From AI Features to Enterprise AI Agents
Enterprise software AI acquisitions now center on building an AI execution layer, where agentic workflow automation systems can autonomously orchestrate, decide, and act across multiple business applications with minimal human intervention, turning static software into adaptive, outcome-oriented enterprise AI agents. This is a marked change from early AI add-ons that focused on chat, summarization, and recommendations inside a single product. Vendors such as Asana, Coupa, Salesforce, Vertice, and SITA are buying companies that plug directly into operational systems and carry out work, not just propose the next step. The result is a race to turn ERP, CRM, and procurement platforms into “agentic” environments, in which human users supervise AI agents that execute workflows spanning finance, operations, sales, and support. For buyers, the central question is how reliably these agents can complete business-critical processes under real-world constraints.
Asana’s StackAI Deal and the Rise of Human-Agent OS
Asana’s USD 75 million (approx. RM345 million) acquisition of StackAI signals how work management is shifting toward embedded AI execution. StackAI built no-code AI workflow agents that connect tools like Salesforce, Slack, Google Workspace, AWS, DocuSign, and Oracle, enabling cross-system workflows to run end-to-end. Asana plans to combine this with its Work Graph, AI Studio, and AI Teammates to form a “human-agent OS” where agents act on project context, ownership, and history, not isolated tasks. CEO Dan Rogers said the deal helps customers “agentify” complex business processes rather than automating only simple intake or routing flows. For Asana, which has seen its market value fall since the rise of general-purpose AI, owning an execution layer differentiates it from standalone automation tools that sit on top of, rather than inside, core work management platforms.
Vertice, Vendr and Agentic Procurement Intelligence
In procurement, Vertice’s acquisition of Vendr shows how enterprise AI agents are being trained on rich commercial data. Vertice already supports finance and procurement teams with agentic workflows, AI-powered insights, and specialist buying expertise, processing over USD 75 billion (approx. RM345 billion) in spend and delivering more than 20 percent savings while doubling procurement speed. Vendr contributes benchmarks and market insights drawn from real-world software buying. Together, the platforms form an AI-powered procurement intelligence dataset spanning more than USD 75 billion (approx. RM345 billion) in indirect spend across 32,000 vendors and 250,000 negotiated contracts. According to Vertice founder and CEO Roy Tuvey, “Vertice and Vendr have shared a vision for AI in procurement: to build purpose-designed AI agents trained on real-world data and tailored to specific procurement use cases.” Those agents are meant to guide and eventually execute negotiations, renewals, and vendor comparisons directly within Vertice.
SITA and Big Blue Analytics Bring Agentic AI to Airline Disruption
The push toward AI execution is not limited to office workflows. SITA’s acquisition of Big Blue Analytics, creator of the OCC Assistant Manager (OCCam) platform, brings proven AI disruption optimization into mainstream airline operations. OCCam is an AI-enabled decision system that responds to disruptions by evaluating every active constraint at once—aircraft, crew, passenger itineraries, maintenance—and generating coherent recovery plans in minutes. Airlines using OCCam have reduced disruption costs by up to 30 percent, a major gain in an industry where a mid-size carrier can lose between USD 70 million (approx. RM322 million) and USD 80 million (approx. RM368 million) annually to irregular operations. Instead of sequential reassignments that create rework, OCCam’s agentic approach optimizes across the full network and presents ranked options, allowing operations control centers to approve and apply system-wide recovery plans at scale.
Toward Agentic ERP and CRM: Coupa and Salesforce Join the Race
Other recent enterprise software AI acquisitions point to the same destination: ERP and CRM platforms with built-in AI execution. Coupa’s purchase of Rossum brings a specialized transactional large language model, trained on tens of millions of documents, into its source-to-pay stack, so AI agents can interpret complex invoices and drive spend automation across the platform rather than sit in a separate OCR tool. Salesforce’s agreement to acquire Contentful gives its Agentforce offering a native content layer. Agentforce will be able to query structured content, assemble tailored experiences, and deliver them across Customer 360 without manual publishing steps. Together with Asana and Vertice, these deals show major vendors converging on a model where data, workflows, content, and action endpoints are tightly integrated, enabling AI agents to start, progress, and complete workflows that touch multiple enterprise systems.