Beyond Keywords: How Agentic Commerce Search Understands True Shopper Intent

Twenty years ago, I helped build search architectures that went on to dominate modern retail. Back then, the search bar trained consumers to strip away natural language and search for products using as few keywords as possible. Now? Consumers want to type exactly the way they talk. 

Salesforce research shows that 39% of consumers have already used AI to find products. And last holiday season, Salesforce estimated that AI- and agent-referred traffic drove 21% of all holiday orders globally, adding up to $263 billion in sales. Yet, on most ecommerce sites, the product search function is stuck in the early aughts. If a customer searches for an “outfit I can wear to a country concert on a hot day,” traditional search engines return a frustrating page with zero results. Search architecture that was initially built to decipher strict keywords can’t parse intent from natural language. 

In the rush to keep up with evolving expectations for product search, ecommerce and merchandising teams have been forced to manually create thousands of rigid rules and synonyms. But it doesn’t need to be this way.

Introducing Agentic Commerce Search: Built for the new era of commerce

I co-founded Cimulate in 2023 because I realized that AI was changing product discovery in fundamental ways and legacy search engines couldn’t meet the moment. In February, Salesforce officially announced the acquisition of Cimulate. Since joining Salesforce, we’ve already tripled our product investment. Agentic Commerce Search is now a top priority for the Agentforce Commerce roadmap.

What does Agentic Commerce Search mean for your business?

• For Salesforce B2C Commerce customers: You now have direct access to intent-aware search, tightly integrated with your core commerce platform. You can simply toggle it on, draw on your existing catalog data, and start optimizing. 
  
• For non-Salesforce customers: Whether you operate on Shopify, SAP, Adobe, Commercetools, or a homegrown stack, Agentic Commerce Search remains fully available to you via headless APIs.

Modern search features enhance the customer experience

Now that Agentic Commerce Search is fully integrated with B2C Commerce and Shopper Agent, merchants have full access to a complete, AI-native discovery suite designed to handle the entire shopping journey:

• Search: Transform your search bar into an AI-native search engine that decodes shopper intent over exact keywords.
  
• Browse: Continuously optimize your product category page rankings based on shopper behavior, catalog changes, and your specific business objectives.
  
• Recommendations: Keep customers engaged across your site with AI-driven algorithms that automatically adapt to the user journey and site location.
  
• Merchandising Rules: A robust toolset for strategically curating results toward key partnerships and business objectives.
  
• Shopper Agent: Deliver rich, conversational shopping experiences natively within our Shopper Agent. Agentic Commerce Search is now integrated into Shopper Agent, bringing the most powerful digital concierge to the market and giving shoppers a richer experience. (Requires Agentforce Shopper Agent)

Key benefits of Agentic Commerce Search

Agentic Commerce Search is an AI-native search and discovery engine purpose-built for the complex demands of modern retail. Instead of merely matching keywords, it natively interprets natural queries and brings context-aware shopping experiences directly to your site. 

• Understand True Shopper Intent: Instead of relying on a massive volume of historical clickstream data, Agentic Commerce Search simulates millions of shopping journeys across your catalog to build a custom Small Language Model (SLM) for every customer. This gives every customer 10 times the behavioral signal, creating a massive data advantage. The engine easily understands complex conversational language, slang, cultural references, and even typos.
  
• Eliminate Manual Merchandising Overhead: Stop babysitting your search bar. By replacing rigid rules with an AI-native framework, our customers see an average 85% reduction in manual merchandising rules. Your merchandising team can step away from reactive data fixes and focus on high-level strategic growth.
  
• Drive Immediate Revenue Growth: When shoppers find exactly what they want, they buy. With Agentic Commerce Search, customers have seen improvements in the range of a 33% increase in clicks per visit, 17% increase in add-to-carts, and a 12% boost in revenue per user.
  
• Faster Time to Value, Lower Total Cost of Ownership: Unlike third-party search solutions that require lengthy implementations and ongoing licensing costs, Agentic Commerce Search is natively integrated with B2C Commerce, so you can get up and running in days, not months. By eliminating thousands of manual merchandising rules and replacing expensive point solutions with a platform-native capability, SFCC customers in particular see a significant reduction in the total cost of managing search.

How Agentic Commerce Search closes the data gap

Most brands don’t have the traffic and clickstream scale of retail giants like Amazon, which means AI often underperforms based on limited data. That’s why we developed a proprietary method for simulating millions of shopping journeys on a customer’s catalog. This simulation builds synthetic data on what converts and then closes the data gap. Now, our customers can leverage the same data advantage that was once reserved for retail behemoths — without requiring the same traffic or scale.

