Telstra — Smart Care
Reducing drop rate and Enhancing User Experience of Smart Care Flows
This case study delves into the remarkable journey of MyTelstra Smart Care, a digital solution designed to revolutionise the customer experience. Telstra leveraged innovative technologies and streamlined processes to create a seamless self-diagnostic system.
Problem
The issue arises when customers abandon the Smart Care flow and instead contact an agent for support. This frequently happens when customers are required to complete multiple steps in the Smart Care flow. It not only occupies agents with issues that customers could resolve through self-service but also leads to longer queues at the call center, affecting other customers.
Goals
Reduce the number of customers who do not complete the Smart Care flow by leveraging automated diagnostic tools to minimize user actions.
Re-engage with customers who drop off the Smart Care flow to decrease the number of customers contacting agents for troubleshooting their fixed service issues.
Overview
To deliver a product that meets the specific needs of Telstra NBN customers, it was crucial to develop a deeper understanding of their requirements. By gathering intel, data, and insights internally from customer data centers and experts, we gained a holistic view of customer use cases, their NBN journey, and the features and services they utilize. This allowed us to prioritize customer-centric designs and solutions that solved their problems rather than focusing solely on business needs. By coupling customer data with iterative user testing, we continuously refined our designs to cater to the specific needs of our customers. This approach empowered us to confidently present solutions to senior stakeholders and facilitate informed decision-making.
MyTelstra Agile Squads
Within MyTelstra, multiple Agile squads were dedicated to launching features and functionalities in the app. Ensuring collaboration and information sharing across squads was essential to maintain consistency and avoid redundancy. I played a pivotal role in championing design operations across the design teams, ensuring the systematic gathering of past research, backend functionality, and solutions.
Customer Journey Analysis
Using metrics from Adobe Analytics, we identified the most visited flows within the Smart Care experience. It was observed that users tend to drop off at specific stages:
During the initial two steps when they are asked to select the affected service and describe the symptoms.
At any stage of the journey when they are required to perform actions on their end. To ensure this behavior wasn't due to technical issues, granular data analysis was conducted.
60%
of the users exit the page when landing on the "Service Selection” screen
User Interviews
Following the initial data analysis, we invited six users to navigate the Smart Care flow. The interviews revealed the following insights:
Customers were confused by the labeling of symptoms.
Participants expressed surprise at being asked to select symptoms and felt unsure about which option to choose.
Most participants expected the provider to have a basic understanding of the issue they were experiencing.
Many users were comfortable following a step-by-step process but preferred to avoid lengthy flows with more than five or seven steps.
Proto-Personas
Insights gathered from data analysis, interviews, and existing research were used to create proto-persona profiles and scenarios. Four proto-personas were developed based on goal and expectation affinity. A profiling spectrum was created to highlight the characteristics of each group.
It was determined that two of the proto-personas, named "Top" and "High," were highly likely to own a Smart Modem. These customers prioritize long-term trust and high-end service and benefit from the Smart Modem's ability to connect via the 4G network during NBN connection issues. On the other hand, the "Basics" and "Practical" proto-personas were less likely to own a Smart Modem. They experience more outages due to their location and are more inclined to switch providers for cost-saving reasons.
Each persona had a distinct experience based on their behaviors, expectations, and affinity with the product. Scenarios were created to explore potential opportunities, ideas, and questions.
Improved Experience
Based on the gathered insights, a new customer journey was designed and tested. The following improvements were implemented:
Users no longer need to select the service from a list in the "Get Help" portal or app. Instead, they can start a service diagnosis directly from the service screen in the app. This aligns better with user mental models, as they naturally navigate to the NBN service screen to find diagnostic tools.
Once users report an issue with a service, an automatic diagnostic tool runs several tests in the background and notifies the user when the results are available.
Depending on the diagnosis result, users may be prompted to take specific actions tailored to the problem, reducing unnecessary steps.
Enhanced Customer Communication
To alleviate the stress and frustration customers experience while waiting for network restoration, personalized messages have been added. Specific enhancements include:
Customers with a Smart Modem receive advice on utilizing the 4G backup network and guidance on checking if the device is correctly installed, presented in two simple steps.
Customers with a mobile plan from Telstra are encouraged to use their mobile device as a network backup and informed they will receive extra data.
A dedicated section in the app is now available for scheduling and inspecting outages. Users can opt-in to receive notifications related to outages in their area.
Feasibility
On the development side, it was crucial to understand the backend systems to have understood what we can currently deliver and what we can push to deliver in future releases.
Understanding the backend constraints and maintaining consistent and open communication with developers and solution engineers ensured the team were able to deliver customer-centric solutions in tight timeframes.
NEXT PROJECT