Improving the order matching for both drivers and shippers
This enhancement enables logistics drivers to pick orders accurately,
thereby shippers can have their goods transported in a more effective way.
This enhancement enables logistics drivers to pick orders accurately,
thereby shippers can have their goods transported in a more effective way.
GOGOX revolutionised the logistic industry by seamlessly connecting shippers with freelance drivers, ensuring efficient and reliable delivery services. In this project, we enhanced their shipper and drivers matching process, with the goal of elevating user satisfaction and fostering user retention by addressing common complaints.
Product Manager x 1
Product Designer x 1 (Me)
Data Scientist x 1
Frontend Engineer x 2
Backend Engineer x 2
Conducted researches
Utilising design techniques e.g. flow design, prototyping, mockups, for idea validation
Revised UI to enhance order picking experiences
Redesigned the matching flow
Initiated Design System usage in the team
Looking at the platform performance, we discovered that there are more drivers dropping orders and not all shippers successfully got matched with a driver.
20% of orders dropped by driver
Shippers' orders being abandoned by one or multiple drivers out of every 5 orders.
85% orders completion rate
A decline in shippers' reliance when traditional companies offer 100% guarantee
I dig dived into data to see what are the reasons drivers decided dropping an order.
The most selected reasons selected by the drivers are:
Drivers are unable to fulfill shipper's requirement
Failed to agree the order quotation with shippers
Drivers mistakenly picked the order they don't really want
We hypothesised that the low order matching success rate is attributed to unclear shipper requirements, making it challenging for drivers to evaluate their ability to fulfil the job.
By encouraging shippers to provide comprehensive details should anticipate a reduction in order drops and improved matching outcomes.
We went for some low effort design approach to boost the clarity of shipper requirements,
Quick Solution
Added more fields such as "Notes to drivers", "Tunnel preferences"
➜
Outcome
No improvement on the order drop rate and completion rate
To better investigate the reason of high drop rate, I start initiating user interviews by shadowing drivers and go through their daily routine and jobs.
I conducted shadowing sessions with 6 drivers, closely observing their routine tasks. The observations yielded surprising insights that surpassed our team's initial expectations, highlighting the significance of addressing this issue. 😮
Short clips of recording how drivers pick an order (actual speed without adjustment)
💔 During the shadowing sessions, a notable observation was that drivers tended to click through orders robotically without thoroughly reading the order details. This behaviour raised concerns and emphasised the need for improving the matching process.
Quotes from interviews with drivers
Based on our conversation, drivers are not reading the order details because
Drivers are compelled to skip the confirmation step as it isn't yet officially confirmed. This urgency arises from the risk of losing opportunity to earn money.
The avg. response time is a mere 2.5 second from a shipper placing an order to a driver confirming. This makes it practically impossible to read thoroughly and confirm the details
With a 1:5 ratio of orders to drivers, the competition for orders is intense. Drivers tend to prioritise picking orders first and making decisions later.
We reviewed the legacy flow and framed the below 2 problems :
1. Insufficient time to read order details
2. Validate order availability only upon user clicks, lack of real-time status
Provide a sense of secure and stable environment to allow drivers select orders should be considered to make sure the experience can be seamless and efficient.
Here’s the walkthrough of our revamped flow
In the original design, real-time order status was not reflected in the app upon clicking buttons. This caused drivers to fear losing an order and often led them to hastily press the "take order" button without confirming the details.
To enhance the order picking experience, in the new design, we added real-time order status updates within the UI. We clearly indicate when an order has been picked and promptly remove expired orders, ensuring a more efficient order picking process.
With the legacy flow, drivers are forced to click robotically because if they don't do so,
they will have a high chance losing the order in "Order details" page.
In our new design, we added a step that the order will be locked and only visible by the driver. This design aims to to reduce the chance that an order will be "stolen" by others.
With the implementation of this new logic,
Likelihood of experiencing orders being picked by others has significantly decreased by 66.7%
Chance of "Order being stolen" reduced from occurring 3 times to max. once only
-10% order drop rate
+5% order completion rate
From the below charts,
it shows how the new design improves the platform's performances
* Drop rate definition — Drivers dropped an order after confirm picking it
* Completion rate definition — Percentage of orders that have been fulfilled successfully
