Generative marketing

Generative marketing means using generative AI to accelerate the campaign experiment cycle—targeting, orchestration, creative, and insights—while a marketer stays in the loop. The goal is not to replace strategy or judgment; it is to reduce friction so teams can try more ideas in less time.

At GrowthLoop, AI Studio is the product home for this assistive layer: agents help you describe audiences in plain language, explore suggestions, attach goals and context, and land in the same audience and journey tools you already use for refinement and export.

GrowthLoop uses AI Studio on your data to speed up marketing work: describe audiences in natural language—no SQL required for the core flows—and review every suggestion before you activate anything.

For a deeper narrative and examples, see Generative Marketing on GrowthLoop University.

AI Studio agents

A team of AI agents works with you in Studio, with you as the guide. The chat interface is where you build trust and context step by step—so moving toward more autonomous agent actions always stays transparent and easy to validate.

Agents in production today

These agent roles power AI Studio experiences in production:

Agent	Role
Supervisor	Oversees the other agents and decides when to activate one or more of them to move growth work forward.
Audience	Builds precise, activation-ready audience segments from history, traits, and attributes; suggests refinements as insights emerge.
Insights	Uses historical and real-time performance data to inform recommendations you and other agents can act on.
Schema	Learns how your data is structured and returns marketer-friendly field descriptions and guidance on what to use for each audience.
App	Helps you look up audiences, destinations, exports, and related objects already in GrowthLoop—so you can see what exists and avoid duplicating work.
Search	Your partner for external context—searching the web for answers that inform campaign and audience decisions.

Availability of each agent in your tenant can depend on your plan and feature configuration. If you are unsure what is enabled, contact [email protected].

Supervisor agent

The Supervisor agent watches the full conversation and makes real-time assessments about when it is time to activate one or multiple downstream agents. It coordinates growth-driving actions so work stays coherent—instead of you having to choose which specialist to invoke for every step.

Specialist marketing agents

The sections below describe what each specialist is for, how it typically operates in AI Studio, and example prompts you can adapt. Your exact UI and step-by-step flow may vary slightly by release.

Audience agent

What it does
Builds activation-ready audiences from first-party signals inside GrowthLoop.

How it works

Identifies relevant data for your brief.
Tests audience size and viability in real time.
Iterates until the criteria you care about are met.
Delivers segments you can open, edit, and use in GrowthLoop.

Example prompts

Generate some audiences for customers at risk of churn.
Generate audiences of customers who may be ready for cross-sell or upsell.
We have a retention problem in the Pacific Northwest—generate audiences to help increase retention by about 5%.

Insights agent

What it does
Lets you ask questions in natural language and get insights that improve targeting and results.

How it works

Analyzes the selected audience in real time.
Breaks down key attributes (behavior, spend, channel preference, and similar).
Surfaces strengths, gaps, and optimization opportunities.
Returns clear, actionable answers to your questions.

Example prompts

Generate some insights about customers who are churning.

Search agent

What it does
Brings external market intelligence into the conversation so campaign and audience decisions reflect what is happening outside your first-party data.

How it works

Gathers competitor, trend, and market signals from the web.
Summarizes what matters into short recommendations.
Connects that external context to your internal strategy and brief.

Example prompts

I am a marketer at [your company]—research my competitors’ recent earnings commentary and identify areas where I may be at risk of losing share of wallet.

App agent

What it does
Helps you find and inspect things you have already created in GrowthLoop—such as audiences, destinations, and exports—so you can confirm what exists before you build something new.

How it works

Answers questions by listing and showing entities in the web application (for example, audiences tied to a destination or export details you have permission to view).
Supports a focused set of follow-up actions in context (such as creating a journey or triggering an export snapshot when available)—not full platform health monitoring, journey step-level diagnostics, or export pipeline telemetry.

Example prompts

Show audiences that have been created for churn prevention.

Schema agent

What it does
Makes customer data easier to understand by translating schemas into marketer-friendly guidance.

How it works

Maps tables, columns, and relationships.
Recommends relevant data for specific use cases (e.g. churn, onboarding, retention).
Speeds onboarding when new datasets or team members come online.

Example prompts

I am a marketer at [your company]—show me which tables and columns may be useful for a [churn prevention / onboarding / retention / upsell / cross-sell] campaign.

Next: put this into practice in Use AI Studio, or read the AI Studio overview for readiness and navigation.