How McQueen Autocorp Smartly Handles 2500+ Complex Queries/week with the Combination of a Cyborg and its Human Counterparts
Expertise Delivered
Industry
Our client,
McQueen Autocorp is a leading US-based used car selling company with a buyer network spread across the entire United States. McQueen Autocorp offers a nationwide modern car selling service, facilitates sellers with an instant offer for their car(s) using price prediction algorithms and real-time bids from an extensive network of licensed buyers.
Disclaimer - The name McQueen Autocorp is a placeholder as there is an NDA signed between both parties.
Challenge
The Client is a car trading platform where visitors generate instant quotes for their car and the platform connects the buyers and sellers through its strong network of 2000+ buyers selling almost 1,000,000+ cars annually.
In this whole process when a Seller accepts the quote given by the Buyer, both parties are expected to schedule a time slot for the Seller to pick up the car up from the Buyer.
The Client had introduced an AI-based Dispatch Bot which would coordinate with the Buyers and Sellers (via SMS) if there was no pickup being performed or no response being received from either of the parties.
Although the bot was working as per expectations, due to the open input of text boxes and human nature, Buyers/Sellers were responding to the Dispatch Bot with long responses which did not match with chatbot's trained knowledge base. This resulted in failure scenarios for the chatbot which ultimately led to a standstill in the transaction.
Due to this,revenue potential for the Client was being kept on hold as the sale did not go through.
Solution
To address this issue, Maruti Techlabs proposed a solution where the conversations which received such long responses (which the chatbot did not understand) would be passed to an experienced support rep, as they were better positioned to understand the context and respond with appropriate responses.
For this purpose, Maruti’s team created another bot called Cyborg which would come into the picture during instances where the dispatch bot could not respond/understand the response from either the Buyer/Seller and hand over the reins to a human agent for clarification.
This handover to the human agent would occur on Slack (an internal communication tool used internally by the client), where Cyborg would post the message on a specific channel.
For the human agent to clarify the context of the response, Cyborg provided the agent with the following options:
- Confirmation - Proceed with the pickup
- Yes - Consider the affirmative reply and send reply to the Seller
- Complete - System treats reply as affirmative
- Not Sure - Unrecognized reply, system asks the question again
- Stop - Unsubscribe the phone number
- Stop & Cancel - Unsubscribe the phone number and Cancel Offer
- Flag - Create “SMS” cases in salesforce and agent replies manually to the
The human agent would choose an option that matched what they assumed the Buyer/Seller was trying to say and based on this, Cyborg would pass the next instruction to the Dispatch bot.
With Cyborg, multiple such situations were being handled manually by human agents and the transactions were successfully being performed.
Experience unmatched efficiency and streamlined operations, all at your fingertips.
Technology Stack
Our Development Process
We follow Agile, Lean, & DevOps best practices to create a superior prototype that brings your users’ ideas to fruition through collaboration & rapid execution. Our top priority is quick reaction time & accessibility.
We really want to be your extended team, so apart from the regular meetings, you can be sure that each of our team members is one phone call, email, or message away.
