Profiting e-comm businesses with the help of human like interactive AI chatbot
Time line
Sep '21- Feb '22
Team
1 Designer (Myself), 3 Developers,
1 Business Analyst,
Buddhi is a chatbot envisioned to help reduce the workload of any e-commerce customer support team by handling the usual questions from the buyers. This bot is conversational and connects with other services to respond appropriately
my role
Sole designer of the project handled all the design needs right from, Analysing the opportunity, Set the design direction for the design & team, conducting user research, Drive ideation workshops and design showcases, collaborating with engineers, Manage design scope, Monitor design delivery
The business opportunity
Problem
BUSINESS GOALS
To design an MVP
To design a solution for this business opportunity and to turn the idea from 0-1 as a fully functionable cutomer support bot.
Keep it minimal yet scalable
This is being an MVP we didn't had enough scope for design, yet the business wanted to plan design in a way to scale this app for future versions.
Key flows to automate
To reduce support workload and boost efficiency, we prioritized automating the most frequent and repetitive user queries for maximum impact
Initial Stakeholder Alignment Session
Focus of the Session
Discuss the pain points faced by small to mid-sized e-commerce companies regarding customer support costs
Evaluate the feasibility of reducing support overhead via automation.
Understand stakeholder expectations for scalability, integrations, and go-to-market potential.
Present findings from preliminary market analysis and business research.
Key Discussion Points
Market gap
Most existing chatbots do not support real-time integration with inventory, shipping, or personalization tools.
Customer support overhead
Businesses with $10M in annual revenue are often forced to scale support teams as they grow, leading to significant cost burdens.
Scalable MVP approach
Stakeholders agreed on starting with a focused MVP that can later scale in features and integrations.
Automation as ROI
Stakeholders expressed strong interest in a solution that could automate 70–80% of routine queries creating clear, measurable value.
User interviews
Actual end Users interviewed
E-commerce Business Owners: Understanding their operational challenges, support cost concerns, and expectations from automation.
Customer Support Team Members: Exploring their daily workflows, repetitive tasks, and frustrations with current tools.
End Customers (Shoppers): Gathering insights on their common support queries, chatbot experiences, and expectations for conversational support.
Method
Remote interviews via video calls
Regions Covered for Interviews
USA & Canada
Duration
60-75
Minutes per participants
Participants involved in interviews
6
From all teams
User Personas
“If a bot can handle the routine stuff and escalate real issues to me, I can focus on what really matters.”
“I’m fine with chatbots if they’re helpful and polite, but if they keep repeating themselves, I give up.”
Key Insights from User Interviews
Repetitive Support Workloads Are a Major Pain Point
Support agents reported that 60–70% of their daily workload involved repetitive queries like order status, return policies, and promotions.
Business Owners Need Scalable, Cost-Effective Solutions
Owners want to avoid growing their support team as they scale. Automation is seen as a way to cut support costs and improve efficiency.
Customers Expect Fast, Context-Aware Assistance
End users are frustrated by generic chatbot responses. They want quick, accurate answers especially for personalized questions like “Where’s my order?” or “Can I change my subscription?”
Poor Escalation and Human Handover Frustrates All Users
When a bot fails to resolve an issue and there's no smooth handoff to a human agent, it results in customer dissatisfaction and lost trust.
Integration Is Not Optional It’s Essential
All user groups emphasized the need for the chatbot to integrate deeply with platforms like Shopify, order systems, and CRMs to be truly useful.
Competeitor Analysis
Key Insights from competietor analysis
Most Chatbots in Were Rule-Based and Shallow
The market was filled with basic FAQ bots lacking dynamic responses or contextual understanding.
High Setup Complexity and Cost Were Barriers for SMBs
Many tools, such as Intercom and Haptik, were designed for enterprise users, leaving small-to-mid-sized e-commerce brands underserved.
Weak Integration with E-commerce Ecosystems
Most tools didn’t support deep integration with systems like inventory, shipping, or subscriptions—limiting their ability to automate meaningful interactions.
Subscription Management Was Overlooked
Despite the rise of subscription-based e-commerce models, most chatbots ignored this critical user need, presenting a major opportunity for Buddhi.
Opportunity for a Niche, E-Commerce-Focused Solution
There was a clear gap for a lightweight, affordable, and purpose-built chatbot that can solve specific e-commerce support needs right out of the box.
Workflows to Automate
Key Insights That Informed the Ideation
Support teams spend the majority of their time responding to order-related, sales and subscription queries.
Customers want quick, contextual help with minimal back-and-forth.
Business owners want to reduce support team costs without compromising service quality.
Existing chatbots lack domain depth and proper integration with backend systems.
SUMMARY
Final Solution
Design Principles Followed
Conversational Clarity: Keep dialogues short, helpful, and context-aware.
Modularity: Components like order tracking, promotions, and subscriptions were designed as reusable building blocks.
Fail-safe UX: Clear fallback options, user controls, and easy escalation to human agents.
Tone & Personality: Friendly but professional and adaptable to different e-commerce brands.
Key Interface Decisions
Conversational UI Framework
We designed Buddhi with a guided conversational UI, where users could:
Type freely or
Use quick reply buttons for common intents like “Track my order” or “Cancel subscription”.
Context-Aware Response Blocks
Each automated flow (e.g. order status) includes:
Step-wise info fetching (e.g., asking for email/ID)
Integration call feedback (e.g., “Fetching your order status…”)
Real-time data presentation (e.g., delivery ETA, subscription summary)
Action buttons (e.g., “Pause”, “Cancel”, “Modify”)
Fallback and Escalation Flows
We implemented:
A fallback mechanism when the bot doesn’t understand the query
Explicit handover options like “Talk to an agent”
Passive cues (e.g., “Need help? You can always ask for a human.”)
Brand-Adaptive UI
To ensure that Buddhi could work across different e-commerce brands (fashion, food, electronics, etc.), we built a neutral yet customizable UI system, allowing:
Custom color themes
Brand voice tone setting (formal / casual)
Welcome message editing and logo placement
UI's and other design deliverables
BRANDING
OUTCOMES
85% Query Resolution Without Human Intervention
Customers received accurate and fast answers to most questions without needing to speak to an agent.
4.6/5 Average Satisfaction Score (Early User Testing)
End users appreciated the bot’s clarity, speed, and seamless handoff to human support when needed.
Faster Onboarding Time
Businesses were able to set up the chatbot and integrate it with Shopify under a day.
Consistent Support Quality
Automation ensured every user received consistent, brand-aligned messaging no matter the time or volume of traffic.
What This Proves
A lightweight, well-integrated chatbot like Buddhi can deliver enterprise-grade efficiency to small and medium e-commerce businesses.
Strategic automation of specific flows brings measurable improvements in cost, speed, and customer satisfaction.
The MVP laid a strong foundation for further development such as adding return/refund automation, personalization, and multi-language support.
