Design for Trust

SOLVING FOR TRUST

In the trucking Industry

Indian Trucking Landscape

India’s logistics sector is ~14% of the overall GDP. With a market share of $150B, the trucking industry has 2 key players- Truckers and Shippers united by a common goal of moving loads (shipment) from one place to another

Next Billion Users

We are building the largest freight network in India

Target user base for this project

BlackBuck is an online marketplace that matches a shipper with a trucker. Think Uber for trucks where we match shippers (demand) with the right truck owners (supply)

Online Marketplaces

In an online marketplace like Amazon, Flipkart has an average transaction value of Rs.2500, whereas freight shipping it's around Rs.50,000. The actual value of goods transported is 50 times this amount up to 20 lacs.

Our Marketplace Approach

We'll be discussing this aspect for this project

For the right matchmaking between shippers and truckers, we require data points like

So how can we match the right trucker to a shipper?

Boss App

Solving for Trust

With such high costs involved, would you trust any random stranger on the platform? It’s no wonder that freight movement is mostly done by family-run businesses, with demand brokers and truck drivers being a part of their extended circle of family and friends.

These micro-networks are spread throughout the country and shippers and truck owners rely on them for their business. These networks tend to be inefficient due to demand-supply mismatches across seasons and locations. But the lack of trust between fleet owners and shippers makes them take the same inefficient options. It was clear to us what our problem statement was

" How might we make the truck owners trust and make them explore new connections? "

Before getting any further, please familiarize with these terms which are going to be used in the case study moving ahead

What does trust means to a truck owner?

We visited trucks owners to get a sense of their day to day lives and understand what makes them trust a transporter (mediator between shippers and truck owners) and also understand goals and pain points

INSIGHTS SUMMARIZED

True judgment of a transporter can only be made after working with the transporter 1-2 times after getting assurance of the transporter’s payment behavior

Document and background verification only helps to make the decision of 1st booking with a new transporter easier

Bad payment experience (late payments) with a transporter is a huge red flag for Fleet Owner to not to continue working with the transporter

What other FOs talk about a new transporter is crucial in deciding to try the transporter in the first place

With all the other data points from the contextual inquiry, we were able to form a Persona of a truck owner whom we like to call Mohanlal

Screen Shot 2021-03-27 at 7.15.49 PM.png

With all this information, it was easy for us to create a journey map of load booking experience

With the preliminary research, we understood the key actors who play role in the truck/fleet owner's journey which were Partner, Load, Lane and Platform

Diving Deeper

We needed to get an in-depth understanding of trust factors within each actor and understand the decision-making factors of the truck owners. We decided to individually call up Truck Owners and interview them

In-depth questionnaire

Insight Mapping

With all the insights mapped and grouped, we came up with a Trust Framework

Transporter Profile Design

The first version

Where is it currently

How these features leverage Trust through
Digital Reputation Platform

Transporter Profile Construction

1. Goal

The goal of Transporter Profile is to Instill Trust in Transporters and BlackBuck. And this can be done by answering some key questions

2. Understanding Best Practices

Secondary research to understand profile pages best practices

First Design

How did it perform?

Insights

The downstream conversion was extremely poor. Out of 8719 users who visited SRP (entry point to profile page), only 29 users ever pressed the call button, hence poor conversion

Gathering Feedback

I went out to test our first profile page with the users in the outskirts of Bangalore

Some of our users did not understand what a prototype is and simply refused to test it

Insights

"The key takeaway was that users were still unclear about the value prop of the profile page and still did not understand how to judge the transporter that will help them make a better decision"

and the research continues...

Listening to Trucker - Transporter calls

We partnered with Knowlarity who help us listen to phone calls that were made from our app. We wanted to understand how these two parties talk to one another

Actual call recording of a conversation between a Truck Owner and a transporter

Mapping insights into sheets for every calls listened

Insights

Transporter did not have the requirement for the right truck type
The load was already booked
Transporter did not pick up phone / Call did not go through
The transporter did not talk properly

We further called up these truckers to get more understanding of their interactions with the transporters and get a deeper understanding

This helped us bucket Transporter Traits into -

This activity was done by a team of three including a UX writer, a Product Manager, and me. We listened to at least 10 to 20 calls everyday and it took us less than a month to complete the research.

