Pacvue today announced Pacvue Prism, a commerce-native platform the company describes as the industry's first Agentic Commerce Grid, designed to replace channel-level budget management with goal-level media accountability across every touchpoint a consumer encounters.

The announcement came on June 22, 2026, from Cannes, France, where the advertising industry gathers annually for the Cannes Lions festival. Pacvue, which the company describes as powering over $12 billion in advertising spend across more than 100 global retail media networks, is positioning Prism as a structural answer to a problem that has grown more acute as the commerce customer journey has fragmented beyond what legacy tools were designed to handle.

What is the Agentic Commerce Grid

At the center of the announcement is a concept Pacvue is calling the Agentic Commerce Grid: a brand's unified view of media performance where every channel connects to underlying commerce signals and ties to verifiable outcome metrics. According to Pacvue, the grid is built on each brand's own data, configured to their channel mix, and optimized for their business objectives. Pacvue Prism is the platform purpose-built to power it.

The ambition is explicit. Pacvue is not simply adding channels to a dashboard. The company is arguing that the fundamental model of channel-level budget allocation - where a brand assigns separate dollars to search, retail media, social, and CTV, then measures each in isolation - is structurally broken. The Agentic Commerce Grid proposes to replace that model with a single connected layer that ties all spend to the same underlying commerce outcomes.

"Commerce media is moving from channel optimization to commerce accountability," said Rahul Choraria, CEO of Pacvue. "The brands that win from here will operate at the grid level: every touchpoint connected, every dollar accountable. The Agentic Commerce Grid, powered by Pacvue Prism, will become the lens through which every CMO evaluates their business."

The problem Pacvue is solving

The customer journey no longer follows a predictable path, according to Pacvue. People discover products on social, research through traditional search or large language models, engage through streaming, and buy through a retailer or a social shop - often within the same session. Existing marketing tools were built for a world where awareness, consideration, and conversion were separate stages owned by separate teams. The result, according to Pacvue, is fragmented measurement, siloed budgets, and an inability to answer a basic question: what is any given unit of spend actually driving?

This is not a new diagnosis. The fragmentation problem in retail media has been documented extensively, with IAB Europe releasing incremental measurement guidelines in November 2025 specifically to address a situation where brands working across multiple retail networks could not compare results in consistent terms. Commerce media spending is projected to reach $100 billion by 2028, yet Koddi research found that 63% of commerce media networks cite measurement as their biggest barrier, while only 12% can seamlessly activate and measure campaigns across onsite, offsite, and in-store environments simultaneously.

What makes the fragmentation problem harder to solve is the rate at which new channels are entering the planning conversation. The Trade Desk integrated with Pacvue and Skai for retail media in Q1 2026. Pinterest, now controlling CTV performance measurement through its acquisition of tvScientific, became part of the Prism alpha channel set. ChatGPT has entered the media planning conversation as an advertising surface. Each new channel adds a layer of complexity rather than reducing it.

Technical architecture of Pacvue Prism

Pacvue Prism is built around five stated capabilities that together define what the Agentic Commerce Grid does in practice.

First, brands can build and operate an Agentic Commerce Grid as a brand asset, configured to their specific commerce signals, channel mix, and chosen outcome metrics. This is the planning and governance layer. Rather than inheriting a default measurement model from each platform, brands configure the grid according to what their business actually cares about.

Second, Prism deploys a unified agentic infrastructure to plan across every channel, optimizing toward the brand's business goals and continuously improving as results come in. This is the execution layer. Agentic, in Pacvue's framing, means the system can act on the plan - not just report on it - using automated decision-making within guardrails brands define.

Third, the platform makes media accountable through what Pacvue calls proprietary halo measurement. This is built on regression modeling combined with custom attribution and pixel tracking, connecting each media investment to a measurable and relevant outcome. Halo measurement is a term of art in retail media, referring to the ability to capture sales impact that extends beyond the directly attributed conversion - for instance, whether a CTV campaign for a brand lifts sales of that brand's full product line at a retailer, not just the specific product advertised.

