Google Cloud's gen AI offerings · Topic 4 of 6
Agent Search and Customer Engagement Suite
How Agent Search, Google Search grounding and the Customer Engagement Suite, from Conversational Agents to Agent Assist, improve customer experience.
- 4 min
- Includes a diagram
- Agent Search: search and generative answers over your own content; prebuilt RAG.
- Google Search grounding: fresh public information.
- Conversational Agents serve customers; Agent Assist helps human agents; Conversational Insights analyzes conversations; CCaaS is the contact center platform.
What it is
Gen AI can transform how customers find information and get help. The exam names two groups of offerings. External search offerings: Agent Search on Gemini Enterprise Agent Platform (previously Vertex AI Search) and Google Search used for grounding. And the Customer Engagement Suite: Conversational Agents, Agent Assist, Conversational Insights and Google Cloud Contact Center as a Service.
How it works
External search offerings
| Offering | What it does | Business value |
|---|---|---|
| Agent Search (on Gemini Enterprise Agent Platform) | Google-quality search and generative answers over an organization's own websites, documents and data | Customers find answers on the company's site or app; also serves as a prebuilt RAG engine for gen AI apps |
| Google Search (grounding) | Grounds gen AI answers in fresh, public web information through Google Search | Answers about current events and public facts are up to date |
Agent Search typical use cases: website search for a retailer, a knowledge base for customers, product discovery, and answering questions grounded in a company's documents. It handles both structured data (such as product catalogs) and unstructured content (such as PDFs and web pages).
"Search over our own content" is Agent Search. "Answers need current public information" is grounding with Google Search.
Customer Engagement Suite
The Customer Engagement Suite is Google Cloud's set of AI products for contact centers and customer service.
| Component | What it does | Business value |
|---|---|---|
| Conversational Agents | Virtual agents (chat and voice) that understand natural language, answer questions and complete tasks such as booking or checking an order | Resolve common requests 24 hours a day without a human; lower cost per contact |
| Agent Assist | Helps human agents in real time during a conversation: suggested answers, relevant articles, summaries and next steps | Faster, more consistent service; shorter training for new agents |
| Conversational Insights | Analyzes conversations at scale: topics, sentiment, trends and agent performance | Find why customers call, spot problems, coach agents |
| Google Cloud Contact Center as a Service (CCaaS) | A cloud contact center platform that routes calls and messages across channels | Run an omnichannel contact center without on-premises equipment |
How they work together
A customer contacts the company. A Conversational Agent handles the request or, when needed, hands off to a human. During the conversation, Agent Assist supports the human with suggestions and a summary. Afterwards, Conversational Insights analyzes all conversations to find trends, and the whole flow runs on the CCaaS platform.
Measuring the value
| Metric | Offering most likely to move it |
|---|---|
| Containment (requests resolved without a human) | Conversational Agents |
| Average handle time, first contact resolution | Agent Assist |
| Customer satisfaction drivers, call reasons | Conversational Insights |
| Channel coverage and operating cost | CCaaS |
A worked scenario
An airline's call volume doubles during storms. It deploys Conversational Agents to rebook passengers automatically, gives human agents Agent Assist with suggested responses and policy articles, runs Conversational Insights to learn that baggage questions are rising, and adds Agent Search to its website so travelers find baggage rules themselves. The contact center runs on Google Cloud CCaaS.
How the exam asks about this
- "Help human agents with real-time suggestions." Agent Assist.
- "Analyze thousands of calls for sentiment and topics." Conversational Insights.
- "A virtual agent that handles common requests." Conversational Agents.
- "Search and generative answers over the company's own website and documents." Agent Search.
- "A cloud-based omnichannel contact center." Google Cloud Contact Center as a Service.
What to remember for the exam
- Agent Search: search and generative answers over your own content; prebuilt RAG.
- Google Search grounding: fresh public information.
- Conversational Agents serve customers; Agent Assist helps human agents; Conversational Insights analyzes conversations; CCaaS is the contact center platform.
Agent Assist does not talk to the customer. It supports the human agent. The virtual agent that talks to customers is a Conversational Agent.
Common mistakes
Swapping Agent Assist and Conversational Agents. One helps humans, the other serves customers.
Using Google Search grounding for private company documents. Private content needs Agent Search or another enterprise data source.
Treating Conversational Insights as real time help. It analyzes conversations to find patterns.
Check yourself
Which component gives human agents real-time suggested responses?
Agent Assist.
Which offering provides search and generative answers over a company's own website and documents?
Agent Search on Gemini Enterprise Agent Platform.
What does Conversational Insights do?
It analyzes conversations at scale for topics, sentiment, trends and agent performance.
Topic 4 of 6 in Google Cloud's gen AI offerings
Test yourself on Google Cloud's gen AI offerings
The full Google Cloud Generative AI Leader question bank has 512 questions, with timed practice exams and an explanation for every answer.
This study guide is independent and is not affiliated with, endorsed by, or sponsored by Google Cloud. Vendor names, certification names and exam codes are trademarks of their respective owners and are used for identification only.