Snapp! Chat Module Feature

Chat Header Picture

Introduction

Snapp! is known as the Uber of Iran with more than 2 million rides per day and 30 million users. The company is the largest and fastest growing internet company in the Middle East.
The chat module was released in August 2020 for the latest version of Snapp. It enables passengers to chat directly with drivers. In the design, as shown below, I have placed it beside the Call button. The reason behind it was that both of them are for contacting the driver so they should be next to each other.

Old Chat
The first version of Chat Module in Snapp

My Role

By selecting two segments of our users, I developed a questionnaire that our Call Center would ask of each group. I revised and modified the designs based on feedback and insights.
In addition, I worked alongside a UX Writer, 2 Product Managers, Client Developers, Data Scientists, and our Call Center members.

Problem Statement

After rolling out the feature, the analysis showed that the usage of chat in the Passenger app was about 4%. While on the Driver side it was 8%. It was evident that drivers are using chat more than passengers, and passengers may not even notice the messages, let alone reply.

Problem Icon

Solution

We had a session with the respective PM and decided to run a survey to find the reason why passengers are not using Chat as much as we expected. We also agreed to increase awareness by running marketing campaigns.

For the survey, first I gathered all of our hypotheses for the low usage of this feature. Then I chose two segments of our users based on whether Chat was available for them and if they have used Chat or not. For each segment, 40 users were asked to take part in answering a questionnaire. Users were segmented based on the following criteria:

  • Those who normally take more than 20 rides per month and Chat was enabled for them and they have used it for at least 1 time.

  • Those who normally take more than 20 rides per month and Chat was enabled for them and they never used it.

Lastly, I made two questionnaires and handed them off to our Call Center (CC) team so they can start calling those users and ask them the questions.

Key Results of the Survey

  1. Most of the eligible users didn’t trust that their messages will be received
  2. Most of the eligible users always appeared at the origin earlier so they didn’t need to call or message the driver
  3. Some of the eligible users use to call the driver for telling the right location or don’t cancel and other things that are difficult to say through chat
  4. Some of them didn’t know about the feature and this could be fixed with SMS or showing a coach mark
  5. Some of the eligible users saw the chat button but didn’t use it, because of being in a hurry

Exploration

The most important thing was to make chat more visible to our users. Here are some iterations for doing that:

Problem Icon

Designs

After analyzing the results of the survey, the key actions we had to do became clear. I designed a solution for everything the users needed.

Making chat more visible
Read and delivered indicators
Sending Location

Results

After running campaigns and increasing awareness on the passenger side, the usage went up. The number of calls between passengers and drivers was also reduced by a good number.

Sending Location
Passenger call rate