To compete, sales agents need a deep understanding of their customers -- staffing changes, corporate strategy, conference schedules, where their competitors are selling and more -- meaning there are thousands of pages on company websites to read, each containing potentially useful clues; tweets from company executives and industry publications to scan; and executive speeches and press releases to monitor. Who has time to cross reference a customer's retail inventory against all of Intel's existing product lines? Who can keep track of it all? A sophisticated advanced analytics program, SalesAssist uses artificial intelligence and machine learning to study publicly available customer data and proprietary sales records. Then it goes beyond summaries and trends, arriving at key -- and most importantly, actionable -- insights about customer goals and needs that Intel's salespeople can use to close deals.

The second component is a suite of machine learning and natural language processing (NLP) models for segmenting potential customers. Web pages are fed into a multi-label convolutional neural network (CNN) text classification model that was developed by Yoon Kim. We further boost it by utilizing a pre-trained multi-lingual BERT language model developed by a team at Google to help scale across languages and classes with scarce training data. The data we use to train the model is enriched by crawling tens of thousands of company sites with labeled industry information found on Wikipedia. For companies without labels, we take advantage of the vast labeled Wikipedia corpus by employing semi-supervised learning.