Fourth, Prism compounds performance over time by connecting to proprietary tech stacks, including data lakes, data warehouses, CDPs, CRMs, and custom AI agents via the Model Context Protocol. This is the data infrastructure layer. MCP - the open standard originally developed by Anthropic - enables AI models to communicate with external tools and data sources through a standardized interface. Pacvue has been building toward this architecture since May 14, 2026, when it launched its own MCP server connecting its retail media data to ChatGPT, Copilot, Gemini, and Claude. The MCP connection means that every new signal entering Pacvue Prism can feed back into future decisions, rather than sitting in a separate warehouse waiting for a human analyst to pull it.

Fifth, brands activate across search, programmatic, social, and new-frontier formats - including conversational and shoppable - through a unified Pacvue platform with an agentic experience.

Integration landscape

The alpha launch of Pacvue Prism includes several integrations that span the full spectrum from established formats to emerging surfaces.

Existing integrations, confirmed in the announcement, include ChatGPT and The Trade Desk. The ChatGPT integration specifically extends from Pacvue's announcement on May 5, 2026, that it had become a technology partner for ads in ChatGPT, naming Kepler as its inaugural agency partner.

New integrations added for the alpha include Pinterest, tvScientific by Pinterest, Firework, and Shopsense AI. Pinterest and tvScientific by Pinterest represent a combined entry into social discovery and CTV performance. As PPC Land reported, Pinterest is simultaneously pursuing its own MCP strategy, with Pacvue listed among the alpha partners for Pinterest's Business Assistant and MCP rollout announced on June 17, 2026. That overlap is notable: Pacvue is not only integrating Pinterest's ad inventory into Prism but is also part of the architecture through which Pinterest intends to distribute its own AI-powered planning tools.

The Firework integration handles shoppable video. Through this connection, Lowe's can link brands to shoppable video inventory, with those formats connected directly to retail product detail pages. This is a significant capability because it closes a loop that has historically been left open in video advertising: a consumer sees a product in a video and the path to purchase is still a detour, involving a search, a new browser tab, and a separate shopping flow. Shopsense AI adds AI-powered shopping functionality from within streaming content.

Yahoo DSP is listed as an additional integration that is underway, though not yet available at alpha launch.

Early performance figures

Alpha participants are actively testing Pacvue Prism and have reported early results, particularly on investments that historically optimized for reach and impressions rather than commerce outcomes. According to Pacvue, those alpha participants have seen: plus 34% impressions, minus 23% CPM, plus 12% total sales, and plus 24% new-to-brand growth.

These figures require careful reading. Pacvue does not disclose the methodology behind the numbers, the size or nature of the alpha participants, the category verticals involved, the time period of measurement, or what baseline the figures compare against. The pattern of strong alpha metrics is consistent with Pacvue's previous product launches: Pacvue Agent, announced April 14, 2026, carried figures including workflows up to 200x faster, 80x quicker time to insight, and 54% improvement in campaign performance - figures that also came from early adopters without disclosed methodology.

What the alpha metrics do suggest, even without methodology disclosure, is that brands moving budgets away from channel-level optimization toward outcome-level planning are seeing shifts in cost-per-impression alongside increases in total sales. The minus 23% CPM alongside plus 12% total sales combination, if validated at scale, would indicate that the reallocation of budget toward higher-signal commerce inventory generates both efficiency and volume gains.

Measurement architecture

One of the more technically specific elements of the Prism announcement is the measurement layer. Pacvue describes it as proprietary halo measurement built on regression modeling in addition to custom attribution and pixel tracking.

Regression modeling in this context refers to econometric methods - marketing mix modeling approaches that estimate the causal contribution of each media input to observed sales outputs. These methods have been gaining industry attention precisely because they address a limitation of last-touch attribution: the inability to capture the influence of channels like CTV or brand social on purchases that happen elsewhere, sometimes weeks later.

