ServiceNow is a leading American enterprise software company headquartered in Santa Clara, California, with over 29,000 employees globally. The company reported impressive fourth-quarter 2025 subscription revenue of $3.57 billion and projects fiscal 2026 subscription revenue guidance between $15.53 billion and $15.57 billion. ServiceNow has made substantial investments in artificial intelligence and automation to significantly enhance workflow efficiency and enterprise productivity. Notable strategic moves include the acquisition of Passage AI to bolster conversational AI capabilities, an expanded partnership with NVIDIA to develop autonomous AI agents, and a commitment of $1 billion through its venture arm to support AI-related startups and enterprise software innovation. Additionally, ServiceNow has invested CA$110 million to promote AI adoption in Canada's public sector, establishing an AI Center of Excellence and related infrastructure. The company actively leverages its internal AI platform, 'Now on Now,' to achieve demonstrable ROI, showcasing how organizations can move beyond theoretical AI implementation to practical, scalable automation solutions.
Reducing Agent Documentation Time With Embedded Generative AI
Research from San Jose State University highlights that customer service agents spend a significant portion of their time (35-45%) on repetitive documentation tasks. This inefficiency costs U.S. enterprises an estimated $2.6 billion annually. A Harvard Business School study corroborates these findings, noting that excessive documentation diverts agents from more valuable problem-solving activities. To address this, ServiceNow introduced 'Now Assist' for IT Service Management (ITSM) and Customer Service Management (CSM), directly embedding generative AI into agent workspaces. Unlike traditional standalone chatbots, Now Assist automates the summarization of incident histories, drafts resolution notes, and facilitates the creation of knowledge articles seamlessly within existing agent workflows. The platform utilizes machine learning and generative AI to handle these mundane aspects of support cases, eliminating the need for agents to switch between external tools. ServiceNow reports that Now Assist generates notes within seconds, allowing agents to review and refine them rather than starting from scratch. This functionality has reportedly reduced the time required for each resolution note by approximately 80%. Specifically, ITSM agents save an average of 4-6 minutes per use, while CSM agents save 12-16 minutes per use. These results underscore the tangible value of integrating generative AI directly into enterprise workflows for enhanced productivity.
Predicting Customer Escalations Before They Happen
Traditional reactive support models, which depend on manual monitoring of tickets and events, often fail to identify deteriorating customer experiences before they escalate. This lack of a scalable mechanism for predicting at-risk accounts leads to inconsistent and often belated proactive outreach efforts. ServiceNow has revolutionized this approach by implementing machine learning to predict and prevent customer escalations. A case study by ServiceNow details how the company transitioned from a reactive stance to a proactive one, leveraging its Predictive Intelligence and Event Management capabilities. Instead of waiting for customer complaints or threats of escalation, the system proactively identifies vulnerable accounts, enabling early intervention. The solution is built upon ServiceNow’s Predictive Intelligence framework, which incorporates a machine-learning model, and Event Management for real-time ingestion of performance-related data. The supervised model is trained using historical escalation patterns, including tickets, surveys, CSAT scores, and engagement signals, with real-time system alerts from Event Management enhancing its predictive accuracy. The workflow involves three key steps: building and training the model using structured features from historical data, deploying real-time risk scoring for continuous customer assessment, and automating proactive interventions for high-risk accounts. The system automatically generates priority alerts, assigns follow-up tasks to support or account teams, and suggests recommended playbooks. Continuous feedback from each engagement further refines the model’s predictions over time. This approach has led to significant business improvements, including faster response and resolution times, higher customer satisfaction, and smoother renewals. Before implementation, only about 11% of customer engagements were proactive; post-implementation, this figure rose to approximately 68%. The system effectively engaged hundreds of customers annually, preventing a large proportion of escalations while maintaining a low false-positive rate of around 3%, thereby optimizing engineering resource allocation.
