Two new product updates aim to help retailers use agentic AI more effectively by improving shopper trust and making product detail pages (PDPs) easier for machines to understand.
A new agentic commerce platform from Cimulate connects traditional site search with newer discovery channels, including answer engines and conversational commerce.
Cimulate’s CommerceGPT adds new features to improve the traditional site search experience and give digital teams tools to support two emerging shopping paths — offsite discovery in answer engines like ChatGPT, Perplexity, and Claude, and onsite shopping via conversational commerce. Cimulate’s latest offerings include Human Feedback, Commerce AEO, and Co-Pilot Analytics.
“Digital commerce is no longer confined to a search bar, keywords, and Google,” John Andrews, co-founder and CEO of Cimulate, told the E-Commerce Times. “Consumers are shopping through agents and expecting conversational experiences on brand sites. They need a single platform that connects old and emerging digital channels, and that’s the gap we’re bridging with this release.”
Marissa Jones, SVP of product at Bazaarvoice, said PDP optimization is more than a cosmetic refresh. Done right, it can drive conversions, build trust, and boost revenue. That means formatting PDP content so it’s machine-readable for AI, not just human-readable.
“Many PDPs still rely on content that looks rich visually but is effectively invisible to AI systems. Retailers that are winning have audited their PDPs to ensure ratings, reviews, and Q&A are machine-readable and properly structured,” she told the E-Commerce Times.
Platform Updates for Agentic Shopping
Table of Contents
Cimulate redesigned its platform to address three pain points arising from new commerce trends. New features reinvent site search for modern commerce, provide better insight and control into how products appear within agentic systems, and enable more efficient integration of AI shopping assistant conversations.
LLM-powered search interprets natural language, context, and intent while reducing reliance on manual rules. The Human Feedback tool allows merchandisers to tune AI’s understanding of relevance. With this feature, the platform complements the scalability of machine intelligence with the precision of human judgment.
While Cimulate’s Conversational Co-Pilot already enables shoppers to engage with an AI assistant that acts like a seasoned in-store associate, the Co-Pilot Analytics feature provides greater visibility into how conversations perform. The solution showcases what customers are asking, which interactions convert and why, and how to improve AI shopping assistant conversions. The insights turn chat transcripts into a powerful new signal for merchandisers, digital marketers, and brand leaders.
According to Andrews, the Commerce AEO tool goes beyond traditional SEO. It focuses on how products surface in AI answer engines, not just search rankings. Instead of treating AI answers as a black box, it analyzes the signals answer engines rely on when choosing which products to recommend — including how well a product matches user intent, how clearly its attributes and use cases are described, and how trustworthy and citable its supporting sources are compared to competitors.
“When a competitor’s product is chosen, the dashboard shows which product won, what factors mattered most in that response — for example, attributes, positioning, or authority — and where your product fell short. The result isn’t just a rank change, but a clear explanation of what the model preferred and actionable guidance on how to improve your product’s chances in future conversations,” he explained.
Fresh Reviews Help AI Trust and Ranking
Bazaarvoice’s Jones suggested that brands must treat reviews as a living signal. Fresh, authentic reviews drive both trust and AI ranking.
“Brands that refresh reviews quarterly see stronger shopper engagement. For instance, Iconic London reported a 126% increase in conversions and a 361% rise in time spent on site after focused 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 weaker trust and lower AI visibility than fresh content.
“Beyond just LLM search, many retailers are now building integrated agentic commerce apps that essentially display a summarized PDP within the LLM experience. The product suggestions in those agentic apps have also been shown to favor recent content, as it recognizes this as a signal that consumers are actively purchasing and liking these products,” she explained.
Jones also recommended that brands invest in real shopper visuals to fuel multimodal AI. Real-life photos and short videos improve AI visibility and shopper conversion.
“PDP galleries with visual UGC have driven up to a 250% increase in time on site, 150% higher conversion rates, and a 15% lift 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 have seen measurable lifts in add-to-cart rates by syndicating shopper photos and creator videos across DTC and retail partner PDPs, giving both humans and AI vision models better context.
Jones sees one of the most significant AI risks as inconsistency. Strong PDPs on hero SKUs don’t compensate for empty long-tail listings. What is needed is standardizing PDP quality across the entire product catalog. “Large food and beverage manufacturers enforce baseline PDP standards. Think minimum review counts, visual coverage, and Q&A across every SKU, so AI systems don’t view the catalog as fragmented,” she said.
Where Product Data Still Breaks AI
Cimulate’s Co-Pilot Analytics component identifies product data gaps when customers request features or use cases that are not currently documented in product descriptions. It listens for cues in the conversation that indicate missing or unclear product information, such as features, use cases, or constraints the customer cares about but cannot easily find.
“When that happens, it dynamically looks beyond the retailer’s site to surface the most relevant information in the moment, helping the customer move forward instead of getting stuck. Analytics then looks at this behavior in aggregate,” Andrews explained.
He noted that analyzing sentiment, engagement, and downstream funnel signals helps assess whether the responses actually satisfied customer intent or introduced friction.
“Combined with visibility into the full 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 Solve — Yet
Cimulate’s onsite assistant can answer many questions without forcing shoppers to click through multiple pages, creating a “zero-click” experience. Still, it cannot guarantee that a brand will be credited or that a “buy” action will be integrated into that external AI agent’s response. That is not a use case the platform addresses yet, noted Andrews.
However, it does show the semantic clusters where products are winning or losing, he assured. CommerceGPT automatically leverages semantic clustering and related behavioral and conversion trends to drive results for similar searches and products.
“We currently expose our analytics on a per query, per product basis, but can run explicit cluster reports for our clients where desired. Cluster-based dashboards are on our roadmap for late 2026,” Andrews said.
He continued, explaining that Cimulate ensures product persona consistency by aligning both onsite search and external AI agents on the same product signals and positioning.
PDP Visibility Tips for 2026
Jones offered suggestions to improve the bottom line in the New Year. One is to optimize for the “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-assisted shopping environments ensure user-generated content is widely syndicated across retail partners, updated regularly, and clearly attributable to real shoppers. With new, integrated agentic shopping apps built by retailers to appear in LLMs, it is important that brands syndicate their UGC to retailers to ensure their products are well represented.
Her second suggestion is to align the PDP strategy with emerging generative experience optimization (GEO). AI-driven discovery is shifting from classic SEO to GEO. PDPs must be optimized for text, visuals, and structured data together.
“Retailers that syndicate structured UGC feeds across large retail networks are seeing stronger visibility in AI shopping assistants and higher-value shoppers who convert at meaningfully higher rates than traditional search traffic,” she concluded.
