Getty Images/iStockphoto
E-commerce AI the focus of Salesforce Agentforce updates
Salesforce updates Agentforce Commerce with AI tools for personalization.
Salesforce's latest updates to Agentforce Commerce -- formerly Commerce Cloud -- inject AI-powered communications, personalization and purchasing tools into numerous shopping channels, including email, SMS text messaging and WhatsApp.
The idea, Salesforce said, is that most marketing communications, including text, email and messaging app messages, are currently one-way with, depending on the channel, "Do not reply" verbiage or email addresses. To continue shopping, the customer clicks on a link to claim an offer or discount code. New Salesforce features will enable customers to respond to these messages -- and AI agents can manage the conversations, create personalized offers, upsell when opportunities present themselves, and close the sale by taking payment and facilitating orders.
In concept, Salesforce's Two-Way Email, Two-Way SMS and Two-Way WhatsApp features could potentially convert marketing blasts into two-way conversations. In reality, the tools will help marketers manage upward of 1,000 or even 10,000 responses they receive after sending out a blast to hundreds of thousands of customers or more, said Liz Miller, analyst at Constellation Research.
In her experience as a marketer, some small percentage of customers typically reply to, say, a coupon email offer. Many of these people would simply say, "Thanks!" But some would want to negotiate the terms of a purchase, such as expanding the items it might cover. Others would be confused if they had received multiple offers and weren't sure how or if they could be combined. Still others needed more information before using the coupon.
The trouble is, on the scale of a million emails blasted, a couple of percentage points' response adds up to a lot of individual requests for a marketer or customer service team to address. Salesforce's two-way automation provides a means to feed those conversations to an AI agent, potentially scoring a few sales along the way.
"It's solving the manpower issue of sorting through 6,000 email replies to get to the five that matter to experience," Miller said. "If I'm a marketer, you know what the least important thing in my day is? Sorting through 6,000 emails. I've got other things that I've got to do. I've got to go drive growth. I've got to go make some business."
AI search, personalization features also on tap
Also in the works are AI-powered Agentforce Commerce tools that personalize buying experiences for users' customers over mobile and web channels:
- Agentforce Commerce Guided Shopping contains conversational analytics to determine an agent's action and can be enabled to offer promos and other pot-sweeteners to close sales.
- Contextual Search gives agents and customers more targeted results based on the ongoing customer dialog.
- Marketing and Commerce Connected Journeys plug customer data into triggered events such as an email signup or re-engagement after cart abandonment.
On the surface, Miller said, this batch of Agentforce Commerce features might appear to be incremental updates. However, using them effectively will make most retailers rethink processes and drive deeper collaboration between marketing and commerce teams.
"They are going to force a new strategy for marketers themselves," Miller said. "We have to create intentional strategies that deploy content within these spaces in very, very different ways. I don't think we're prepared for that."
Agentic AI is evolving so quickly that users need to stop and think about how it is rewriting tried-and-true processes.
"Marketers are still being bombarded with the promise of AI and Agentforce, and how you can have all these agents," Miller said. "It is a really good time to hit pause for just five seconds, step back and [consider] how, in a world where I can have a bidirectional conversation with email that leads a customer to a shifting contextual website that has guided search, what do I do to make sure that experience slaps every time?"
Some of the Agentforce Commerce features are available today as part of the Salesforce Winter '26 release, and others will become available in February. Contextual Search is in open beta.
Looking at holiday shopping data aggregated from Salesforce customers, Nitin Mangtani, senior vice president and general manager of Agentforce Commerce, said that consumers are taking their time to review more content about the things they're about to buy before making a purchase. In Salesforce's view, tools such as guided search and AI personalization might help retailers make the sale in those deciding moments prior to purchase.
"During the 2025 holiday season, shoppers spent 35% more time on brand sites before buying compared to the previous year, while traffic to AI-search channels like ChatGPT doubled," Mangtani said. "This signals that consumers are becoming more discerning; they have a growing need for more product information and reassurance on the path to purchase, making conversational commerce a critical lever for conversion."
Salesforce released and previewed its generative AI tools for retail at the National Retail Federation's NRF '26: Retail's Big Show.
Don Fluckinger is a senior news writer for Informa TechTarget. He covers customer experience, digital experience management and end-user computing. Got a tip? Email him.
