The problem
- Help center on the platform, back in 2017, was very static.
- Not contextual in nature to the customer's complaint. Same set of queries were presented for both active/delivered orders, which meant that customers had to go through the effort of identifying the correct issue and describe in detail.
- No real-time information on the order was given to customers to ease anxiety
- Customers often resorted to clicking on help to find order status
- For any issue, customers had to chat with/talk to an agent. This takes around 1-2 mins for an agent to respond, and 3-5 mins to identify and resolve the situation. The longer the interaction between the customer and the agent, the more the cost per call to Swiggy.
- There were many issues which didn't require so much manual effort but still went through this, and Swiggy was bleeding money on all calls by customers for the smallest of issues.
Top 5 customer issues (across active and delivered orders)
Project brief
- Identify the gaps in the help & support experience for customers to get contextual assistance for their queries. This includes heuristic evaluation and talking to customer care agents to diagnose the biggest causes of customer calls.
- Build a chat-bot experience for the key issues, to minimize human interaction and quickly resolve issues.
Roles & responsibilities
At the time, I was managing a design team on projects across other platforms, but I took full ownership of the Help center flows and designed the end-to-end process with a lot of support from the Product Manager on detailing the different issues and ways to resolve them.
I also worked with my peer, who was involved in the agent workbench designs on the customer care side, to ensure smoother handshake between the new help framework on the customer app and the corresponding updates on the call center agent's dashboard.
The work duration was from August to November 2017, and the final designs were shipped in Dec 2017. We rolled out the chat bot experience to all customers Pan India on Feb 2018.
Design process