The materials science giant — and world’s largest Google Gemini installed base — has achieved broad benefits from an AI strategy that emphasizes enterprise-wide training and ‘humans in the loop.’ Now CIO Nick Colisto is looking to scale up. Credit: Avery Dennison Avery Dennison’s culture-first approach to AI adoption has paid off considerably in productivity and financial gains for the global labeling and RFID solutions manufacturer. The company’s program to drive AI use throughout the enterprise, emphasizing “humans in the loop,” engaged more than 250 company executives and 1,000 employees in over 20 AI pilot projects in 2024. By addressing cultural issue first — demonstrating how employees can bring AI alive — the materials sciences giant has enjoyed an abundance of financial benefits from AI innovations developed in-house. These include enhanced inventory forecasting and predictive maintenance, but perhaps most important, the structured rollout of Google Gemini to more than 22,000 employees globally. Avery Dennison boasts the largest installed base of Gemini pilot testers in the world. Over time, its work with Gemini has translated into an average savings of 10 hours per month per employee. That’s no small feat in an industry full of enterprises talking big about AI but doing less to train all employees on how to get tangible value from AI models. “It was a flagship win because it touched all of our employees and gave us a lot of credibility with AI very quickly,” says 2021 CIO Hall of Fame honoree. “People found the power in it. We had a 37% adoption rate in the very beginning, and that has only grown.” Inventory forecasting, predictive maintenance, and Gemini are just three of more than 20 AI projects created as part of Project Loop, a nod to keeping humans in the AI loop, which earned Avery Dennison a 2025 CIO 100 Award for IT innovation and leadership. The company, which manufactures and distributes pressure-sensitive adhesive labels, apparel branding labels, and tags and RFID inlays, has also deployed marketing platform Jasper.AI, which has provided yet another win, speeding up content creation by a factor of two to three times in the marketing department. Synthesia’s video creation tool, also used as part of the pilot, enables employees to produce impactful content at scale, reducing production costs and saving valuable time, Colisto says. Education, training, and collaboration drive adoption To prepare for its leap into AI, Avery Dennison created a Digital Innovation Center of Excellence (DICE) and partnered with a major online training program to upscale its IT workforce through IT Academy, hackathons, and workforce development days. Non-IT employees were provided with training and education on AI features in Google Workspace applications, providing robust Gemini training and exposure to AI tips and techniques. Following the workforce education on generative AI, Colisto launched ideation sessions and brain “writing” exercises, during which viable business use cases were chosen from hundreds of ideas proposed throughout 2023. Then, in 2024, the CIO and Avery Dennison’s global workforce moved from concept into action, creating 21 generative AI pilots that improved the productivity of all Google Workspace users and enhanced several business processes, including maintenance, operations, customer engagement, and workplace safety. The abundance of commercial-grade generative AI tools available from Google and four other major AI platform partners enabled Avery Dennison to move full steam ahead. “We don’t want to build a lot of AI,” Colisto says. “Our main strategy is to leverage the large, established, enterprise-grade platforms developed with billions of dollars in innovation.” Many other CIOs are increasingly going this same route, foregoing homegrown AI proofs of concepts in favor of commercial solutions. In addition to leaning on the innovation of Avery Dennison’s five key AI providers, the company also looks at startups, he says. “I’m sure there are some niche development opportunities but there’s a lot of waste, too.” The CIO was clear that there was no desire to reinvent the wheel — but there were notable exceptions. As part of Project Prophesy, for example, the materials science giant worked with a consultant to develop a custom AI application that improved the accuracy of demand forecasting, reducing errors by more than 60%. This was achieved by identifying and collecting data that impact inventory forecasting, such as shipping, port activity, weather, credit data, and even GDP statistics, to optimize inventory planning and reduce costs associated with overstock and stockouts. Such improvements, Colisto says, are expected to save the company millions of dollars annually in working