Customer Relationship Management - Case Study
The client is into the business of managing IT Infrastructure & Cloud Services.
Brand: Reputed Client in IT Industry
Core Platform : Web
Programming Language: .net Core
Framework: Azure Bot Service (AI), Azure Cognitive Service (AI), Azure SQL Database
IT companies, Software companies, Development agencies
Across Europe and the USA.
- No predefined business logic.
- The business was entirely based on FAQ, NLP & NLU logic to provide required information and details to resolve client queries.
- Keep updating and adding FAQs list.
- Get accurate search results at the earliest time possible.
- Address customer queries with personalization.
The client's requirements for getting various tasks done from the chatbot were feasible; the only challenge was to keep adding FAQs and get faster and accurate results effortlessly.
To design the highly customized Chatbot web application, our experts took the extensive analysis of the nature of the client's business, its target audience, the challenges faced, and the goals to be achieved. Further, the analysis of B2B and B2C clients helped to get the solution to many technical queries and the issues related to the cloud platform services. The analysis also revealed that the client needs to keep a 24x7 support team to answer queries to their client, which was time and cost expensive, thus to reduce first and second level people and cost, we need to introduce the Chatbot.
To determine the role of the Chatbot and set goals, we started gathering detailed requirement from a client and analyzed their current business process of support, which also helped us to prepare the list of intents, AWS services FAQs they want to cover in BOT, and Custom FAQs which currently they have in place. We then implemented the business logic to handle the context of interaction and Work Flow (Sequence of Questions and Answer) to meet the business requirements successfully.
Before the actual project started we collected the following documentation to ensure we are building the right application
- Design documents
- Sample Data/Physical Printed Forms of current manual process
- Organization hierarchy and its possible accessibility
- Answering repetitive queries.
- Capitalize the infinite potential of a chatbot to interact with more users faster than a human.
- Solve day to day activities of the support team.
- Make customer service efficient.
- Streamline the interactions and reduce the cost of customer service.
- Automate and reduce common activities of its managed services of hosting service providers on the Cloud platform.
- Channelize the cloud platform related service FAQs performed through the ChatBOT.
- Reducing the time spent on technical FAQ by Juniors in the support team of their managed services.
Azure BOT Services developer
.net Core developer
- Develop a BOT that can assist the support team for their questions.
- Address client's repetitive queries.
- Faster and relevant interaction with more users.
- Solving day to day activities of the support team.
- Enabling efficient customer service.
- Conversational Maturity
- Text Chat BOT
- Retain data and context seamlessly
- Use of Azure BOT and Azure Services only.
- Accuracy level 80% or higher at an initial level.
- Can perform reasoning without human intervention.
- Deeply integrates with CRM.
Key Takeaways and Learnings
Azure BOT default was not assisting much in NLP, for NLP purpose Azure Cognitive service was required to be capitalized..
.Net Core was a more efficient and better way to develop client-side functionality instead of TypeScript or any other scripting language.
Reduction in cost
- Increase in response time of support by 5 times.
- Reduction in cost by over 20%.
- Helped the business to provide a 24*7 support system.
- Provided seamless customer experiences.
- Reduced the strain and complexities of the support team.