Particular Audience

Report: Retailers Must Own Retail Media AI Decisioning

Particular Audience published an industry guide, "Retail Media AI Architecture: From Prediction to Decisioning," arguing retailers have roughly two years to establish control over the AI decisioning layer that will determine how products are surfaced and monetised inside LLM-driven commerce experiences. The guide defines five architectural layers (Causal Measurement, Predictive Models, Decisioning Systems, Generative Interfaces, Agentic Orchestration), warns that exposing raw catalogues to third‑party LLMs risks retailers becoming inventory in others' auctions, and recommends building or buying a decisioning/control layer and participating in open standards such as the Model Context Protocol (MCP). The article includes quotes from Particular Audience CEO and founder James Taylor and background on Particular Audience’s positioning and offices.

Particular Audience Promotes Langton McCombe to Sales Director, North America

Particular Audience promoted Langton McCombe to Sales Director, North America. McCombe joined the company in 2019 and has progressed through multiple internal roles; he relocated from Sydney to Vancouver to lead Particular Audience’s US market entry during the COVID-19 pandemic. In his new role he will manage commercial and client delivery efforts across North America to grow retail partners’ revenue and adoption of PA’s personalization and retail media solutions. Particular Audience positions itself as an AI-native retail media and personalisation platform, operating DiscoveryOS and working with retailers including Target, The Good Guys, Petbarn, Hamleys and Hotel Chocolat. The company supports open standards such as the Model Context Protocol (MCP).

Particular Audience Unveils Game-Changing AI Search Testing Tool

Particular Audience announced Search Model A/B Testing, a new capability that lets retailers A/B test and govern the AI models driving on-site search and sponsored results. The feature extends the company’s Adaptive Transformer Search (ATS), introduced in 2023, which the vendor says reduced zero-result searches for clients to below 0.5% (industry average ~20%). The new release enables retailers to compare model architectures, sentence embedding structures, foundational LLMs, and reinforcement-learning training data (real and synthetic) against commercial metrics such as conversion, average order value, margin contribution and zero-result rate. Particular Audience positions the capability as a way to unify organic and sponsored relevance, create new monetizable long-tail inventory for Retail Media Networks, and move retail media toward intent-level optimization rather than keyword bidding. James Taylor, Founder & CEO, framed the change as turning an opaque relevance algorithm into a governed, testable decisioning layer.