Two new product updates aim to help retailers use agent-based AI more effectively by improving shopper trust and making product detail pages (PDPs) easier for machines to understand.
Cimulate’s new agent trading platform combines traditional web search with newer discovery channels, including search engines and conversational commerce.
Cimulate’s CommerceGPT adds new features to enhance traditional web search and provides tools for digital teams to support two emerging purchase paths – off-site discovery in answers like ChatGPT, Perplexity and Claude, and on-site shopping through conversational commerce. Cimulate’s latest offerings include Human Feedback, Commerce AEO, and Co-Pilot Analytics.
“Digital commerce is no longer limited to search boxes, keywords and Google,” John Andrews, co-founder and CEO of Cimulate, told the E-Commerce Times. “Consumers shop through agents and expect conversational experiences on brand sites. They need a single platform that connects legacy and emerging digital channels, and that’s the gap we’re bridging with this release.”
Marissa Jones, vice president of product at Bazaarvoice, said PDP optimization is more than just a cosmetic refresh. Done right, it can increase conversions, build trust and increase revenue. This means formatting the PDP content to be machine readable by AI, not just human.
“Many PDPs still rely on content that looks rich visually but is effectively invisible to AI systems. Winning sellers have audited their PDPs to ensure ratings, reviews and Q&As are machine-readable and properly structured,” she told the E-Commerce Times.
Agent Shopping Platform Update
Cimulate has redesigned its platform to address three pain points arising from new business trends. The new features are reshaping web search for the modern store, providing better visibility and control over how products appear in agent systems, and allowing for more effective integration of AI shopping assistant conversations.
LLM-based search interprets natural language, context, and intent while reducing reliance on manual rules. The Human Feedback tool allows marketers to fine-tune AI’s understanding of meaning. With this feature, the platform complements the scalability of machine intelligence with the accuracy of human judgment.
While Cimulate’s Conversational Co-Pilot already allows shoppers to interact with an AI assistant that behaves like a seasoned in-store associate, Co-Pilot Analytics provides greater insight into how conversations work. The solution shows what customers are asking, which interactions are converting and why and how to improve AI shopping assistant conversions. These insights turn chat transcripts into a powerful new signal for marketers, digital marketers and brand leaders.
According to Andrews, Commerce AEO goes beyond traditional SEO. It focuses on how products appear in AI-answered search engines, not just search rankings. Rather than treating responses as a black box, AI analyzes the response signals that engines rely on to choose products to recommend—including how well the product matches the user’s intent, how clearly its attributes and use cases are described, and how credible and quotable its supporting resources are compared to the competition.
“When a competitor’s product is selected, the dashboard shows which product won, which factors mattered most in that response—for example, attributes, location, or authority—and where your product fell behind. The result is not just a ranking change, but a clear explanation of what the model preferred and helpful guidance on how to improve your product’s chances in future conversations,” he explained.
Fresh reviews help AI trust and rate
Bazaarvoice’s Jones suggests that brands need to take reviews as a live signal. Fresh, authentic reviews increase trust and AI ratings.
“Brands that update reviews quarterly see stronger customer engagement. For example, Iconic London reported a 126% increase in conversions and a 361% increase in time spent on site after targeted PDP optimization and UGC integration,” she said.
Jones offered as an example that global CPG brands refresh review volume quarterly using post-purchase emails and sampling programs to maintain relevance. Reviews older than six months consistently show lower trust and AI visibility than fresh content.
“In addition to LLM search, many retailers are now creating integrated merchant apps for agents that essentially display aggregated PDPs within the LLM environment. Product suggestions in these agent apps have also been shown to prioritize the latest content because they recognize this as a signal that consumers are actively shopping and that they like the products,” she explained.
Jones also recommended that brands invest in real shopper visuals to support multimodal AI. Real photos and short videos improve AI visibility and shopper conversion.
“PDP galleries with visual UGC achieved up to 250% increased time on site, 150% higher conversion rates and 15% increase in average order value. Target, for example, is a masterclass in scaling this approach across large catalogs,” she said.
For example, beauty and personal care brands saw measurable increases in add-to-cart rates by aggregating shopper photos and creator videos across DTC and PDP retail partners, giving both human vision and AI models better context.
Jones sees one of the most significant risks of AI as inconsistency. Strong PDPs on hero SKUs won’t make up for empty longlists. There is a need to standardize PDP quality across the entire product catalog. “Major food and beverage manufacturers are pushing core PDP standards. Think minimal checks, visual coverage and Q&A across each SKU so AI systems don’t see the catalog as fragmented,” she said.
Where product data continues to disrupt AI
Cimulate’s Co-Pilot Analytics component identifies gaps in product data when customers request features or use cases not currently documented in product descriptions. It listens for cues in the conversation that indicate missing or unclear information about the product, such as features, use cases, or limitations that the customer is interested in but cannot easily find.
“When that happens, it dynamically looks beyond the seller’s site to show the most relevant information at that moment, helping the customer move forward instead of getting stuck. Analytics then looks at that behavior in aggregate,” Andrews explained.
He noted that analyzing sentiment, engagement, and follow-up funnel signals helps assess whether responses actually satisfied customer intent or created friction.
“Coupled with insight into the entire purchase journey, this reveals consistent gaps in product descriptions or catalog structure. Merchants can then use these insights to enrich their catalog, clarify product data and proactively fill the gaps that matter most to customers,” he said.
What CommerceGPT doesn’t address — yet
Cimulate’s on-site assistant can answer many questions without forcing shoppers to click through multiple pages, creating a “zero-click” experience. However, it cannot guarantee that the brand will be credited or that the “buy” action will be integrated into the external AI agent’s response. This is not yet a use case the platform is targeting, Andrews noted.
However, it shows semantic clusters where products win or lose, he assured. CommerceGPT automatically uses semantic clustering and related behavior and conversion trends to produce results for similar searches and products.
“We currently expose our analytics on a query and product basis, but we can run explicit cluster reports for our clients if needed. Cluster-based dashboards are planned for late 2026,” Andrews said.
He went on to explain that Cimulate ensures product persona consistency by aligning both on-site search and external AI agents to the same product and location signals.
PDP visibility tips for 2026
Jones offered suggestions for improving the bottom line in the new year. One is optimizing for a “Triple-A” content framework: accessible, authentic, abundant. PDPs that perform best with AI-driven discovery consistently meet all three criteria.
For example, retailers with strong performance in AI-powered shopping environments ensure that user-generated content is widely distributed among retail partners, regularly updated, and clearly attributed to real shoppers. With new integrated shopping apps created by retailers to appear in LLM, it is important for brands to pool their UGC to retailers to ensure their products are well represented.
Her second suggestion is to align the PDP strategy with the emerging generative experience optimization (GEO). AI-driven discovery is moving from classic SEO to GEO. PDPs must be optimized for text, visuals and structured data together.
“Retailers who syndicate structured UGC feeds across major retail networks are seeing increased visibility with AI shopping assistants and higher-value shoppers who convert significantly faster than traditional search traffic,” she concluded.