Reimagining Interior Designing With Conversational AI: A Case Study

Reimagining Interior Designing With Conversational AI: A Case Study
Image © Reimagining Interior Designing With Conversational AI: A Case Study


Providing personalised customer experience has now become the main concern for businesses across all domains, whether it be eCommerce, retail, healthcare, BFSI or online services. And, one of the key ways businesses are providing personalised and comprehensive customer experience is by deploying conversational AI chatbots. Case in point — Bangalore-based Design cafe, a home interior design company, that has been struggling with serving a large customer base during the pandemic due to less number of customer care executives.

As the pandemic hit, there has been a considerable increase in the number of customers who were demanding services even at odd hours on WhatsApp or website chat. However, due to less number of customer care executives during the lockdown, there was a massive delay in response time, becoming a challenge for Design Cafe to address customer issues. Therefore, the company was looking to adopt an approach where they could be available to customers at all hours.


Sign up for your weekly dose of what's up in emerging technology.

To meet such an increased demand for virtual interactions, amid pandemic, Design Cafe decided to partner with conversational AI startup, Yellow Messenger in order to build a chatbot that can help them improve their customer experience on both the web chat as well as their WhatsApp messaging platforms. 

To understand the case better, Analytics India Magazine spoke to Ajay Maheska, the director – marketing at Design Cafe, along with Rashid Khan, the co-founder and chief product officer and Sajin Sam Mathews, the project manager of Yellow Messenger.

“Our digital assets, like websites, social media and paid advertisements, have been a major source of acquisition for new customers,” said Maheska from Design Cafe. “Once the pandemic hit, we noticed a significant increase in the volume of interactions on chat and WhatsApp channels that was leading to bandwidth issues at the call centre.”

Yellow Messenger proposed a conversational AI chatbot assistant with lead generation, push notification features along with live-agent support for Design Cafe’s customers.

Also Read: How Government Of India Used Conversational AI During COVID-19: A Case Study

The Tech Behind

Yellow Messenger designed the AI-enabled chatbot to provide yet another touchpoint for customers to reach out to Design Cafe to book meetings with designers as well as seek information 24/7. To facilitate this, Yellow Messenger created a product based on deep learning led artificial intelligence. 

From a product standpoint, the team followed the jobs-to-be-done (JBTD) framework when solving customer problems, and leveraging AI on the key parts of the platform helped in converting the jobs-to-be-done to jobs completed. One of the key ways Yellow Messenger leveraged AI was to understand customers’ language. From a birds-eye-view approach, language understanding has been divided into two aspects — firstly, taking natural language from end customer as input and giving out the identified intents, i.e. actions to be performed and entities, i.e. data required for action to be performed; secondly, leveraging the identified intent and driving the next best action (NBA) for the customer.

The machine learning models are primarily based on TensorFlow on the backend and are front-ended by a Python flask web server for managing real-time traffic. On the other hand, for training and batch jobs, the company leveraged tools like Celery for queue and tasks.

Customer Experience Flow

Deep learning-led artificial intelligence works on the basic premise of leveraging previous experiences like conversations, tickets, CRM data, etc., to come up with a model to understand language. The key component of achieving high performance in terms of understanding and triggering the right action is dependent on data. 

Explaining this, Khan stated, “Yellow Messenger platform in a given quarter handles more than one billion conversations across the platform, which helps in building more robust models, in-turn driving better conversations on the platform, and building a learning and improvement loop.”

The bot has also been trained on FAQs and shows various images and videos as a response to customer queries. “As per the requirements, we proposed three different conversational flows based on the inputs given by their marketing and product team. And each of the flows was brainstormed and modified according to their exact requirement, and agreed upon the best/most suitable bot flow,” added Mathews from Yellow Messenger.

Bot interface 

The architecture of the bot was built as a three-tier platform architecture – cognitive platform, developer platform and integrations platform. The cognitive platform was built mainly on Python and deep learning, where the company leveraged TensorFlow for production workloads. “Quite recently we built out a generative model using OpenAI’s GPT-3 model,” stated Khan.

The developer platform is built on top of NodeJS, for the backend, and ReactJS, for the front-end. The integration platform leverages third-party APIs to be natively available in our flow builder and marketplace. “We have open support for REST/SOAP-based APIs and are adding support for GraphQL,” added Khan.

Also Read: Chatbots In Mental Health. Friendly But Not Too Friendly. 


Design Cafe wanted a solution that can help in generating leads through various social media campaigns and give an in-depth understanding of the source of each lead. “The idea was to market their solutions and products by reaching out to new prospects through Paid, Organic and Direct leads,” stated Maheska. 

Post the deployment, Design Cafe noticed good traction in the leads being captured, and the admin was able to capture the source of each lead, analyse each conversation, check the analytics of the bot usage/adoption, track the information of each lead as well as notify the customers on various proposals. Users were also able to scan QR code to chat with the bot. And, the live-agent chat support was able to support the customer with personal queries/doubts.

Design Cafe witnessed significant adoption of the chatbots across website chat and WhatsApp platforms, with more than 50% WhatsApp users being able to schedule meetings with designers without human intervention. In October 2020, more than 15% of our new customers had scheduled meetings via the chatbot. It not only improved the customer experience during COVID but also increased the number of meetings booking with designers at night without any human intervention.

To further refine customer experience, Yellow Messenger is working on leveraging the survey module, campaign manager, and Google calendar integration features, in addition to different channel integrations like FB messenger and Google Business Messages.

More Great AIM Stories

Sejuti Das
Sejuti currently works as Associate Editor at Analytics India Magazine (AIM). Reach out at

Our Upcoming Events

Conference, in-person (Bangalore)
MachineCon 2022
24th Jun

Conference, Virtual
Deep Learning DevCon 2022
30th Jul

Conference, in-person (Bangalore)
Cypher 2022
21-23rd Sep

3 Ways to Join our Community

Discord Server

Stay Connected with a larger ecosystem of data science and ML Professionals

Telegram Channel

Discover special offers, top stories, upcoming events, and more.

Subscribe to our newsletter

Get the latest updates from AIM