How we improved Bolt users’ airport travel experience

Despite being a ride-hailing company operating in over 40 countries, Bolt still needed to catch up with our competitors regarding our share of airport revenue. This was a crucial area to address, as airport trips generally have a higher average fare than other types of trips.

Furthermore, airport rides represent a gateway for riders to use Bolt throughout their trip, making it a significant opportunity to improve our competitive position and drive revenue growth across our markets.

When I joined Bolt as a Product Manager, I was tasked with building and executing an airport strategy to address this challenge. It quickly became apparent that this would require significant cross-functional collaboration.

It was crucial that the Rider Experience team could connect effectively across other product areas such as geo, driver, and marketplace, as well as with non-tech stakeholders.

Problem

The airport pick-up experience for our Bolt users at the time was not ideal. Upon arrival, riders would encounter a confusing list of terminals for their pick-up spot.

Unfortunately, the assigned pick-up location wasn’t always accurate, leading to confusion for both riders and drivers. To facilitate pick-ups, drivers would resort to manually calling riders. Some entrepreneurial drivers even had pre-filled copy-and-paste messages to redirect riders to the correct pick-up location.

The airport pick-up experience could clearly have been better, with significant areas for improvement. Our analysis showed that airport rides had a lower finish rate than city rides, primarily due to higher cancellation rates and longer waiting times for riders.

Building a framework

To solve this problem effectively, the team employed a multi-layered approach.

Firstly, we established a data analytics system that accurately tracked the revenue generated from airport rides. This was crucial, allowing us to make informed decisions based on reliable data.

Next, we identified user points, improvement areas, and specific success metrics. Using this information, we created a roadmap of potential features, prioritising them based on estimated impact and effort.

Once we’d drafted the vision, strategy, and roadmap, we worked closely with stakeholders to finalise feature requirements and iterated improvements based on user feedback.

A data-driven approach

To start this, we built a proper tracking system that accurately measured the number of airport rides within a specific market.

Previously, some airports needed to be included in our data. Depending on the market, certain rides were wrongly classified as airport rides and vice versa, leading to overcounting or undercounting of airports.

To address these issues, the Geo team conducted a comprehensive review of the airport geofenced areas and established a standardised labelling terminology that our local teams could utilise.

After this extensive process, we finally obtained accurate and actionable data, which offered us a clearer understanding of Bolt’s performance in the airport ride market.

The updated tracking system confirmed our previous concerns regarding Bolt’s performance in the airport ride market. Our percentage of airport rides was significantly behind our competitors, and some markets showed lower finish rates than expected.

Further analysis showed that demand for airport rides came in waves, resulting in an imbalance of supply and demand at times. Additionally, we observed many deadheading trips, where drivers dropped off Bolt riders but returned to the city empty-handed.

Identifying opportunity areas and building a roadmap

After obtaining reliable data, our focus shifted toward identifying opportunity areas. As a Product Manager at Bolt, I liaised with various operations-level stakeholders, particularly Regional and Country Managers, as they had invaluable knowledge of problem areas on the ground.

I conversed with our local market teams to understand user problems and pain points from both the rider and driver sides. At the same time, I worked with the analytics team to validate the identified pain points.

Based on the information gathered, we grouped the pain points into 3 distinct pillars: poor airport experience, inefficient matching, and low brand awareness.

From these pillars, the team’s mission was to “Communicate and provide riders with a hassle-free and safe airport transportation experience and become the default airport transport option.”

While each pillar had its success metrics, the 2 north star metrics for the airport domain were the finish order rate (% of trips requested by riders that result in a finished ride) and the percentage of ride-hailing revenue generated from airports.

To achieve this goal, the team developed an extensive feature roadmap prioritised by the estimated impact and development effort required. I also worked to align legal and marketing stakeholders so we could build and market our solutions effectively.

Ultimately, we settled on four core features that would address rider and driver pain points and drive the most impact while moving to the backlog the less impactful or lower confidence items.

These features were:

• Airport Directions
• First-In-First-Out (FIFO)
• Rematch
• Airport Notifications.

Poor airport experience

As Product Managers working in an office environment, we may be prone to bias. To avoid this, it’s essential to interact with the product experience as closely as possible, especially during the discovery phase.

I first headed to London Heathrow to gain a first-hand understanding of the arrivals’ experience and identify areas for improvement. Over a few days, it quickly became apparent that there were significant opportunities to improve the rider experience.

The leading rider pain points were:

• Incorrect pickup points.
• Absence of pick-up area descriptions.
• Lack of walking directions.

To fix the incorrect pickup points, the Geo team conducted extensive due diligence of all airport points of interest (POI) to ensure accurate pick-up and suggested drop-off points. They marked (and rapidly fixed!) faulty areas and noted airports lacking curbside pick-up.

Pick-up area description

We also realised that the absence of pick-up area descriptions was a significant issue for our riders and drivers, particularly in airports where kerbside pickup was unavailable.

