Lenovo is a leader in personal computers, tablets, smartphones, servers, and storage segments with operations in 170+ countries worldwide.
Lenovo’s warranty costs were rising given the expanding product footprint. Product quality teams wanted to detect and respond to issues faster. Lenovo wanted to listen to and extract customer sentiments across a variety of social and traditional channels to respond faster to issues gathering momentum.
Core Compete architected and deployed a solution that listens to a variety of communication channels to extract sentiments at a detailed component level. Detailed insights are then integrated with structured data from quality databases to generate alerts when a product quality issue showed potential to snowball.
To drive maximum value from the big data and analytics solution, Core Compete brought together the right components in four critical areas.
Domain Expertise – Unstructured data from 27 channels spanning 177 countries including product reviews, forums, blogs and social posts. Structured data related to customer complaints and product quality. Designed and implemented data ingestion and quality management strategies.
Analytics Innovation – The team designed and implemented industry-specific taxonomy for products, advanced linguistic and statistical NLP models and workflows to sift through huge volumes of social data in multiple languages to accurately conclude drivers of negative customer sentiment on a global scale.
Technology Know-How – Core Compete designed and developed the solution on AWS, based on SAS Contextual Analytics to deliver intuitive, actionable insights with visualizations that were developed in Qliksense. The solution was directly integrated with an on-premise database that housed the quality database.
Agile Delivery Methodology – A 12-month roadmap was developed to tackle the problem in an incremental manner. The roadmap considered data quality, analytical complexity, business value and change management. Phase 1 of this solution encompassing ~900 products and over 1100 features was completed in 12 weeks.
The solution is being used by over 300 end users from different departments across the globe. Lenovo was able to shrink issue detection times by 50% resulting in a 15 % warranty cost reduction.