Agentforce Commerce: From Chatbots to Always-On Digital Staff
Salesforce Agentforce Commerce is an ecommerce platform that embeds specialized AI agents into B2B and B2C workflows so they can independently handle shopping assistance, procurement, and merchant operations across web, apps, and messaging channels without relying on separate chatbots or disconnected admin tools.
The key takeaway is blunt: if you run ecommerce and still treat AI as a sidecar chatbot, you are about to be outpaced by merchants treating AI agents as digital staff. Salesforce has moved Agentforce Commerce into general availability with three “employees” in the box: Shopper Agent for customer experience, Buyer Agent for B2B procurement, and Merchant Agent for operational efficiency. These are wired into catalog, inventory, orders, and customer data from day one, not bolted onto the front end. The result is a credible alternative to other AI agents ecommerce platforms that promises actual automation, not another FAQ bot. As peak shopping season looms, this is less a feature release and more a line in the sand.
Shopper Agent: AI Customer Service Agents That Can Close the Sale
Shopper Agent is Salesforce’s answer to the question every CX leader keeps asking: can an AI customer service agent handle the entire purchase journey, not just deflect tickets? Salesforce claims this agent is transactional and lives up to the “personal shopper” idea by checking live inventory, confirming carrier cutoffs, offering store pickup, and closing the sale in one conversation. Because it carries full context and history into later service interactions, it also tackles the classic complaint of customers repeating themselves.
This is where the strategy gets interesting. Rather than ceding the transaction to discovery platforms like ChatGPT, Gemini, or social apps, Salesforce wants commerce to stay on owned properties. The AI sits on your site, your app, your messaging channels, while you remain the merchant of record and every order lands on the same platform that runs service, loyalty, and marketing. For retailers, the choice is now stark: either deploy capable AI agents on your own surfaces or watch third-party AI platforms reshape your customer relationship for you.
Buyer and Merchant Agents: B2B Commerce Automation Grows Up
On the B2B side, Agentforce Commerce is a clear signal that spreadsheets and portals are about to lose their central role. The Buyer Agent orchestrates the entire procurement process inside WhatsApp and SMS conversations, removing friction that has plagued B2B ecommerce for years. A buyer can simply text, “Need 40 cases of the 16-oz fasteners, same as the March order,” and the agent responds with product images to confirm the SKU, applies contract pricing, and completes the order—no portal login, no phone call.
The Merchant Agent, meanwhile, targets the operational mess behind the scenes. It lets B2B teams organize catalogs, create boost and bury rules, and tune sort orders in plain language instead of clicking through layered admin menus. This is B2B commerce automation that respects how teams actually work: talk to the system, describe the merchandising outcome, and let the AI handle the rule-building. For merchants facing shrinking headcount but rising complexity, these agents are less about novelty and more about survival.
Architecture and Adoption: Headless, Shared Context, and Platform Risk
Under the surface, Salesforce is betting that AI agents only deliver real value if they sit on top of shared context. Earlier in the year, the company introduced shared context as a design principle across its agentic commerce suite so a continuous conversation can span marketing, commerce, and service without losing data between systems. In this release, those agents are natively wired into inventory, order management, and customer data, which means a Shopper Agent can handle a service issue or confirm a delivery date without throwing the customer over the wall to another platform.
Technically, Agentforce Commerce can now run fully headless, so the same catalog, pricing, accounts, and Buyer Agent power any front end. Agentic Commerce Search, built from an acquired intent-aware engine, pushes discovery beyond keyword matching and can even be deployed on non-Salesforce storefronts. There is platform risk here: you are tying more of your customer journey to a single vendor. But with 78 of the Top 2000 online retailers already on Salesforce, representing more than 192.60 billion in web sales in 2025, the company clearly believes it can turn that installed base into an AI-first commerce ecosystem.
What Merchants Should Do Before Peak Season
Timing is not an accident. Salesforce has confirmed general availability of Shopper Agent, Buyer Agent, and Merchant Agent ahead of the coming shopping season, and is urging brands to have their Shopper Agent live on owned properties in time for that surge. Native integration into ChatGPT is already here, with Google Search and Gemini app integrations due in summer, along with guest order tracking and intelligent FAQs on product pages. This cadence is aimed squarely at retailers racing to automate customer interactions and order management before demand spikes.
So what now? First, treat these AI agents as roles, not widgets. Define where Shopper, Buyer, and Merchant Agents own outcomes—and where humans still must. Second, prepare your data; messy catalogs and inconsistent contract terms will cripple any AI. Third, pilot in one or two high-friction journeys, such as repeat B2B orders or last-mile delivery questions, and measure abandonment, handle time, and revenue impact. The merchants who win this peak season will be the ones who let AI agents do the repetitive work while humans focus on exceptions and strategy, not the ones waiting for a “perfect” future release.