The IAB has been actively working on standardizing incrementality measurement in commerce media since November 2025, when it released guidelines defining incrementality as the causal impact of marketing by identifying additional business outcomes directly driven by a campaign or tactic, compared to what would have occurred without that marketing activity. Pacvue's halo measurement architecture is directly addressing the same gap those IAB guidelines were written to solve.

Custom attribution and pixel tracking complement the regression modeling, providing direct connection of media investments to outcomes at the individual touchpoint level. Together, the three layers - regression, custom attribution, and pixel tracking - are designed to cover the measurement gaps that exist when any single method is used alone.

Commercial positioning and competitive context

Pacvue describes itself as the leading Commerce Media OS, a designation that positions it differently from category peers. Skai (formerly Kenshoo) announced its agent-native operating system in May 2026, using nearly identical language to describe a unified environment for agentic campaign execution across retail media and beyond. The two companies serve overlapping enterprise advertiser bases and are competing directly on the promise of cross-channel commerce intelligence.

The operative question in this competition is whether the Agentic Commerce Grid concept - connecting every touchpoint to commerce signals in a single governed environment - is technically deliverable at enterprise scale, or whether the data volume, identity fragmentation, and integration complexity make it a roadmap aspiration rather than a current product reality. Pacvue has not disclosed the number of brands in the alpha program, the integration depth with each partner channel, or the data governance mechanisms governing how commerce signals flow between proprietary tech stacks and the Prism platform.

According to Nic Jones, Director of Media at PetIQ, the goal-level framing is genuinely significant: "I'm excited all over again by seeing Pacvue Prism, because most organizations assign budgets at the channel level, but in the long term, this could eliminate that process. You're no longer dealing with channel level budgets. Now you're dealing with what is the goal of these dollars. That in itself is really powerful."

Jason Colon, Executive Vice President of Commerce Media at Horizon Commerce, framed the appeal differently, focusing on operational complexity: "Brands are managing more channels, more signals, and more complexity than ever before, and they need technology that enables better orchestration, faster decision-making, and more agile execution."

Co-Founder and President Melissa Burdick articulated the thesis from Pacvue's side: "For years, marketers have managed channels. The next era is about managing outcomes. As discovery fragments across social, retail media, streaming, and AI, brands need an intelligence layer that connects every touchpoint to commerce. That's what Pacvue Prism was built to do."

What it means for the marketing community

The marketing and advertising community has been watching fragmentation intensify over the past 18 months. The agentic AI buildout in ad tech has produced a dense cluster of product launches in the first half of 2026 alone. Yahoo DSP, DoubleVerify, LiveRamp, MiQ, Mediaocean, and others have all released agentic planning or execution infrastructure in recent weeks. The pattern is consistent: every major layer of the ad stack is attempting to position its own data as the intelligence layer through which automated decisions flow.

Pacvue Prism sits at a particular intersection of this buildout. It is not a media owner, a DSP, or a data clean room. It is a managed environment for enterprise advertisers and agencies, built on top of retail media infrastructure that already handles a claimed 12% of global retail media ad spend outside China. That scale gives Prism's connected channel coverage commercial weight that a newer entrant building the same architecture from scratch would not have.

The inclusion of ChatGPT and conversational formats alongside traditional retail media channels is also a signal of where Pacvue sees the immediate competitive frontier. PPC Land has tracked the ChatGPT advertising infrastructure since the pilot began with a closed roster in February 2026. By May 5, OpenAI had opened its self-serve Ads Manager to all US businesses and named Pacvue as one of a short list of technology partners. Placing ChatGPT within the Agentic Commerce Grid is an assertion that conversational AI advertising is now mature enough to sit alongside Amazon, Walmart, and Pinterest in a unified commerce planning environment.

Whether it performs at that level is what the alpha and subsequent broader availability will test. Pacvue Prism is available today to enterprise brands and agencies.

