With a growing product portfolio and increasing support volume, it is essential to adopt automation to scale support delivery. Red Hat constantly evaluates how to empower customers to self-solve using search and by building tools for engineers to resolve cases faster. Using self-solve rate and time to close(TTC) as primary KPIs will determine success. In this session we will cover the evolution of different search techniques in our Solution Engine and the customers’ search journey. We will identify the challenges to provide an accurate and relevant solution for customer issues before opening a support case. We will dive into the query parsing for human vs machine generated data, relevancy model for the wide array of products and evaluation aspects. We will also describe how adopting ML classification techniques helped improve language detection and faster case routing to the specialists. This session will give insights into how to leverage these techniques to promote customer self solving behavior and findability by using search and machine learning.
Speaker: Manikandan Sivanesan, Red Hat