UX Research: Decreasing cancellation rate and retaining customer loyalty.

UX Research and strategy

Elena Astakhova
UX designer
3 min read
What is GoJek
In the bustling landscape of modern urban living, Gojek emerges as a transformative force, offering a myriad of services designed to simplify and enhance the daily lives of its users. With a strategic focus on product diversification, Gojek has carved its niche in the market through a blue ocean business strategy, enabling it to compete on new frontiers. Customers are drawn to Gojek not only for its convenience but also for its embodiment of local culture, versatility, exceptional driver support, and an intangible sense of community – attributes that collectively imbue the brand with a distinctive 'soul'.
However, amidst its array of offerings, one core service stands out as the cornerstone of Gojek's revenue stream – its taxi service. Yet, despite the convenience it offers, users like myself have encountered persistent challenges that detract from the overall experience. Long waiting times, coupled with communication hurdles and uncertainties regarding driver reliability, have emerged as recurring pain points, prompting a deeper exploration into the user experience.
Despite a wide variety of provided services, taxi service remains the main revenue maker. It also helps to get new users - a taxi is the first thing tourists need when they arrive in the country.


Find out what causes drivers to cancel short trips.

Develop and validate hypothesis solutions

Project duration: 1 month

My role: Lead UX designer, researcher

Problem Statement
Ordering a scooter taxi could take 30+ minutes:
drivers cancel trips / don't move/drive away.
I talked to 26 GoJek drivers.

Budget: coffee and croissant per person.

Impact on the organization

  • losing users: every second review on Google Play mentions long waiting time and trip cancellations.
  • revenue loss: reduced user engagement leads to fewer transactions and decreased revenue from ride bookings
  • competitive disadvantage: In a competitive market landscape, the inability to address user and driver concerns puts the organization at a distinct disadvantage against Grab.
  • from the field research I have conducted, I have found that losing drivers: 86% of ex-GoJek drivers (now Grab drivers) I have interviews said that they have changed companies because of long trips.
The human brain processes images 60,000 times faster than text, this is why storyboards are so handy. It comes as a visual sequence that outlines a user's journey with a context and user's feelings. It consists of illustrated frames depicting key interactions and screens from the user's perspective.


Observe and Interview users in their environment.

As I observed GoJek scooter taxi drivers navigating the bustling Indonesian traffic, it became clear that their attention is constantly in flux due to the chaotic environment and physical demands of driving. Distractions like conversations with peers and family video calls are common, posing safety risks and impacting service quality. To address this, systems like Autobit incentivize drivers with performance metrics, helping them stay focused.

Research questions:

I have applied a funnel technique, starting from the broad questions, like "How did you become a driver", to more detailed: "How to cancel the trip in a way that will save your points?"

Top findings
I have found that all uneasy drivers' behavior is caused by dissatisfaction with a certain type of trip: short trips with a far pickup location. Drivers employ a few techniques to avoid these trips. They worry about their rating score, so they have a few techniques that help them get rid of the orders they don't like.

1. Drivers cancel trips because they are afraid passengers will do the same. When a driver sees a short trip with a far pick-up location, they prefer to cancel it first to save time. They are trying to finish as many trips as they can in a short period of time.

2. Drivers employ a strategy of subtly moving away from the passenger. They sabotage trips they are not happy with by moving in the opposite direction. It causes automatic trip cancellation, or passengers cancel it.

3. Drivers don't move to make a passenger cancel the trip if they think that they spend more money on petrol; they believe this trip isn't worth it.

4. Drivers don't move because they don't check the app. They left "Autobit" on and got distracted. This happened more often between 7 am and 10 am.

Solving problem: HMW wrokshop

Time: 1 h
Space: Zoom + Miro
Goal: turn problems into creative solutions.

Participants: 2 developers, 2 designers, 1 Product owner

HMW statement: How might we make economically unfeasible trips more attractive so drivers don't cancel the trip?

Workshop outcomes

The problem needs to be addressed to both drivers and passengers.

For the drivers:

  • remind drivers that these short trips are also valuable for the system and help them to earn rewards and money
  • add more points/rewards for long pick-up trips
  • offer a reward (point/money)

For passengers:

  • update passengers with the order progress
  • offed discount/points for long waiting time
  • encourage them not to cancel the trip

(!) Improve the Autobit system (needs more exploration)

UX Design Recommendations:
(A problem: ordering a scooter taxi could take 30+ minutes: drivers and passengers cancel trips with far pickup locations.)
  • 1
    Improve Driver Engagement
    • Implement visual reminders in the driver app emphasising the value of short trips to the system. Use motivational messages highlighting rewards and earnings from these trips.
    • Increase incentives for drivers undertaking longer pick-up journeys. Display potential rewards prominently within the app.
  • 2
    Improve Passenger Experience
    • Integrate the animation with real-time updates. For example, the animation could show a character moving towards a designated point on a map, reflecting the progress of the assigned driver.
    • Design a playful and informative animation to entertain passengers while they wait for a taxi. This could include a dynamic progress bar with interactive elements or a cute character representing the waiting process.
    • Use the animation to provide visual feedback on the impact of waiting. For instance, the animation could change based on the length of the wait, offering rewards or encouraging messages for patience.
    • Offer discounts or loyalty points to passengers experiencing extended waiting times, encouraging them to remain patient rather than canceling.
  • 3
    Implement other System Improvement
    • Explore enhancements to the Autobit system for more effective matching and trip allocation. Conduct user research to identify pain points and areas for improvement.