ToolsCompaniesAbout

Customer Experience Agent Studio

by Google@googleMountain View, California, USANo ratings yet

Google Cloud platform for building, testing and deploying multimodal customer service AI agents.

Visit cloud.google.com
Tool type
Customer service agent builder
Made for
Customer experience teams and developers
Runs on
Google Cloud console and deployed agent channels
Agents are built in the Google Cloud console and deployed to web, mobile, voice, email, social channels and apps.
Last checked
28 Oct 2026
2 product sources

What it does

CX Agent Studio takes in conversation design work from a visual, low-code interface and produces deployable conversational agents that handle customer sessions over chat, voice, images and text. Builders can start from prebuilt agent templates, use AI assistance during development, and then run multimodal conversation simulations, evaluations and tracing in the same interface before deploying agent application versions.

Teams typically use it to create self-service agents for customer support across web, mobile, voice, email, social channels and apps. Distinctive points in the product material are human-like voices in over 40 languages, direct audio-to-audio translation in 10 core languages, native business messaging widgets for product discovery and transactions, and connectors plus MCP support for reaching backend systems and proprietary data.

It suits customer experience and contact center teams working inside Google Cloud, and it is offered as part of Gemini Enterprise for Customer Experience. Access is through the Google Cloud console, and billing is per session, so chat and voice sessions are priced differently. Chat sessions bill again after 50 turns or 30 minutes of inactivity, and voice sessions move to a per-second overage rate after five minutes. Legacy Dialogflow CX flow migrations require contacting sales.

What it can do

  • Visual low-code conversational agent builder with AI assistance and 35 prebuilt agent templates.
  • Multimodal agents supporting chat, voice, images and text in over 40 languages.
  • Built-in multimodal conversation simulator, automated evaluations and tracing in one interface.
  • Omnichannel gateway across web, mobile, voice, email, social channels and apps.
  • Direct audio-to-audio translation in 10 core languages.
  • Native widgets for product discovery and instant transactions in business messaging.

Works with

  • MCP support for connecting backend systems and proprietary data sources.
  • Out-of-the-box connectors for backend systems and applications.
  • Partnerships with CCaaS vendors to use Google AI capabilities in existing deployments.

Example workflows

Build and test a support agent
Input: a customer support scenario and a chosen prebuilt template. Steps: adapt the template in the visual builder, run multimodal simulations, then run evaluations and review traces. Output: an agent application version ready for deployment.
Deploy a multilingual voice agent
Input: a tested agent application and target languages. Steps: deploy the agent version to a voice channel and monitor sessions. Output: a voice agent answering customers in the configured languages, billed per session with overage after five minutes.

Before you choose

  • Billing is per session: a chat session that switches to voice bills the whole duration as one voice session.
  • Voice sessions switch to an overage rate of $0.0025 per second after 5 minutes; chat sessions bill again after 50 turns or 30 minutes of inactivity.
  • Flow-based Dialogflow CX agents require contacting sales for migration pricing.

Listed under

Get started

The page offers a console entry point and a quickstart for creating an agent application.

  1. Sign in to the Google Cloud console and open CX Agent Studio.
  2. Create an agent application using a prebuilt template or a blank agent, following the quickstart.
  3. Run evaluations and simulations, then deploy an agent application version to the desired channels.
Go to the Google Cloud console

Plans and prices

Prices in USD

Pricing is usage-based per session. Discounted rates of $0.45 and $0.40 apply under Gemini Enterprise Flexible Savings Plans (1-year and 3-year) when eligible. Automated evaluations, AI augmentation and monitoring are included at no extra charge. Check the pricing page for limits and eligibility.

Pay-as-you-go usageUsage-basedChat sessions bill per session; see the pricing page for the rate table and limits.Voice sessions bill per session; per-second overage after 5 minutes. See the pricing page for the rate table.
  • Evaluations, AI augmentation and monitoring included
  • Custom quotes available through sales

Source: cloud.google.com/products/gemini-enterprise-for-customer-experience/cx-agent-studio/pricing, checked 28 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 28 Oct 2026. Each source supports part of the overview; availability and prices can change.

Ratings and reviews

No ratings for Customer Experience Agent Studio yet.

Be the first to rate it

What’s changed

Free account

See what’s changed at Customer Experience Agent Studio

Launches, price moves and pivots, each with a date and a source. Free with an account.

Signals

Free account

See Customer Experience Agent Studio’s signals

Launch traction, hiring and release activity, tracked over time. Free with an account.

Data and security

Free account

See how Customer Experience Agent Studio handles your data

Training on your files, compliance, hosting and the models behind it. Free with an account.

Compare with similar tools

Free account

Compare Customer Experience Agent Studio side by side

Prices, platforms and features against similar tools, with sources. Free with an account.

Where this comes from

How we research
Company website
cloud.google.com: product, pricing and policy pages
YC directory mirror
Batch and company name
Web search
Confirms the official site
Social accounts
Accounts the company links, and its posts on X
Human review
Checked by a person
knowledge_base

Last checked 28 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.