Reconstructing Rating & Reviews

When to Capturing Ratings

We understood from the user research that a lot of talking on the phone happens before a deal is negotiated and truckers usually call a lot of transporters to get the right load, so it was evident that we should capture the Call Experience

We also wanted to capture the fulfillment experience after a deal has been done let's call it Business Experience. The caveat here is that we do not know whether both the parties worked together or not. The soft signal for conversion the call button

Load Booking Journey

Where to Capturing Ratings

After Call- Caller ID

Capture

Caller ID Experience

Caller ID Design

After Business

Transporter Profile

Call Logs

Homescreen

What to capture?

We want to capture the good and bad transporter traits we got from the research earlier

How to capture Ratings

After Call

Information Architechture

I created a rating capture screen with Transporter Context- Image & Name. The Middle area being the primary capture area and the bottom portion consists of a written review and CTA to submit the overall feedback.

Usability Testing

I created 5 prototypes of how feedback capture and went out in Bangalore outskirts to test them out with the users. Users were selected based on their availability as I accompanied a CRO. The designs were tested with 8 users

Iterations & User Feedback

Users took a lot of time to read through the entire section and give their responses because all the questions are written differently.
A lot of users had trouble understanding the questions and options of the first question

Users found this approach to be the most confusing as they did not read the section headings and directly started tapping on the options

Most of the users had neutral feedback about this version.
Some users like it as they mentioned "hasta hua dikh raha hai" ("I can see it smiling")
Some of the users did not understood what the smiley meant and were asking questions about it

The winning option as it was understood by all the users (not the very tech-savvy ones).
Users were easily able to read and answer questions as they were binary

At first glance, most of the users thought it was a display of ratings rather than a rating capture
Apart from that, it did not look that it can be interacted with

Capturing Overall Rating

We needed an entry point to capture and display the overall rating. Although we wanted to capture star ratings, some of us felt a binary method (thumbs up/down). So

We called up 12 users to understand

If they left ratings on other platforms
If they understood binary ratings better than 5-point scale ratings
If they describe transporters on a binary scale (black/white) with little to no gray areas

Insights

None of the users have given a rating on any other platform
8/12 users understood the 5-star rating scale.
The remaining users thought 1 Star means the best rating (commented: Ek number)
They did describe the transporter as very good or extremely bad

The Decision

After a lot of brainstorming, we decided to with the 5-point scale rating system, despite the binary system being the preferred choice of users, why?

Why we went with the 5 point scale ratings?

Binary ratings will negatively affect conversion if there are a lot of badly rated transporters. Users will still call an average rated transporter over a bad rated one

5 point scale system is flexible as we can modify the threshold for every star based on how stringent we want the rating process to be

The problem with the design

During the user test, we found out despite giving a four-star rating, users chose the bad traits in the subsequent screen. This resulted in dissonance when those ratings were displayed on the profile.

Rating Input Problem

Users were giving good overall ratings and selecting bad for all the traits

Ratings Output Problem

Having a good overall rating but bad traits was confusing

Solution

By changing binary input to a multi-select and giving rating based selections we were able to prevent the dissonance between the overall rating and the selected traits

Original

Solution

Options for bad overall rating

Options for good overall rating

Final Call Rating Experience

Capturing Business Ratings

Business rating entails how was the transporter experience during fulfillment. It also entails if the transporter paid the money on time

Since we do not have an explicit signal if a trucker has been in a business with a transporter (future scope). We assume that if the trucker is visiting the transporter profile then he must have done business with him

We wanted to make sure that we only capture only genuine business experience. So we added a step in between to confirm if the trucker and transporter actually went into business

First Design

Problems

Explicit business deal capture was leading to extremely high drop-offs.
Users not understanding Payment and Service Experience questions

Solution

We Removed the explicit business deal capture screen and added a button for it on the overall business experience capture.

Changed the previous leading question to a single word and Good / Bad selection

Showing Ratings

When and where?

There are 4 areas in the load booking journey that we show the overall rating of the transporter

Showing Ratings on Transporter Profile

I created six different versions of rating visualization on the transporter profile and tested with our power users on Whatsapp and also internally

Feedback

Most of the users had trouble understanding the ratings and they took a lot of time to grasp the overall concept. Overall, the designs did not workout for them