GOGOVAN launched a new crowd-sourced same-day delivery service in 2018 summer. Originally, the service targetted small-medium businesses with ad-hoc urgent delivery needs in Hong Kong. I was part of an ambitious project to find the product-market fit for GOGO delivery.

The challenge

To find the product-market fit for GOGO delivery, our goals were to:

  1. Improve order completion rate (activation/retention)
  2. Increase activation/retention for on-demand pro segment (e-commerce merchants)
  3. Increase courier activation
GOGO delivery product roadmap
GOGO delivery product roadmap

My role

I led the design for the delivery experience since the product launch in 2018 summer and collaborated with one designer on the on-demand pro features and two other designers to support delivery on the new client app.

In addition, I worked alongside a product manager and data analyst. We conducted continuous research through user interviews and usability testing.

Reducing order cancellations

During the first week of launch, we noticed a high cancellation rate, where 80% of cancellations were due to shippers placing orders with oversized, overweight or multiple packages.

Process & experience

To understand the problem, I asked the experts from our operations and customer service team.

We found that new shippers assumed our couriers deliver packages using vans, but in reality, our couriers are walkers (they travel on foot and public transport). This led to oversized, overweight or multiple packages that were physically impossible for a courier to carry.

I tested the existing app with people unfamiliar with our service to dig deeper to the reasons why: most of them assumed packages were delivered by vans due to our brand “GOGOVAN” and that the order flow didn’t give them any idea of our courier’s mode of transport.

To tackle this problem, we ran a design sprint to come up with ideas and decide on what to test in our order flow prototype.

Sketches and voting stickers to explore and decide which ideas to test in our prototype
Sketches and voting stickers to explore and decide which ideas to test in our prototype
New service copy and icon to improve clarity
In the prototype, we renamed the service to "Courier Delivery" (搵速遞員) and redesigned the service icon to show a walker/biker to improve the clarity of the service on the home page.
Prototype to test service clarity
We also tested these ideas:
1. Package type options to show what kind of packages are acceptable for couriers to deliver
2. Helper text with a link to direct oversized delivery needs to our van service
3. An option for oversized packages that direct to our van service

After testing the prototype with users unfamiliar with our delivery service, we found that all testers understood the service and the majority expected either a walker or biker to pick up their package. When given a task to deliver larger items, they successfully switched to the van booking service.

The outcome

Due to scope vs outcome, we decided to implement the new service icon, helper text, oversize/overweight options and not add package type options to the order form.

One month after the release, cancellations due to oversized items by new shippers reduced by 37%. Overall cancellation rate dropped 17%.

Scheduled delivery and cut-off times

Through interviews with e-commerce and brick and mortar store visits, we learned that shippers like to consolidate orders and place them all at once at the same time. Some shippers preferred placing multiple orders in the evening for the next day without worrying about it the next morning due to cutoff times. Many shippers also found it difficult to understand the cut-off time for each drop-off option.

Process & experience

To improve the clarity of drop-off times, we worked closely with operations and marketing team to adjust the drop-off and cut-off times that is more meaningful for both shippers and couriers.

We made the times easier to understand and more memorable by:

  • Simplifying the same-day drop-off options to office-hours and after office-hours for residential deliveries
  • Displaying all drop-off options even when they’re not available (in a disabled state)
  • Adding helper text to make the cut-off times more transparent
  • Rounding times to the closest hour
Prototype to test the clarity of drop-off options and cut-off times

We tested the prototype with users and they found the helper text most useful in understanding the options. However, some users found the “available if…” helper text on available options confusing, so we decided to only show it when the option is disabled.

1. Added package type options to show what kind of packages are acceptable for couriers to deliver. 2. Helper text with a link to direct oversized delivery needs to our van service. 3. An option for oversized packages that direct to our van service.
1. Simplified the pick-up time selection for clarity
2. Showing unavailable drop-off options in a disabled state with helper text to remind shippers of the cut-off times
3. Date picker for scheduling deliveries up to six days in advance

The outcome

After shipping the scheduling feature, the percentage of scheduled orders increased by 270%. Scheduled orders make 20–30% of our daily orders.

Finding relevant orders for couriers

As our demand grew, couriers found it difficult to find relevant orders. Supply couldn’t keep up with the demand, and completion rate dropped as a result. To increase completion rate, we decided to tackle the problem of difficulty in picking orders for couriers.

Process & experience

Through focus groups, we learned that couriers pick orders based on distance from pick up location, drop off location and earnings.

I led a design sprint with cross-functional teams to identify the key problems to tackle and decide which ideas to test.

1. Added package type options to show what kind of packages are acceptable for couriers to deliver. 2. Helper text with a link to direct oversized delivery needs to our van service. 3. An option for oversized packages that direct to our van service.
Sketches from the design sprint to tackle couriers' problem of finding relevant orders
Early prototype to explore suggested bundling for couriers

We expolored the most popular idea: a visual map to show couriers available orders based on their location and suggesting bundling to increase efficiency and income.

Courier flow for filtering and sorting available orders
Courier flow for filtering and sorting available orders

We also explored using filters and sorting to help couriers find their optimal orders. We tested the prototype with couriers and made several iterations to suit their needs. After discussing with engineers, we decided to go ahead with this solution based on effort and outcome.

Lessons learned

Test product onboarding with new users unfamiliar with the service

The problem of shippers not understanding our service wasn’t identified during the closed beta period prior to launch because our sales team onboarded each user so they had a good understanding of the service. When new customers started using the service, the problems started to show up.

Listen to users and teammates who communicate frequently with clients on a regular basis.

Conducting frequent user interviews and sync-ups with our operations / customer service team gave us valuable insight into the problems we should prioritize and tackle.

Test designs and get feedback early

Testing early gave us a more concrete direction and prevent tunnel vision. It also helped cross-functional teams understand the product roadmap and priorities. There were many times our hypotheses were proven wrong and saved us time.