CRM and CDP Vendors Put Customer Data Inside Enterprise AI

By Daniel Okafor

Amperity said its proof of concept for Stellantis ingested millions of records in minutes and created a unified Customer 360 in less than three days.

It anchors a week of customer-data announcements from Cordial, Amperity, Oracle and ZoomInfo. The disclosure gives no record mix, sample size or independent validation; the result measures processing speed and profile creation, not campaign performance.

What changes for marketers and sales teams?

Cordial launched a headless customer data platform that lets marketers explore and activate real-time customer data inside AI assistants. It positions the CDP as an open decisioning layer rather than a system that forces marketers into one interface. Named destinations are Claude, ChatGPT, Microsoft Copilot and custom agents.

The platform is designed to connect customer, product, message and performance data across:

• Google Analytics, Adobe Analytics and Shopify
• Salesforce Commerce Cloud, Databricks and Snowflake

For a small marketing team already working in those assistants, the proposed benefit is access to customer context without a parallel CDP screen and a shorter route from campaign planning to activation. Cordial disclosed no deployment-effort, time-saving or conversion figures.

Cordial says Boot Barn and Torrid use the platform to identify opportunities, recommend next-best actions and accelerate campaign planning. Those are customer examples, not performance benchmarks.

CEO Jeremy Swift described the product as an alternative to “legacy marketing clouds” that tether users to their own interfaces. “Cordial’s headless CDP puts real-time customer data where work happens,” he said.

ZoomInfo, meanwhile, plans to bring sales intelligence into Snowflake’s AI Data Cloud. For a small sales team already using that platform, the relevant test is whether the announced intelligence becomes usable in account research and early-stage prospecting. The announcement identifies no affected sales stage, timetable or productivity gain.

Stellantis tests a customer view for years-long relationships

Stellantis selected Amperity’s AI-powered Customer Context Platform for its customer-data transformation and a goal of more personalized experiences across brands and touchpoints. The platform will unify fragmented customer data, resolve identities and help Stellantis understand customers across vehicles, households, brands, dealers, digital interactions and ownership journeys.

The automotive relationship can span years and cross multiple dealers, service visits, financing interactions, connected-vehicle signals and brand touchpoints. For automotive marketing teams, that makes a persistent customer identity relevant to after-sales journeys, loyalty and cross-brand engagement, as well as audience building.

Amperity says the platform will support:

• Audience building and cross-brand engagement
• Campaign measurement
• After-sales journeys and loyalty
• Real-time responses to customer signals

Founder and co-CEO Kabir Shahani said the technology would help Stellantis turn “fragmented signals into real-time understanding” for teams and AI agents. The proof of concept demonstrates ingestion and unification speed. It does not show personalization lift, conversion, revenue or agent accuracy, and Stellantis has announced planned use cases rather than deployed results.

Oracle targets shared context for enterprise agents

Oracle announced new Oracle AI Data Platform capabilities intended to give enterprise AI agents a shared understanding of business entities, relationships, policies and workflows. The company did not publish pricing, availability dates, customer names or benchmarks.

For a large sales organization running agents across marketing, service and analytics, the practical implication is shared business rules rather than agents working from isolated records. That could matter at account planning, service renewal and campaign execution, where conflicting customer information can produce conflicting recommendations. Oracle’s announcement establishes an architectural aim, not proof that agents will make better decisions.

Across the four announcements, the measurable claims remain limited: Amperity reports ingestion of millions of records in minutes and a unified customer view in under three days. The rest are product intentions, selected integrations and named use cases.

Buyers should ask what data each system can access, how identity is governed, how outputs are measured and when production access begins. The next proof point is production: real-time activation inside assistants, consistent identities across brands and agents following shared business policies.