Ironhack’s Prework Challenge 3 : Usability Evaluation & Site Redesign

Improving a Travel-Aggregator mobile App by User Testing @Ironhack UX/UI Bootcamp

9 min readApr 5, 2021

In this third diary entry of my journey to becoming a UX/UI designer I will be exploring : Detailing a Scenario ; Benchmarking with Nielson’s Heuristic Method ; User Testing ; Task Analysis for Insights and Site Redesign. For this case study we jump into the shoes of a world traveler.

Creating the Context

Scenario & Persona

3 lifelong friends in their mid 40s decide to go on an adventure. They want to see one of the world’s 7 wonders but also want to spend quality time together and fit in different personal interests and desires that they can do separately. They want to feel comfortable and spoil themselves but they don’t all initially have the same economical possibilities so the added-value-for-their-money aspect is taken into consideration when choosing a destination (even though they have been saving up). Their budget is 1300 euros each for flight and housing accommodations.

Destination & Trip specifications

Their heart finally sets on Rio de Janeiro in Brazil as it offers a large choice of diverse activities, is a good mix of sightseeing, nightlife and relaxation on the beach and is economically one of the more interesting “wonders of the world” options (euros to real).

They set on a 2 week trip which leaves them time to visit and relax and is manageable for all of them. After further research, they decided to plan their 14 day trip somewhere in September, October or November when it is warm (~26°C) and not too humid. Just before the Brazilian high season. The wardrobe recommendations are light and comfortable summer-ware. Except for eventual hiking boots they are traveling light and only need carry-on.

Although they would prefer a beautiful apartment or a house with a terrace to be able to have a more private setting they are open to getting a luxurious hotel. As for the location it doesn’t matter if it’s not the most central as long as it is not too far from the beach. They do not need any specific care or services except for wifi and laundry possibilities. The main focus of the housing is a beautiful and comfortable setting.

They know that taxis are not expensive in Brazil so they do not plan to rent a car for the full duration of their trip, but they do want to check out the situation for a 2-day car rental in the middle of their stay to be able to go on a day-trip, off the beaten path.

Choosing an app — Benchmarking

To provide an app to test, that performs well for my user scenario, I first did a quick Usability Heuristics Evaluation with Nielsen’s Principles and then looked for specific features & value propositions. Here is what I found :

Hopper: At the heart of the Hopper app is a prediction algorithm for future prices. Their main goal is to help the user book the cheapest flight and accommodation using advanced data science. Although it does offer good prices it does not test well in a few aspects of the Heuristic evaluation applied. It’s not easy to navigate as it hopps forward and backward on the same line without any shortcuts when you need to get back to a specific aspect or to change a setting. Keeping track of the number of people in between flight and hotel searches seems tricky as well. In sum it lacks in flexibility; efficiency; user control and freedom and does not have the most minimal design. What is very clear though is the low-price calendars. And finally, what distinguishes it most is that you don’t get redirected to another app to check out the “homes”. Yeah!

Kayak: “Metasearch for travel? No one was doing it. Until we did’’ Kayak states. It searches hundreds of travel sites to show travelers the information they need to find the right flights, hotels, rental cars and vacation packages. You can also access mobile-only deals. Best in test in the Heuristic Evaluation it also has interesting features like “Explore” and “Trip” meant to unify further the whole travel experience.

Trip Advisor: Is a user-generated content and comparison shopping website. It also offers online hotel reservations and bookings for transportation, lodging, travel experiences, and restaurants. Trip advisor also tests quite nicely on Nielsen’s heuristic scale. It does not provide a low-price calendar though nor other flexibility of the sort to help you choose the dates if these are not specific to start with. What it does have is extensive guidebook content as well as tours options and restaurant options to book, making it possible to plan all your days on the trip down to the tee and in advance.

The feel, or tone, is very different between these three apps but the value propositions were quite similar. Kayak and Trip Advisor did better on the Heuristic evaluation and between these two finalists Kayak offered features more suited to the user scenario. So after considering all the aspects I decided to run with Kayak : )

Testing the KAYAK app

My user persona is trying to unify two out of the three tasks (checking for flight and housing) to be able to balance these two factors. The goal is to find the cheapest deals for a maximum of comfort and pleasure within their budget. What is particular is that they only can take a two week trip but within a larger time frame. The focus for this test is on the budget and the number of days within a larger time frame.

