Our client is an India-based company that provides IT support and monitoring platforms for small to medium-sized businesses across the United States. The company also runs a helpdesk department in the US to avoid language difficulties and reduce the impact of time zone differences.
The client was struggling with two primary challenges:
- When a user request came in, it took too long to assess and assign it to the correct engineer, which would eventually cause problems in meeting service level agreements.
- Some user requests did not require the involvement of an engineer, and our client wanted to fully process these automatically.
The bottom-line expectation was to decrease operational costs, accelerate ticket resolution, and free up human resources for more value-generating activities.
To address these, the company wanted to deploy a chatbot solution where users could report issues or make requests. The chatbot would attempt to provide a solution by presenting a relevant article or running a pre-defined automatic action. If these first steps did not resolve the situation, an IT support engineer would be assigned. Ensure smarter workflows, reduce operational costs, and accelerate incident resolution.
In response, SoftServe provided the client with:
- A chatbot service that is tightly integrated with Google Dialogflow and fine-tuned according to the client’s specific needs.
- Integration with an existing issue tracking service that enables automated triggering of resolution steps, previously made manually by a support operator.
- Telegram-based step-in component that allows a human operator to take over the conversation seamlessly.
Our team used Dialogflow for Natural Language Processing (NLP) and deep learning techniques for incident categorization. We also ensured the chatbot understands freestyle writing to better capture the users’ intentions.
As a result, SoftServe delivered a smart chatbot solution capable of understanding the user problems, requesting additional details, and evaluating them against potential root causes and resolutions.
Our solution empowered the client to correctly categorize problems in 80% of cases and resolve a significant number of cases fully automatically. This results in:
- 20% faster time to incident resolution
- 10% of cases being resolved without human intervention
Due to the rich embedded ML, the number of cases resolved automatically will grow based on the analysis of processed cases and the resulting adjustments.
Looking for smart and cost-effective way to automate routine workflows and deliver increased value to your customers?
LET’S TALK about how SoftServe can empower you with an intelligent chatbot solution—specifically tailored for your business needs.