Agentic Commerce Search builds a custom small language model for each retailer using their specific catalog, behavioral data, and synthetic training data. Then, we use general-purpose LLMs like OpenAI, Claude, and Gemini to simulate search, browse, click, and conversion-style event data across the retailer’s catalog. Ultimately, this allows us to produce 10-20 times more commerce-specific training data for each retailer’s model.

Agentic Commerce Search creates an AI-native layer for commerce that builds intelligence based on what converts. Ultimately, this allows us to deliver contextualized search results that surface what sells on your site.

Customer highlight: West Marine

Here’s a scenario that illustrates how Agentic Commerce Search works at West Marine, the premier retailer of boating, fishing, and marine gear.

West Marine is known for its highly knowledgeable in-store associates who are equipped to troubleshoot problems and get customers exactly what they need. Replicating that elite, consultative experience online was a major priority.

Boating is a highly specific lifestyle. Maintaining a vessel requires precise parts, and if you’re new to the water, figuring out exactly what you need to buy can be incredibly intimidating. Imagine a first-time boat owner landing on the West Marine site and searching for “boat oil.” Traditional search tools rely on exact part numbers or serve up a page of generic marine lubricants, leaving a new boat owner completely lost.

To solve this, West Marine used Agentic Commerce Search to power search, recommendations, and their digital associate, Skipper, effectively translating their in-store expertise to ecommerce. Because the AI understands natural intent and the deep nuances of West Marine’s catalog, Skipper acts just like a knowledgeable crew member. Instead of forcing the shopper to know exact technical specs, Skipper bypasses literal keyword matches and guides the boater to the precise marine-grade oil required for their specific engine. With Agentic Commerce Search, West Marine eliminated thousands of manual rules and doubled conversion rates on their site.

According to Deanna Christy, Senior Director of Ecommerce and Digital Experience at West Marine, “When we onboarded our previous search tool, the sheer volume of manual rules and configurations we had to manage behind the scenes was exhausting. When we moved to Agentic Commerce Search, I said, ‘This is AI, let the AI work.’ Now, the AI does everything for us—we no longer have to manually configure search and recommendations or constantly touch the system, freeing up our team’s time. We saw an uplift starting on day one and the results continue to improve.”

Partners lead the way with AI-native search

We’re working side-by-side with industry-leading partners like
Arkhive Digital, Astound Digital, Bounteous, Capgemini, CloudWerx, Deloitte, Deloitte UK, DemandPDX, EPAM, EveryMind, HCG, Merkle, OSF Digital, Publicis Sapient, Slalom, Solu, Studio Science, The Commerce Team, The Maze Group, and more.. These partners will play a key role in helping customers implement Agentic Commerce Search.  Here’s what our partners are saying:

“At Maze, we’ve seen Cimulate (now Agentic Commerce Search) move the needle in ways that rules-based search never could. Double-digit lifts in conversion, add-to-cart rate, and revenue per user aren’t rounding errors – they’re the difference between a good quarter and a great one. Having implemented Cimulate across both SFCC and Shopify for clients like Boot Barn and LoveShackFancy, we’ve seen search finally evolve from matching keywords to truly understanding customer intent,” said Vince Santo, CEO at Maze.

“Search has been a pain point for clients for over a decade and the answer was always to bolt on another third party,” shared Helena Shannon, CEO at Arkhive Digital. “Cimulate brought a true enterprise, AI-first search to market, one that reads intent and makes the storefront itself more powerful, and Salesforce building that into the platform is exactly the right direction. Enterprise teams are looking for fewer vendors, not more, and they expect more out of the platforms they already run. Agentic Commerce Search delivers that.”

“We’re thrilled to see Agentic Commerce Search become an integrated part of Agentforce Commerce, another exciting step forward in Slalom’s partnership with Salesforce,” shared Rob Pontious, Director of Commerce & Retail at Slalom. “Our teams are equipped and ready to help customers bring AI-native discovery to their storefronts, create more relevant shopping experiences, and unlock new ways for agents to drive commerce.”

Ryan Dowling, CRO at Astound Digital, says, “We’re excited about Salesforce’s continued investment in the future of commerce and retail. By bringing Cimulate into the ecosystem, Salesforce is delivering a powerful, semantic search option right out of the box. We’re encouraged by the reduced total cost of ownership for merchants and the potential improvements to the shopper experience.”

Turn conversations into conversions with Agentforce Commerce

Agentic Commerce Search closes the data gap, eliminates the manual overhead that’s been slowing your teams down, and turns every search query into a signal your business can act on. Whether your shoppers are typing conversationally, browsing with an AI agent, or somewhere in between, your search bar should be smart enough to keep up. The future of search is agentic. And it’s available now.

Ready to transform your search bar into an autonomous, intent-driven revenue generator?