Timeline

  • November 2025 - IAB Europe releases incremental measurement guidelines for commerce media, addressing fragmentation across retail networks that Pacvue Prism is designed to solve (PPC Land)
  • December 11, 2025 - Pinterest announces definitive agreement to acquire tvScientific, the CTV performance platform now integrated with Pacvue Prism (PPC Land)
  • March 17, 2026 - Pacvue announces partnership bringing Reddit Ads into its Commerce Operating System, part of the platform expansion leading to Prism (PPC Land)
  • April 14, 2026 - Pacvue launches Pacvue Agent, its AI agentic layer for Amazon Ads workflows, promising 200x faster execution - a precursor to the Prism launch (PPC Land)
  • April 27, 2026 - Walmart Connect announces Connect Select with Pacvue and Skai as technology integration partners, confirming Pacvue's position in programmatic CTV (PPC Land)
  • May 5, 2026 - Pacvue announces it is a technology partner for ads in ChatGPT, with Kepler as its inaugural agency partner - an integration now incorporated into Prism (PPC Land)
  • May 7, 2026 - The Trade Desk Q1 2026 results confirm Pacvue and Skai retail media integrations within the platform, which is named as an existing Prism integration (PPC Land)
  • May 13, 2026 - Skai announces its agent-native operating system at ShopAble 2026, the most direct competitive framing to Pacvue Prism's approach (PPC Land)
  • May 14, 2026 - Pacvue launches its MCP server, enabling retail media data access from ChatGPT, Copilot, Gemini, and Claude - the MCP infrastructure that underpins Prism's data compounding capability (PPC Land)
  • June 17, 2026 - Pinterest announces its MCP server with Pacvue as an alpha partner, alongside its Business Assistant closed beta (PPC Land)
  • June 22, 2026 - Pacvue announces Pacvue Prism at Cannes Lions 2026, launching the Agentic Commerce Grid platform to enterprise brands and agencies with alpha integrations including ChatGPT, The Trade Desk, Pinterest, tvScientific by Pinterest, Firework, and Shopsense AI; Yahoo DSP integration underway

Summary

Who: Pacvue, the commerce media platform founded in 2018 and headquartered across Seattle, Los Angeles, Chicago, New York, Washington D.C., London, Shanghai, and Tokyo. The company claims to power over $12 billion in advertising spend and 12% of total retail media ad spend worldwide (excluding China). Key executives named in the announcement include Rahul Choraria (CEO), Melissa Burdick (Co-Founder and President), and early client stakeholders from Horizon Commerce and PetIQ.

What: The launch of Pacvue Prism, a commerce-native platform that creates what Pacvue calls the Agentic Commerce Grid - a unified, configurable view of media performance connecting every channel to underlying commerce signals. The platform integrates retail media, search, programmatic, social, and new-frontier formats including conversational AI and shoppable video. Alpha participants report early results including plus 34% impressions, minus 23% CPM, plus 12% total sales, and plus 24% new-to-brand growth.

When: Announced June 22, 2026, at Cannes Lions. Alpha integrations at launch include ChatGPT, The Trade Desk, Pinterest, tvScientific by Pinterest, Firework, and Shopsense AI. Yahoo DSP integration is underway. Pacvue Prism is available today to enterprise brands and agencies.

Where: The announcement was made in Cannes, France. The platform is available globally to enterprise brands and agencies, accessible at pacvue.com/platform/prism.

Why: The modern customer journey fragments across social discovery, LLM research, streaming, and retailer checkout - often within a single session. Existing tools were built for siloed stages and separate teams, producing fragmented measurement and budget allocation by channel rather than by outcome. Pacvue Prism is positioned to replace that model by making every media dollar accountable to verifiable commerce outcomes, using agentic infrastructure, regression-based halo measurement, MCP connectivity to proprietary data stacks, and unified activation across all channel formats from a single platform.