The three tasks the user needs to perform :

Find a cheap round-trip to Rio de Janeiro for 14 days somewhere in September, October or November.
Find housing with a luxurious setting for the whole stay and with individual rooms.
Check out what it costs to rent a car for 2 days in the middle of their stay.

Five-second Test : All of the users recognized the app to be some kind of travel, search engine and could recall the flight button on the top to the left. One remembered the name KAYAK, “it’s a very beautiful name, poetic for a travel app” and another test subject knew the app but only for the web.

Before starting the 20 minutes test session, I did not make up a story on how the user encountered this app and I felt that this would have helped them and me. Something like “ A friend mentioned to you that Kayak has really interesting features to help you plan your trip and a lot of filters that really caters to your specific needs ”. This said, it did permit me to see that if you are not aware of these aspects, even though you fit the user type, you can not easily see that these options are available to you. Which is a shame.

Task Analysis

Observing the users in action to understand in detail how they perform their tasks and achieve their intended goals was very interesting. The users, all in their 40s, from different backgrounds and family situations, were all people that I know well and that I thought knew how they were going to operate for this given task but surprise surprise, I was wrong ; )- and I did get to know theme even better. : )

Insights from Testing

Overall, the users found the app easy to navigate and had no problem going back and forward to sections they were interested in. Most of them defined it as being easy and enabling you to go straight to the point. Unburdened by unnecessary information. Most liked the simplified pure interface. One did a general check of the app before engaging in the tasks, one had already used kayak on a computer and the third did not know the app and went straight for the tasks to accomplish.

To achieve their goal the users were thinking in terms of separate tasks. The redirection to other sites when looking at specifics for a reservation was not surprising to them nor was it appreciated as it demands a re-adaptation.

The main problems & friction areas were not so much in what the app is able to do nor in the user flow to accomplish certain tasks but rather clearly engaging the user in the amazing possibilities it offers and that meet a lot of the needs the users displayed.

After summarizing the results and findings of the tests I was able to pinpoint the things that could be improved and that were blocking the user from having a unified experience and optimal results in regards to context and the personas needs :

Seeing the “see cheapest dates” button in the research for flights.
Users don’t take time to explore “Explore”. The users attempted to use it but quickly abandoned.The map takes a lot of the focus and the filters become secondary to this feature. The “Budget” filter got noticed as an interesting filter option but was also quickly abandoned as it does not fit the need to balance the flight and hotel cost.
It was not practical for many to do the math in their head to subtract from the budget once the flight was found.
The ones that had never used kayak before did not realize there was a low price calendar (colors too faint) even though they were looking for it.
No one used the heart to take flight(s) into consideration.

Redesign proposition for KAYAK

Wireframe Redesign

A Delete button on recent searches. To many different date searches for the same destination is stacked up in “Recent searches” and lead to confusion. Confusion for the app as well it seems, when applying the flight dates to the hotel search.
A “Number of days” subsection in the Explore menu, under the “dates” section is appropriate for my user type. It’s an option that was cruelly missing for my scenario (being able to separately set the number of days for the trip in a larger period of time), without having to enter dates in order for the app to retain the number of days.
Unify the experience even more by adding hotel and car to the “budget” filter. Making it more a Calculator. Staying in-house to further look and scope for hotels (like hopper so nicely does) would also enhance this unified experience but this is not something I chose to work on here.
“Trips” is misconstrued as a collector of information for your tip once it is booked and is not thought of to be used for planning. I suggest a new first page in the image carousel that draws attention to the planning possibility in the Trip menu : Pin to Plan

Graphic Redesign & Visibility

In order to engage the user to profit from the features & filters it offers I propose to use color to draw attention to the options offered. I also suggest testing replacing the heart icon with a pin icon.

Heart to Pin icon_The heart was never used and was considered by one of the users as symbolizing a more “long-lasting relationship” and was not considered for planning, in this context.
I suggest the Kayak orange for “see cheapest dates” button, that takes up a lot of space but that nobody saw when looking for cheapest dates!!! Neither did I for a long time!
Concerning the low price calendar the colors are too faint (in dark-mode at least). The users all had used low price calendars before but most of them did not see the feature and they where looking for it (hastily). To attempt to fix the visibility I applied an orange stroke with no fill for the lowest prices and then used the orange to red color scheme when prices go up.

In conclusion I found that determining the appropriate content scope is very important, not only to get the most out of the test but also to help focus on what aspects to take into consideration in an redesign of the app. People are so wonderfully different and will always give you tons of preferences and ideas for a redesign. Focus permits you to stay within the context you defined and helps you stay on track of the task.