The News
At its Gemini at Work 2026 event, Google Cloud highlighted how enterprises are moving beyond AI experimentation toward broader operational deployment, emphasizing measurable productivity gains, enterprise data integration, and industry-specific applications. Google and Alphabet CEO Sundar Pichai reported that nearly 80% of Google Cloud customers actively use its AI products, while nearly 500 enterprise customers have each processed more than one trillion tokens over the past year. Google Cloud CEO Thomas Kurian reinforced that momentum with customer examples from Nokia, NTT Docomo, Ulta Beauty, Merck, and Honeywell. Beyond introducing Gemini as a universal agent for work, Google detailed enhancements to its Borderless Lakehouse, Knowledge Catalog, AI-assisted data engineering, and specialized agent capabilities for regulated industries. Together, these developments reflect a growing focus on connecting enterprise AI investments to measurable operational and business outcomes.
Analyst Take
Enterprise AI is entering its operational value phase
The most significant takeaway from Google’s Gemini at Work briefing was the breadth of operational results customers are beginning to report. Enterprise AI adoption is increasingly being evaluated through improvements in business processes, employee productivity, and customer experiences rather than model capabilities alone.
Google highlighted several examples of this transition. Nokia reported reductions of up to 80% in telecommunications network troubleshooting time, while NTT Docomo reduced the time required to move from data collection to insight from two weeks to nearly instantaneous, recovering an estimated 450,000 hours annually. In retail, Ulta Beauty reported conversion rates three times higher through its Gemini Enterprise-powered shopping assistant compared with traditional browsing. Bunnings also reported saving employees approximately half a million hours through an internal agent deployment.
These examples demonstrate how AI can create measurable value across different operational environments. While customer-reported metrics should be evaluated within the context of individual deployments, they suggest that structured, repeatable workflows offer particularly promising opportunities for enterprise AI. The broader shift is from asking whether AI can perform a task to determining whether it can consistently improve that task at scale.
Enterprise data accessibility is becoming a competitive differentiator
One of the more technically consequential portions of the briefing centered on Google’s approach to connecting AI systems with enterprise data distributed across cloud platforms, SaaS applications, and on-premises environments. Google’s Borderless Lakehouse and Knowledge Catalog address complementary aspects of this challenge. The Borderless Lakehouse supports access to information across platforms such as Amazon S3, Azure Data Lake, Databricks, and Snowflake, while Knowledge Catalog provides shared business definitions, schemas, and metadata that help AI applications interpret organizational information consistently.
For developers and data engineering teams, the distinction matters. Retrieving enterprise data is only one part of the process; understanding how an organization defines revenue, risk, customer value, or operational performance requires contextual information that may not be apparent from the underlying data. For enterprise buyers, Google’s approach highlights the importance of making existing data environments AI-ready without introducing unnecessary migration requirements or operational complexity. It also positions data infrastructure as an increasingly important component of enterprise AI differentiation.
Google demonstrated these capabilities through a financial services workflow in which Gemini generated PySpark code, trained a machine learning model, and made the resulting analysis available to business users for campaign creation and performance monitoring. Notably, reusable query templates enabled subsequent dashboard interactions without repeated LLM inference, offering a practical approach to improving cost predictability and analytical consistency.
Connecting technical and business workflows creates new productivity opportunities
Another important theme was the potential to reduce handoffs between technical teams and business users. Building an analytics workflow traditionally involves data discovery, engineering, model development, validation, application integration, and reporting, often across multiple disconnected systems. Google’s demonstration illustrated how Gemini could help connect these stages. A data scientist could identify relevant datasets, generate and inspect PySpark code, train a model, and publish it for business use. A marketing team could then apply that model to identify customer segments, prepare campaign materials, and monitor results through a live dashboard.
