Customer Experience Insights

What it does

CX Insights analyzes conversational data from customer interactions across the customer journey, primarily contact center conversations between agents and end users. It gives care leaders and managers scored conversations, highlighted moments, topic and intent trends, and agent performance views.

Teams use it to move beyond sampled quality assurance: Quality AI scores every conversation against business and compliance criteria using conditional scorecards. A Gemini-based topic model builds a taxonomy of call drivers from unlabeled conversation data, and a discovery engine surfaces trends and intent. Smart highlighters flag moments such as customer complaints, supervisor requests, and agent hold actions, and conversations can be classified at scale.

It suits customer care leaders, managers, and quality teams at organizations running contact centers on Google Cloud. It is part of Gemini Enterprise for Customer Experience and managed through Google Cloud. To get started, contact a Google Cloud sales specialist. Charges are per 100 conversations for each feature used, so cost scales with volume and the features you enable. Confirm supported source platforms and input formats with Google Cloud before planning a rollout.

What it can do

• Scores every customer conversation with Quality AI against business and compliance criteria using conditional scorecards.
• Discovery engine that automatically uncovers trends and intent from customer conversations.
• Smart highlighters that flag moments such as complaints, supervisor requests, and agent hold actions.
• Topic model that creates an unsupervised taxonomy of conversation drivers using Gemini large language models.
• Classification of conversations at scale.

Example workflows

Call driver discovery

Illustrative example. Input: a corpus of contact center conversations. Steps: run the topic model to create an unsupervised taxonomy of conversation drivers, then review the discovered topics. Output: a labeled set of call drivers showing what customers contact the center about.

Complete quality review

Illustrative example. Input: all customer conversations from a period. Steps: apply Quality AI scorecards against business and compliance criteria, and review smart highlighter moments such as complaints or supervisor requests. Output: scored conversations with flagged moments for coaching and compliance follow-up.

Before you choose

• Pricing is charged per 100 conversations for each feature used, so cost scales with conversation volume and the features enabled.
• Request a quote or the linked pricing details from Google Cloud for topic model labeling, which is priced separately from analysis.
• Confirm supported contact center platforms, data sources, and input formats with Google Cloud before committing.

Listed under

• Data
• Support
• Analytics and BI
• Voice and phone agents
• Call analytics
• Call center analytics
• Contact center
• Conversation analytics
• Customer experience
• Customer service

Get started

The page directs readers to contact sales for getting started.

1. Contact a Google Cloud sales specialist to discuss getting started.
2. Define the business and compliance criteria to use in Quality AI scorecards.
3. Select the features to enable, such as analysis, highlighters, or topic model labeling, to match your budget and goals.

Billed per feature used per conversation. Topic model labeling price was not shown on the page and requires the linked pricing details.

Source: cloud.google.com/gemini-enterprise-cx/cx-insights, 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.

• Customer Experience Insights | Google Cloud ↗cloud.google.com

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.