To address this, we implemented a short-hand description of the pick-up area that would always display by default within the pickup confirmation screen and when the rider is waiting for their driver to arrive for airport rides.

We also ensured that the short-hand description was visible to drivers, providing them with clear guidance on where to meet the rider for the pick-up. This helped to ensure that drivers could quickly and easily locate the pick-up area, minimising the potential for delays or confusion.

Walking directions

Due to the limitations of many users’ smartphones with poor GPS capabilities, we recognised that the proper solution was to use static walking directions based on fixed points. Building these walking directions required a specific set of data points.

To help achieve this, I created a standard operating procedure (SOP) that other operations teams could use to gather on-the-ground data.

To help increase the feature’s discoverability, we, by default, always displayed the walking directions banner for all airport rides.

When the rider tapped the banner, they would see the starting and pick-up points, step-by-step directions, and a walking path needed to reach the pick-up area.

Riders could also access these directions before ordering in the Confirm pick-up screen.

We initially focused our efforts on major airports within each market that generated significant airport revenue to maximise impact. However, we also recognised the need to future-proof our solution.

We tasked the Geo team with building an internal system that allowed our local teams to make directions for any custom pick-up area, including railway stations, shopping malls, and other high-traffic venues.

Inefficient matching

The airport dispatching system needed fixing because it primarily relied on the distance to the rider and the driver rating to make allocations, disregarding the time drivers spent waiting in the airport area.

As a result, it was possible for a driver who had arrived only a few minutes ago to get a request ahead of one who had been waiting for nearly an hour.

Some drivers resorted to walking up to the terminal to gain an advantage over other drivers. In specific markets, drivers were selectively choosing rides, leading to some riders being unable to secure a ride and sometimes inadvertently collaborating to drive up surge pricing.

In other instances, drivers in some markets were asking riders to cancel their orders and instead conduct an off-market cash ride, which put the safety of our riders at risk.

FIFO & Rematch

To ensure enough supply of drivers, we worked with both the dispatching and driver teams to create a fair and impartial queuing system at airports. Our solution was a First-In-First-Out (FIFO) system, meaning drivers were assigned a waiting position based on when they arrived at the designated waiting area.

If a driver chose to leave the queue or declined an order, they would lose their position after a certain grace period.

On the user experience side, we aimed to provide transparency on the number of drivers waiting in the queue and the designated waiting areas’ boundaries.

We did this to ensure we complied with local regulations — it also enabled drivers to make informed decisions about whether to head to the airport or stay in the city.

To incentivise drivers to accept airport requests and maintain a steady supply of drivers, we implemented a feature called Rematch.

This feature prioritises drivers dropping off passengers near the airport, placing them at the front of the matching queue. As a result, they can quickly receive another passenger and benefit from a steady stream of customers.

By only activating Rematch when the driver is nearby dropping off a passenger, we can achieve lower estimated arrival times and higher rates of completed rides, providing a better experience for both drivers and riders.

Low brand awareness

One of the primary challenges we faced was low brand recall among our riders when travelling abroad.

Our internal data supported this, showing that our market share among riders travelling outside their home market was significantly lower than that of our competitors. External surveys highlighted a lack of awareness about Bolt’s global presence.

Simply put, when riders visited a new location, they typically opted for a competitor they were more familiar with rather than Bolt.

To address this issue, we took a two-pronged approach.

Firstly, we collaborated with our Marketing team to develop out-of-home campaigns showcasing our improved airport pickup experience. We placed these campaigns strategically in high-traffic areas of the airport to capture the attention of existing and potential Bolt riders.

Geofenced notifications

Secondly, we created a geofenced notification system that automatically sent push notifications tailored to specific markets when we detected an out-of-market rider arriving at designated airports.

Thanks to the recency effect, these notifications helped re-engage riders with our brand and place Bolt back into their decision-making process.

Since a traveller’s first ride can influence their decision-making process regarding ride-hailing providers, we found that providing a positive airport experience increased brand loyalty and revenue for Bolt. We noticed a significant secondary lift in revenue as a result of this approach.

Conclusion

The product roadmap has helped accelerate and support Bolt’s double-digit growth rate in airport market share.

The same roadmap has also delivered vital wins resulting in higher finish order rates for airport rides and decreased rider cancellation and no-show rates. It has garnered positive feedback from our drivers.

During this project spanning several quarters, I had the privilege of collaborating with senior stakeholders from the Operations, Management, Legal, and Product teams.

This experience was invaluable for personal and professional growth, as it allowed me to build strong relationships with these stakeholders, whom I can now turn to for guidance.

Join us!

My focus has shifted towards enhancing the operational order experience, improving retention, and developing a package experience, but improving the airport experience still needs to be completed.

Various teams are actively exploring several product enhancements, including integrating scheduled rides specific to airports, expanding the FIFO system to new hubs, and more.

If you want to help build something that your friends and family will be using and impact the commuting habits of millions of Bolt users, join the team! For example, we’re looking for a Senior Data Analyst to help us discover more exciting opportunities. If that sounds like you, we’d love to hear from you.