AIR MILES

I interned at AIR MILES as a Product Designer in Winter 2016.

AIR MILES is Canada’s largest coalition loyalty program, with more than ten million active Collector accounts and approximately two-thirds of Canadian households participating in the program. Collectors earn reward Miles by shopping with select sponsors, which can be used to redeem free flights, hotel accommodations, car bookings, merchandise, and more. AIR MILES has over 100 sponsors, including American Express, Toys R Us, Amazon, and more.

The Problem

The existing car and hotel booking experiences are drastically hurting conversion tunnels

Issues with the AIR MILES Travel booking experiences (flight, car, and hotel) is one, if not the greatest, driver of calls to the call centre, with wait times of up to 4 hours during peak seasons for travel. As a result, AIR MILES needed a redesign of the car and hotel booking experience to address the usability issues that are impacting the conversion tunnel and maintain consistency across the platform following the recent redesign of the flight booking experience.

Role and Tools

For this project, I worked alongside another designer and a UX researcher. I was responsible for conducting heuristic evaluations and competitive analyses and worked closely with the UX researcher to develop plans for user testing and focus group sessions. I also designed and prototyped interfaces based on the findings and worked closely collaborated with the developers to ensure their implementation of my work was as accurate as possible.

• Pen and paper, for ideation, sketches, and user research
• Photoshop for interface design
• Axure for prototyping

Research

Heuristic evaluations, competitive analyses, and lots of conversations with users and call centre employees

To gain insight into the state of the art, I applied the following research methods to identify the areas of improvement for the current experience.

Heuristic Evaluation

I first performed a heuristic evaluation of the existing car and hotel booking experiences against Jakob Nielsen’s design heuristics to identify usability issues. Some findings that came out of the evaluation included poor error recovery and visibility of certain page elements.

User Testing

Next, I conducted user testing with active AIR MILES Collectors to observe and understand how they actually interacted with the platform. To accomplish this, a total of seven Collectors were invited to the AIR MILES office. These Collectors varied in age (late 20s to late 60s) and familiarity with technology (inexperienced with use of computers to expert users) and were a good representation of the diverse range of Collectors that continue to collect Miles today.

The objective of user testing was to determine the following:

Comprehension

• On car and hotel booking details (search, location of car, map, insurance, fees)
• On navigating throughout the end-to-end booking experience

Ease of Use

• Ability to enter a valid search request
• Ability to select a car/hotel nearby the arrival location
• Ability to complete the booking process without guidance

One of the biggest findings from the user testing sessions was the fact that 7 out of 7 Collectors were not aware of the map feature existed within the search results. We also verified that maps are a critical feature for selecting a car rental location, as some Collectors stated that they would not proceed with booking if they did not visually see the pick-up location relative to their entered destination.

Focus Groups

To further validate pain points of Collectors, my team and I drove down to the Care Centre to organize a focus group with five Customer Care Specialists. These specialists understand the needs of our Collectors very well, as they hear about the frustrations and concerns of our customers on a daily basis. Speaking with the specialists that deal directly with Collectors allowed us to gain valuable insight into potential issues that may not have been obvious to spot in other UX research methods.

From the focus group sessions, we learned that Collectors often do not complete a booking because AIR MILES does not actually accept prepaid credit cards as a valid payment option, a fact that only comes to the Collector’s attention at the very end of the flow (when they are prompted to enter their payment information!). We also learned that a significant portion of calls to the centre were from Collectors confirming their bookings–there is up to a 24-hour delay on confirmation emails, but this information was not displayed on the platform either.

Competitive Analysis

Lastly, I conducted competitive analyses of our competitors and other existing car and hotel booking platforms to identify the latest web usability trends in the travel industry.

Some competitors that were analyzed included Expedia, TripAdvisor, Kayak, and Enterprise

Using the findings acquired from the methods listed above, I was able to discover emerging themes and patterns that played an integral role in whether or not the Collector completed their booking successfully. I then organized the findings into six major themes: technical bugs, insufficient search results, usability/findability issues, booking flow issues, error handling issues, and overall comprehension.

The Solution

After a few iterations and testing sessions (both internally and with actual Collectors), the following designs were proposed.

Map

Increasing visbility of the map feature on the search results page was one of the most important items to address. In the previous design, the map was collapsed and had a small link titled "View map" on the far right side of a thin grey rectangle. For better visibility, the final design had the map expanded by default, and the "View map" link was made considerably larger and more noticeable by adding an accompanying location "pin" icon to the right of it to indicate that the map view can be toggled.

Final design for default search results with expanded map

Error Handling

We also eliminated the use of vague error messaging in search results. Previously, when there are no car or hotels available based on the user’s entered criteria, a message simply telling the user to modify their search criteria and try again was presented. Playing on the error prevention design heuristic, the redesigned errors were made to be more informative and provided users with plausible explanations as to why an error occured, and how they can recover from it.

Other design changes that were made addressed the issues that were identified through the user research methods. Some of these issues include:

• Lack of awareness of accepted payment methods upfront (previously only explicitly stated on the actual checkout screen)
• Lack of awareness of the fact that bookings take up to 24 hours until they are confirmed (Collectors would often call immediately after booking to confirm)
• Poor visibility of navigation items and Online Support feature

Next Steps

Measuring success

Unfortunately, my internship ended before most of these designs were implemented. The next step would have been to test whether or not the new design was successful in achieving the intended goals.

If given the opportunity, the most important metrics that I would measure are:

• Conversion tunnel - Are more Collectors completing bookings successfully?
• Bounce rates - Have bounce rates decreased on screens that previously had high drop offs?
• Usage of built-in features - Are Collectors using the Map and Online Support features as a result of better visibility?