Share Now is a free-floating car-sharing service, historically providing an option to users in spontaneous need of a car. To extend the service offering, the company decided to target the conventional car-sharing market by allowing users to plan and book a car days in advance. Experiments around this service in the market paid off and the service developed into one of the main strategic pillars of the company. This case study tells the story of how we gave this amazing service the stage it deserved.

2021 - 2 Months

◇ The Goals

◇ Starting Point

There was a lot of uncertainty around the success and impact of an extension of the service offering in the direction of pre-booking. Also, the technical and operational challenges seemed quite daunting.

Therefore, we installed a dedicated learning team, feeding findings and observations directly to the design and development teams. Choosing embedded web views as front-end technology enabled us to push iterations and fixes to the customer also outside of app release cycles, making us adjust the product even faster. The only native part was the entry point into the funnel.

After a few months of learning and iterating both on the front stage as well as the back stage, we had worked out most kinks and automated a lot of the internal processes making the service reliable and fast. The entry point into the funnel was moved from its initial location in the menu onto the map, already increasing exposure noticeably.

We could have left it at that. However during numerous user tests both related and unrelated to pre-booking, we learned that users were quite confused that such a powerful feature was positioned in a way that was often perceived as an afterthought.

◇ Bottom Navigation

So far, the Share Now bottom navigation has been used more as a toolbar than a true tap bar. This may have been ok with only one main functionality available to the user (that being spontaneous rentals). With a new service being offered to the customers, the notion of the main navigational element of the app being misused gained traction.

With the introduction and manifestation of pre-booking however, a genuine tab bar was the better use of the app's main navigation.

◇ Main Objectives

Objective 1

To turn the bottom navigation into a true tab bar, we decided to offer the user a choice between taking a car off the streets spontaneously or planning for a future trip. We experimented also with different labels, and active states and included the entry into the user’s account as a third option.

Objective 2

With clearing out the elements from the bottom navigation to make room for spontaneous rentals and pre-booked trips to shine, we needed to find a good home for the former occupants of that bit of screen real estate.

We did some competitor analyses looking at companies like former DriveNow, Google maps, Miles, Sixt Share.

With almost all elements being related to the map and how cars are presented (except Accounts), we opted to experiment with placing them directly on the map. To not clutter the space too much, we chose to cluster what made sense from a conceptual POV.

Synergies with another team

During alignment conversations with the delivery team working on the rental experience, we learned that they were working on a feature that had some potential overlap. The team had learned from user feedback that finding gas stations and charging stations was difficult as they were not displayed on the map according to real-life users. Gas stations and other map elements were already enabled to be toggled on or off in the menu, but this feature was too hidden for most users to notice.

◇ Testing

To learn whether or not we were on the right track before going into testing, we went ahead with a round of qualitative user testing to validate the concept.

🤷

🤷 The assumption around map layers: Users struggle with finding these optional because they can not find the settings in the menu structure.

🤔

🤔 Hypothesis: Clustering all map-related elements in a single entry point on the map will help with discoverability and will reduce the friction around this topic for users.

🤷

🤷 Assumption around former toolbar: With these elements not being main actions, users don’t interact with them very frequently

🤔

🤔 Hypothesis: Clustering them in a designated entry point will free up real estate for the main JTBD while keeping the options within arms reach.

Scope and test setup

→ Play with Prototype

Learnings

◇ Iterating

Being successful with our 1st goal, we focused our efforts on iterating around the second goal to reduce the learned confusion with the admittedly quite complex top navigation element.

1. Combining filter and list in one view would provide more meaning to the otherwise useless list view.
2. The search functionality’s prominence would be reduced.
3. Map layers would receive a label for clarity

→ Play with Prototype

◇ Implementation

The adjustments made to the initial concept were not seen as in need of another round of testing by our UXR colleagues since the main goal regarding the bottom navigation was a success. The changes to the top navigation did not seem to have a massive impact on a successful outcome.

As a next step, we set together with the engineering team for final grooming to reduce complexity from an engineering POV. This meant a few minor adjustments:

◇ Next steps

One thing that kept coming up in testing was the notion that the car visible in the default map screen would be the very car that one would plan the trip with when taping on the plan. This is however not the case and was quickly discovered by users on the following screen. However, we have worked on a North Star vision on how to potentially tackle this and improve the overall maturity of the product.

Investigating further into this would be a good next step to improve the pre-booking exposure even more.