capital and generating operational efficiencies. “We were able to leverage all these different data sets externally and help optimize inventory planning and reduce costs in our forecasting,” says the CIO, pointing to the difficult challenges of forecasting inventory during and after the pandemic. “This tool really helped us navigate through that.” The gift that keeps giving Measured benefits of the company’s creation and deployment of AI pilots throughout 2024 — some of which were launched in specific regions — has given Avery Dennison the green light to scale up globally through 2025 and beyond. For instance, applying AI to predictive maintenance in one of the company’s India plants reduced unplanned downtime by 25%, while improving operational schedules and customer satisfaction and decreasing maintenance costs, the company said. Another AI pilot the company launched in Latin America was shown to strengthen customer relationships and improve revenue predictability. Predicting customers’ next purchase dates in the region achieved 90% accuracy, enabling proactive customer engagement, better inventory alignment, and reduced churn, according to the company. Avery Dennison’s VelocityEHS ergonomics platform is yet another of the 21 pilot projects to provide measurable improvements — in this case by identifying and mitigating high-risk tasks to cut down on workplace injuries. The Mentor, Ohio-based company is now working to scale these and other AI solutions across the global enterprise, Colisto says. One major initiative this year is to measure the outcomes of all this AI activity and determine ways to keep innovating from within — with contributions from its entire workforce, not just a small band of developers. “2023 was a year of learning and ideation, 2024 was a year of building, and 2025 is about scaling,” Colisto says. “We’re scaling the experiments with our business units and functions around the world to monetize them, gain more value in those experiments, and measure them this year.” Next up? The CIO said the workforce is pushing ahead on the next round of “moonshots” to advance AI use for other functions within the corporate walls and in customer-facing solutions. “We’re looking at the challenges we see, both externally and internally, that we think are addressable with AI,” he says. SUBSCRIBE TO OUR NEWSLETTER From our editors straight to your inbox Get started by entering your email address below. Please enter a valid email address Subscribe by Paula Rooney Senior Writer Paula Rooney is a senior writer at CIO.com, where she focuses on how CIOs deploy AI, cloud, and digital technologies to transform their organizations. A veteran IT journalist, Paula has reported for PC Week, CRN Linux.com, The Register, TechTarget and ZDnet, and UBM, among other outlets. She holds a master’s degree in journalism from Columbia University and was most recently recognized with ASBPE Regional Silver and Regional Bronze awards for her enterprise news story “LA Public Defender CIO digitizes to divert people to programs, not prison.” More from this author featureIBM’s massive SAP S/4HANA migration pays off May 21, 2025 7 minsfeatureBank of America’s big bet on AI started small Apr 25, 2025 8 mins Show me morePopularArticlesPodcastsVideos news Salesforce to acquire Doti to boost AI-based enterprise search via Slack By Anirban GhoshalNov 14, 20253 mins Artificial IntelligenceMergers and AcquisitionsSalesforce opinion MCP doesn’t move data. It moves trust By Shail KhiyaraNov 14, 20258 mins Data GovernanceGenerative AIIT Governance opinion AI, without fear: An open note to IT By John WeiNov 14, 20256 mins Artificial IntelligenceIT LeadershipRisk Management podcast CIO Leadership Live ASEAN: Scaling AI Responsibly at Mondelez By Estelle QuekNov 9, 202519 mins CIODigital TransformationInnovation podcast CIO Leadership Live Australia with Connie Beck, Group Director - Transformation & Digital, Urbis Nov 5, 202518 mins CIO Leadership Live podcast CIO Leadership Live ASEAN: AI at Scale-Balancing Innovation with Integrity By Estelle QuekNov 4, 202522 mins CIODigital TransformationTechnology Industry video Inside Genesys' AI Studio: How agentic AI enables business. users to build better bots Nov 12, 202512 mins Generative AINo Code and Low CodeRetail Industry video CIO Leadership Live ASEAN: Scaling AI Responsibly at Mondelez By Estelle QuekNov 9, 202519 mins CIODigital TransformationInnovation video CIO Leadership Live Australia with Connie Beck, Group Director - Transformation & Digital, Urbis Nov 5, 202518 mins CIO Leadership LiveIT Leadership Sponsored Links CIO Think Tank Roadmap Report: Information management in the age of AI Reimagine tomorrow. Become a future-ready, AI-powered enterprise with EY Consulting.