we came up with an MVP of all the basic content a customer would like to see.

   • We hope to test engagement in this area and explore how we can further improve this area in phase 2

Questions, Nearby hotels, Blog

1. If users have questions, they have the option to send a message or call. This provides help for users when needed and avoids increasing page’s bounce rate.

2. Nearby hotels allows the user to see the other luxury hotel options around the area. That way they’re not limited to only the property they are viewing.

3. Our own blog articles categorized by destinations can also be found in this section.

Room options

1. When viewing room options, room options opens to a new page. Our A/B testing resulted 5 out of 7 users prefer having a separate booking room options experience. From a UX perspective, it is also a better experience for our users when they are not distracted by hotel information and are able to focus on finding a room option.

   “ I really like this new layout. The options are clear, the perks are all there and I can see the price … I think you got the right level of information here. By this point, I only want to see price and perk options. The total price is also a nice to have as I often have a budget in mind.” – Maurice

Conclusion

What happened

1. Increased 38% traffic from hotel page to checkout

2. Increased 43% engagement on the hotel page

3. Increased average time spent on the hotel page by 15 mins

4. Page scroll increased by 12%

5. Baymard UX audit score increased from 5/32 to 18/32.

Next Steps:

1. Although our new design increased checkout traffic by 38%, the bounce rate on the checkout page also increased by 27%. Our next steps is to investigate why users are leaving checkout.

2. Keep track of our KPIs and observe new data and user behaviours on the new design.

3. Continue to improve our UX audit score.

What I learned

I’ve learned the importance of research and sharing research results with other stakeholders. Having other stakeholders involved early in the process decreases the chances of complications and limitations later in the project. With early involvement of other team stakeholders, it allowed me to understand their business goals and perspectives.

UX design was a brand new role at Kiwi Collection. By conducting research and sharing results with the entire company, we’re able to introduce a new aspect into our company culture, where UX design brings value to the business. Not only did we get to dive deeper to get a better understanding of our users, but we were able to understand the long occurring frustrations our users have had while booking with us.

Though we were able to achieve the initial goal of increasing traffic to checkout by 30%, there is still lots to learn about our users. With the new design, we’ll have to complete new research in order to understand why our bounce rate at checkout has increased as well. We’re hoping to generate more hotel content to bring a real life luxury hotel experience into a digital space.