Agentforce is Salesforce’s new operating system, not another CRM feature
Salesforce Agentforce is an enterprise AI agent platform that connects data, workflows and channels so autonomous agents can handle ecommerce, customer service and fan engagement tasks across the business, instead of being limited to traditional CRM record-keeping and ticket routing. This matters because it signals a strategic shift: Salesforce is trying to turn CRM from a system of record into an AI execution layer that directly drives sales, service and revenue operations. The company’s recent moves cluster around a single thesis: AI agents will sit between customers, employees and systems, orchestrating actions end-to-end. With Agentforce Commerce, Help Agent and high-profile deployments in Formula 1, Salesforce is making it clear that its future growth depends on enterprise service automation, not on selling more licenses for classic CRM screens.
AI agents in ecommerce: from B2B procurement to shopper concierges
Salesforce’s latest Agentforce Commerce release makes AI agents a first-class citizen in ecommerce, not a plug-in bot. For B2B customer engagement, the new Buyer Agent and Merchant Agent are designed to run the entire procurement flow inside channels like WhatsApp and SMS. One example: a buyer can text “Need 40 cases of the 16-oz fasteners, same as the March order,” and the agent confirms the SKU with images, applies contract pricing and completes the order, without a portal login or phone call. On the B2C side, a Shopper Agent and agentic commerce search aim to turn sites into conversational storefronts, backed by deep integrations with ChatGPT, Gemini and Google AI Mode. Crucially, every order lands on the same Salesforce platform that already runs service, loyalty and marketing, which tightens the loop between sales, support and post-purchase engagement instead of scattering data across separate admin tools.
Help Agent’s outcome-based pricing could reset service economics
Agentforce Help Agent, launched on June 25, is Salesforce’s clearest shot at enterprise service automation — and its business model is more disruptive than its UI. The AI service agent runs across voice, web, portals and messaging, with guided setup and prepackaged actions for case management, appointments, order updates and account changes. The real break from legacy CX economics is pay-per-resolution pricing: organizations pay only when Help Agent resolves a customer issue end to end, with no charge if the customer escalates to a human or leaves negative feedback. One quotable line captures the intent: “Salesforce said the model ties costs directly to successful outcomes rather than consumption or activity.” In practice, that means service leaders stop paying for ‘attempts’ and start paying for verified resolutions — a direct response to an industry-wide pivot toward outcome-based pricing in AI agents. If the model scales, it will pressure rivals who still bill per interaction, not per solved problem.
From F1 fans to quote-to-cash: Agentforce pushes into the entire revenue stack
The Agentforce story is not confined to support desks. In Formula 1, Visa Cash App Racing Bulls is deploying Agentforce 360 to personalize fan engagement, support VIP hospitality and connect team operations, while an AI agent called TORO surfaces real-time insights and visualizations during race weekends. F1’s wider use of Salesforce already shows how AI agents help fans self-serve on complex topics like rule changes, driving 70–75% self-service adoption and cutting average handling time by 50%. This is B2B customer engagement with a twist: fans behave like high-value customers, and teams behave like complex B2B organizations. Meanwhile, Salesforce is pushing Agentforce deeper into quote-to-cash, contact center and revenue operations, backed by an aggressive acquisition streak that includes data, content and automation platforms. The goal is obvious: if AI agents can handle commerce and service, they can eventually orchestrate pricing, contracts, renewals and billing across the revenue lifecycle.
The $3.6B Fin bet shows Salesforce knows its weak spots
Salesforce’s agreement to acquire Fin for USD 3.6 billion (approx. RM16.6 billion) is the clearest admission that its homegrown agents were not good enough — yet. Fin brings an out-of-the-box AI customer service platform with more than 30,000 customers and reported end-to-end resolution rates around 76%, compared with Agentforce’s roughly 60% baseline. That gap matters in a market where autonomous AI service agents are moving from pilots to production and enterprises are tired of inflated ‘deflection’ metrics that ignore whether real work got done. One quotable takeaway from practitioners is that client conversations have flipped from 80% about contact center migration to 80% about AI, which underscores that Salesforce is chasing demand, not inventing it. The Fin deal, expected to close in Q4 of Salesforce’s fiscal 2027, is a high-stakes bet that better resolution rates plus outcome-based pricing will make Agentforce a credible alternative to rival platforms in financial automation and broader customer experience transformation.
Conclusion: Salesforce is redefining CRM around autonomous outcomes
Taken together, Agentforce Commerce, Help Agent, the VCARB partnership and the Fin acquisition confirm that Salesforce is trying to redefine CRM around autonomous outcomes, not logged activities. AI agents now sit inside ecommerce checkout flows, fan experiences, and multi-channel support, with outcome-based pricing tying revenue directly to resolved issues. At the same time, Salesforce’s FY2026 revenue of USD 41.5 billion (approx. RM191.1 billion) and Agentforce’s USD 1.2 billion (approx. RM5.5 billion) in ARR across 18,500 customers show that this is more than an experiment; it is a growth engine the company is already scaling. The open question is whether enterprises will trust AI agents with core revenue, compliance and service workflows. But the direction is clear: CRM as a passive database is fading, and Salesforce intends Agentforce to be the automation layer that actually does the work.