The technical significance is not that AI replaces developers or data scientists, but that it can reduce repetitive implementation work while preserving visibility into code, data, and model behavior. ECI Research’s Google GovTech Survey Results found that 47.2% of respondents identified developer velocity and ease of integration as the most important factor in final technical selection, assuming baseline security requirements were satisfied. This reinforces the value of platforms that simplify connections between enterprise systems and accelerate delivery. For development and platform engineering teams, the longer-term opportunity lies in creating reusable capabilities that support multiple business functions rather than treating every AI deployment as a standalone project.
Industry specialization strengthens the enterprise AI value proposition
Google’s introduction of specialized Gemini capabilities for financial services and legal workflows, with government, healthcare, and retail offerings planned, reflects another important direction for enterprise AI adoption. Rather than relying exclusively on general-purpose models, these offerings incorporate industry-specific skills, trusted data sources, and workflow integrations. In a financial services demonstration, Gemini evaluated the impact of a hypothetical acquisition on a $100 million investment portfolio, combining market information, internal holdings, regulatory considerations, and risk analysis to produce recommendations and supporting financial documents.
Customer examples further illustrated the importance of specialized capabilities. Merck described using Gemini Enterprise to support molecular research and drug discovery, with approximately 75,000 employees having access to the platform. Honeywell highlighted its use of Gemini Enterprise within its Forge Intelligence platform to support predictive maintenance by combining sensor information with operational context to identify potential equipment failures earlier. These applications demonstrate why domain knowledge and data integration are becoming essential to enterprise AI performance. General-purpose reasoning capabilities provide the foundation, but practical business value often depends on understanding specialized terminology, operational requirements, and established workflows. For Google, industry specialization creates opportunities to expand Gemini Enterprise into higher-value applications. For buyers, it reinforces the importance of evaluating AI solutions against the specific accuracy, integration, and operational requirements of their industries.
The economics of AI increasingly depend on repeatability and scale
Beyond productivity gains, Google’s briefing highlighted the importance of controlling the operational costs associated with enterprise AI deployment. Capabilities such as reusable skills, project spending limits, and query templates that execute without repeated model inference address a practical challenge: AI costs can become less predictable as agents perform longer-running tasks and operate across multiple business functions. The ability to establish reusable workflows offers a potential efficiency advantage, particularly for recurring reporting, financial analysis, and operational monitoring. Rather than repeatedly generating the same outputs through model inference, organizations can combine AI-assisted development with more predictable execution methods.
For IT decision-makers, the economic equation extends beyond token pricing to include integration, infrastructure, maintenance, and human oversight. Google’s approach reflects a broader industry shift toward evaluating the total cost of completing a business process rather than the cost of an individual AI interaction.
Looking Ahead
Google’s Gemini at Work briefing illustrated how enterprise AI is progressing toward more integrated, outcome-driven deployment. Customer productivity results, data infrastructure enhancements, and industry-specific applications suggest that the next phase of adoption will be shaped by how consistently organizations can translate AI capabilities into measurable business improvements.
Over the next 12 to 18 months, an important indicator will be whether enterprises can extend successful deployments across additional departments while maintaining accuracy, predictable costs, and operational consistency. Google’s combination of enterprise data infrastructure, specialized AI capabilities, and workflow automation provides a foundation for supporting that expansion.
For developers and data engineering teams, the priority is building reusable integrations and workflows that make AI capabilities easier to deploy and maintain. For enterprise leaders, the focus is identifying applications where AI can deliver measurable improvements without introducing unnecessary complexity. Ultimately, the strongest indicator of enterprise AI progress will not be the number of agents deployed, but the business outcomes those agents consistently deliver. Google’s Gemini Enterprise strategy reflects that shift, emphasizing integrated data access, industry specialization, and operational efficiency as increasingly important components of enterprise AI value.
Stay Ahead of Application Development Trends
Get weekly analyst insights, research notes, event coverage, and AppDevANGLE updates delivered directly to your inbox.
Subscribe for Weekly Insights
Join technology leaders, practitioners, and GTM teams following the trends shaping modern software delivery.
Looking for deeper research access?
Explore ECI Research reports, survey insights, and market analysis through the ECI Research Portal.
