ANALYSIS. STRATEGY. RESEARCH. DESIGN. INTERFACE DESIGN. ANALYTICS.
Scenario
Problem
When a customer is trying to complete a transaction and runs into difficulty, they don’t want to waste time scrolling through FAQ pages. They want an answer immediately.
As a consequence, the customer experience radically improves, and satisfaction levels skyrocket. Customers appreciate prompt answers to minor problems so they can carry on with their day. Happier customers are more loyal customers and boost the brand’s image overall.
For everyday banking operations, chatbots are the ideal first port of call. If a customer encounters an issue when trying to pay a bill or transferring money, they need a solution there and then. They don’t want to wait around on the end of a phone line for a solution. It’s a low-stakes problem that chatbots can answer just as well as advisors can.
This eases the workload of advisors in financial organizations who can focus on problems that require more sensitivity or expert knowledge.
Goal /my task/
Determine the optimal approach and interaction model to set user exceptions, and ease user frustration by solving the user’s problem or answer queries through the bot. If the bot cannot solve or answer a problem, it will refer the customer to the relevant support agent. The agents can focus on giving more complex advice and less on answering repeat or simple queries.
My Design Process /approach/
A collaborative, user-centric iterative design process was used to guide distribution from kick-off, and even through development and lunch. Team calibration and collaborative design thinking were critical to both idea generation and also facilitating the core team’s alignment and decision-making. User data including research, interviews, and testing, providing invaluable input and ongoing feedback that helped us craft, then iterate an experience that not only delivered on core business needs, but also satisfied users.
Phase1:
Discovery & Planning
As a first step, it was critical to gather internal and external data that would help frame the problem and understand both sides and factors influencing the challenge ahead.
Research /discovery/
Let’s start with the basics. I think sometimes it’s important to make a step back for a short while before diving into more complex matters. In this case it really helped us. Believe it or not, but reading through all those fundamental definitions opened our eyes on to a few creative solutions and boosted the entire ideation process.
Conversational framework
Interviews and conversation
Stakeholder interviews and impromptu chats were facilitated with core team members to get a better sense of individual stakeholders' concerns, their understanding, and attitudes towards the project.
Takeaways
Having this discussion with stakeholders and the core team early in the process allowed us to reach a consensus and a shared understanding of why we were constraining our MVP launch to a Watson Chatbot with IBM.
Phase 2:
Ideation & Synthesis
Design thinking techniques were used in various ways capacities to generate a feasible ideas.
Workshop /ideation/
I have facilitated workshop sessions as a brainstorming exercise with the product team, as well as key stakeholders. The goal of the sessions was to reinforce alignment with the core team on the goals of the projects, as well as the problem I was solving, to discuss existing research and assumptions.
The outcome of this session was two-fold. Firstly, it helped to drive further team alignment and a feeling of inclusion and alignment across the entire core team. Secondly, the information and insights produced during this session were synthesized and used for further idea creation and discussion.
Further outcome - ​Use cases Prediction for the chatbot
-
Getting a quick answer in an emergency
-
Resolving a complaint or problem
-
Getting detailed answer or explanations
-
Finding a human customer services assistant
-
more
Messaging system flow
Chat conversation - prototyping & usability testing
Chatbot Analytics: Essential Metrics & KPIs to Measure Bot Success
Bot Metrics
Retention Rate: this is the percentage of users that return to using the chatbot in the given time frame. This important because we need to keep the customer engaged to extract valuable insight into their preferences by making them spend time on the chatbot.
Intent analysis
Intent analysis is a step ahead of sentiment analysis. While the latter is a common text classification tool that analyses an incoming message and tells whether the underlying sentiment is positive, negative, or neutral, intent analysis tells you all about the user’s intention behind the message.
Goal Completion Rate (GCR)
This captures the percentage of successful engagement through the chatbot.