LiveRamp introduces agentic AI for marketing automation: orchestration, NL segmentation, and marketplace search
On October 1, 2025, LiveRamp rolled out three AI capabilities built for real marketing work: agentic orchestration, AI-Powered Segmentation, and AI-Powered Search for its Data Marketplace. The company positions itself as the first data collaboration platform to give autonomous agents governed access to identity, segmentation, activation, measurement, clean rooms, and a 900-partner network.
Translation for teams: fewer manual steps, faster execution, and stronger use of first-, second-, and third-party data-without breaking enterprise guardrails. Customers can enable these features under their existing AI policies.
What was announced
- Agentic orchestration: Connect your own agents or partner agents via APIs to plan and execute multi-step workflows across LiveRamp's stack and 900 partners-under enterprise governance.
- AI-Powered Segmentation: Build precise audiences from conversational prompts across first-, second-, and third-party data. Explore, create, and activate in minutes.
- AI-Powered Search (Data Marketplace): Use natural language queries to surface high-value third-party segments and speed up activation.
As Matt Karasick, head of product, put it: "AI agents are only as smart as the data they can access, and LiveRamp is uniquely positioned to fuel them with superior signals at the right time to personalize in new ways."
Why it matters for marketers
Budgets are flat, channels are fragmented, and manual workflows don't scale. Agentic systems promise planning, execution, and optimization without constant oversight. McKinsey's 2025 outlook notes $1.1B in 2024 equity investment into agentic AI and a 985% surge in related job postings from 2023 to 2024, signaling real momentum for this operating model. Source
Competitors are moving, too. Adobe launched an Experience Platform Agent Orchestrator on September 10, 2025. Amazon followed on September 17 with agentic features that autonomously manage marketplace operations and advertising.
How LiveRamp's agentic orchestration works
Agents plug into LiveRamp's identity resolution, segmentation, activation, measurement, and clean rooms. They can plan multi-step workflows, make decisions from real-time signals, and execute across partner destinations-without constant human intervention. The governance framework enforces security and policy controls while allowing autonomous operation.
The architecture supports multi-agent workflows and partners from multiple vendors, addressing the need for coordinated systems across fragmented martech stacks. Access to 900 partners means agents can move from strategy to activation to measurement within a single governed environment.
What you can run right now
- Prospecting and lookalikes: Prompt the agent for "in-market households likely to convert in 30 days," then activate to top destinations.
- Suppression and frequency control: Auto-generate suppressions for recent purchasers to cut waste and improve ROAS.
- Lifecycle programs: Trigger reactivation, cross-sell, and win-back segments based on first-party events.
- Data Marketplace discovery: Ask for "affluent outdoor enthusiasts in the Midwest," review options, and test the winner.
- Closed-loop measurement: Agents run A/B tests, compare incrementality, and shift budget based on performance.
Data, privacy, and control
RampID sits at the center as a pseudonymous identifier to connect data with ad platforms while preserving privacy controls. Clean rooms and standardized measurement keep sensitive data protected during collaboration. LiveRamp's RampID integrated with Amazon Marketing Cloud in February 2024 to support measurement and audience activation on Amazon DSP.
Market signals and risks to watch
- Investment is heating up: $1.1B in agentic AI equity investment in 2024; job postings up 985% year-over-year (McKinsey, July 2025).
- Adtech impact: Ari Paparo suggests autonomous systems could take on setup, targeting, and optimization roles long held by DSPs, a category he calls "one of the most complex" ever built.
- Procurement velocity: LiveRamp's Data Marketplace posted $35M revenue in the quarter ended June 30, 2025, up 9% year-over-year.
- Execution risk: Gartner expects over 40% of agentic projects to be canceled by 2027 due to cost, unclear value, and weak risk controls. Treat governance as product work, not paperwork.
Ecosystem momentum
LiveRamp added 25+ partner integrations alongside the announcement, including Netflix, expanding activation destinations and data sources for agent-driven workflows. Partner commentary reinforces demand: Dstillery's Evan Hills notes AI-enhanced third-party data "unlocks even more value," and Predactiv's Michael Gorman says these investments help marketers tap the best solutions across the ecosystem.
How to prepare your team
- Codify AI policy: Define where agents can operate, which data they can touch, and escalation rules for approvals.
- Start with high-signal use cases: Audience creation, suppression, and budget shifts tied to objective metrics.
- Connect data at the ID layer: Ensure RampID coverage and clean room protocols for sensitive joins.
- Instrument measurement: Predefine lift, CPA, or ROAS goals; let agents reallocate against those guardrails.
- Select pilot partners: Choose 2-3 destinations where activation speed and feedback loops are strongest.
Timeline highlights
- Feb 2024: RampID integration with Amazon Ads for Amazon Marketing Cloud
- Nov 18, 2024: Media Intelligence Tools for standardized cross-platform measurement
- Jul 27, 2025: McKinsey highlights agentic AI momentum ($1.1B 2024 equity investment)
- Aug 6, 2025: Q1 FY2026 revenue up 10.7% to $194.8M
- Sep 10, 2025: Adobe launches Experience Platform Agent Orchestrator
- Sep 17, 2025: Amazon brings agentic AI to seller platform and ads
- Oct 1, 2025: LiveRamp debuts agentic orchestration, AI-Powered Segmentation, and AI-Powered Search
Key takeaways for marketers
- Agentic orchestration turns LiveRamp into a command layer for autonomous workflows across identity, activation, and measurement.
- Natural language segmentation and marketplace search lower the barrier to advanced audience work.
- Governance is the unlock: clear policies, permissioning, and measurement make agent automation safe and useful.
- Ecosystem competition is accelerating; interoperability and vendor-agnostic agents will matter.
LiveRamp (NYSE: RAMP) operates from San Francisco with global availability across a 900-partner network. Customers can enable the new capabilities in line with their internal AI policies.
Level up your team's AI execution
If you're building an agentic roadmap for marketing, explore focused training and certifications to accelerate adoption. Start with the AI Certification for Marketing Specialists from Complete AI Training: view program. You can also browse courses by marketing job function here: courses by job.
Your membership also unlocks:
