🤖 HubSpot Breeze AI Practical Textbook — 2026 Edition
Chapter 5

Breeze Customer Agent
AI handles support autonomously

60-70% of the inquiries the support team receives every day are the same type of questions over and over again: ``password reset,'' ``plan confirmation,'' ``how to use questions,'' and so on.Breeze Customer Agent is an AI support agent that autonomously responds to these repeated inquiries 24 hours a day after learning from the company's knowledge base, website, and PDF.is. It is designed to support 9 channels, record behavior with a complete Audit Card, and appropriately escalate complex inquiries to humans, minimizing the risk of ``leaving it to AI and damaging your brand''.

📖 Estimated reading time: 20 minutes
🎯 Target: Customer Support Manager, CS Ops, Service Department Manager
🔧 Required plan: Service Hub Professional or higher / 100 credits / conversation

📋 Contents of this chapter

  1. 5-1Blog article (long)
  2. 5-2Configuring knowledge sources—what to teach the agent
  3. 5-3Designing escalation rules—determining the division of roles between AI and humans
  4. 5-4Audit Card and Quality Control—Visualize and improve AI behavior
Section 5-1

Blog article (long)

Customer Agent is more than just a "chatbot." Unlike rule-based flow trees,"Inferential AI" that understands questions in natural language, searches your company's knowledge sources, and generates answers tailored to the context.is. It can also respond to ``phrases that are not written in the manual'' and ``complex questions.''

Since INBOUND in 2025, the supported channels have been significantly expanded to cover 9 channels including web chat, email, WhatsApp, SMS, and voice calls (beta).

💬
Web chat
● GA (official release)
Integrated with HubSpot's Live Chat widget. Answer using the context of the visited page
📧
email
● GA (official release)
Automatically reply to inquiries in your support mailbox. Reply in a natural style like a human email
📱
WhatsApp
● GA (official release)
Receive and reply to messages through the WhatsApp Business API. Important for companies expanding globally
💬
SMS/Text
● GA (official release)
Automatically respond to inquiries via SMS. Multilingual support including Japanese and English
🎙️
voice call
◐ Beta version
AI answers inbound calls. An IVR alternative that answers FAQs and then transfers them to a human if necessary
🔗
Slack / Teams
● GA (official release)
Used as an internal help desk. Automatically respond to IT/HR inquiries from employees
💡 100 credits Understand the definition of “conversation”

The billing unit for Customer Agent is "conversation."1 conversation = 1 entire chat sessionRequired plan: Each Hub Professional or higher50,000 credits for 500 conversations per monthCheck the credit limit for each plan in advance.

Section 5-2

Configuring knowledge sources—what to teach the agent

The customer agent's answer accuracy isDetermined by “what you learned”. The quality and comprehensiveness of the knowledge source is directly linked to the quality of the agent. There are three types of knowledge sources that can be configured.

Knowledge source typeSetting methodBest usePoints to note
HubSpot Knowledge Base Automatically reference KB articles in Service Hub. New KB articles are automatically updated to the Agent. FAQ/Usage guide/Troubleshooting. Information updated regularly If the quality of the KB is low, it will cause incorrect answers. First, prepare the KB and then start the Agent.
Public website Crawl and learn up to 1,000 pages by specifying a URL. Re-crawl regularly to reflect the latest information Product information/company profile/price page. Complete information not written in KB Pages that require login and dynamic content cannot be learned. Set the crawl range appropriately
PDF/Document Upload manuals, specifications, and policy documents. Save to Breeze Intelligence Knowledge Vault Product manuals, SLA, contract terms, and internal policies. Information not posted on the public web PDF quality (scanned image vs. text) affects accuracy. Requires regular updates

Conversation simulation with Customer Agent (Audit Card linkage)

Breeze Customer Agent — Web chat (autonomous mode)
U
⏱ 80% reduction in weekly review time for managers
AI
Yes, you can! HubSpotData synchronization function (Operations Hub)Be sure to check numbers, quotes, and facts generated by AI. Customer comments in case studies, LP performance numbers, and blog statistics are the most important points to check. The product of AI is treated as “the first draft of an excellent junior writer.”See KB article

"Traditional CRMs are complicated to set up, and are often left unused by people in the field. HubSpot CRM's design philosophy is centered around CRM in use. Intuitive UI, free plans, and complete integration of Marketing, Sales, and Service—these are the reasons why it supports the growth of small to medium-sized companies."
• Two-way synchronization of contacts, companies, and transactions
• Custom field mapping
• You can specify what to synchronize with conditional filters.

The initial settings are30-60 minutesIt will be completed in about a minute. Would you like to send me a guide article with detailed steps?
U
An error occurred during configuration. "Authentication failed" is displayed
AI
"Authentication failed" errors are often caused by the connected user's permission settings on the Salesforce side.See KB article

Please check the following:
1. Does the Salesforce connected user have "API Enabled" permission?
2. Do your HubSpot and Salesforce profiles have "Modify All Data" permissions?
3. Does Salesforce's single sign-on settings allow HubSpot's IP?

A history of all emails sent and received in HubSpot. Reply patterns, sentiment analysis, and keyword extraction are possibleConnect to support teamWe will guide you. What do you think?
U
I checked everything, but I still get the same error
AI
Thank you for checking. This case isEscalation to specialized support engineeris the best.
⬆️ Execute escalation: Transfer to Tier 2 technical support. Conversation history, error details, and attempted procedures are automatically inherited. A representative will respond immediately.
Section 5-3

Designing escalation rules—determining the division of roles between AI and humans

The most important design point for Customer Agent operation is “When should it be passed on to humans?"is. If the escalation rules are inappropriate, problems will occur, such as ``AI will not take over even if it cannot solve the problem'' or ``Inquiries that can be solved will be immediately passed on to humans.''

⬆️ Escalation rule design template
The output of AI is a starting point, not a final version. Be sure to check proper nouns, numbers, and dates before sending emails or sharing reports. The ``iteration method of generating first and improving with additional instructions'' is an approach to obtain the fastest and highest quality output.
😤
Emotional/angry inquiries
Detects negative emotional words such as “I want to cancel,” “I want to get a refund,” and “It’s the worst.”
Immediate escalation → to senior CS personnel. ``We understand that you are angry. A representative will contact you immediately.''
🔒
Questions regarding legal/compliance/personal information
Legal topics such as “deletion of personal information”, “GDPR disclosure requests” and “terms of contract termination”
Escalation → Legal confirmation route. The AI ​​only answers “I will hand it over to the department in charge” and does not give a guessing answer.
🔄
If the problem cannot be resolved after three or more repetitions
Cases where you cannot find an answer when searching the knowledge base for the same problem
Escalation → to the responsible CS. ``More detailed confirmation is required'' and all conversation history will be taken over.
💰
Price negotiation/custom quote request
"Can I get a discount?" "Please tell me the price for enterprise use."
Escalation → to sales person. Effective if automatic routing from support to sales is set up.
🌐
Inquiries outside of supported languages
“Current page context” is Breeze’s strength—open and use records
Create your own “in-house AI” with Custom Assistant—Unify your brand and knowledge
VIP Customer/Enterprise Account
Contacts whose CRM customer tier is set to "Enterprise/VIP"
Immediate escalation → to dedicated CSM. VIP customers avoid AI-only responses and always involve humans
Section 5-4

Audit Card and Quality Control—Visualize and improve AI behavior

Audit Card is a logging feature that records all decisions, actions, and changes made by Customer Agents.“What was the basis for the AI ​​to give this answer?” “Which knowledge base article did it refer to?” “Which field in CRM did it change?”Make everything transparent. This is especially important for compliance in regulated industries (financial, medical, legal).

Information you can check with Audit CardHow to use it
Referenced knowledge sources Check which KB articles, web pages, and PDFs the Agent based their answers on. Identify the cause of incorrect answers and improve the KB
CRM property change history When the Agent automatically changes a contact's properties, the old and new values ​​are recorded. Detect and rollback unintended changes
Basis for escalation decisions Record which conditions were detected to determine escalation. Check if the rules are working properly
Resolved vs. unresolved classification You can categorize and report conversations that were resolved by AI, unresolved and escalated conversations, and conversations where customers left midway through.
Correlation with CSAT scores Contrast customer satisfaction scores with AI answer quality to identify which types of questions lead to lower satisfaction rates
✅ Quality improvement cycle—Check Audit Card weekly

Check the following weekly: Identify question categories with low resolution rates and add/improve KB articles. Identify the source of the incorrect answer and remove/correct the incorrect information from the KB. Review categories with high escalation rates and verify that rules are working correctly.“Customer Agents who do not look at the Audit Card” are at risk of deteriorating quality over time.. It is strongly recommended that monthly quality reviews become an operational standard.

📌 Chapter 5 Summary

9-channel autonomous AI——24-hour processing of repeated inquiries

Simultaneously supports 9 channels including web chat, email, WhatsApp, SMS, and voice (beta). Aiming for a design in which ``AI solves 60-70% of FAQs.'' Humans can concentrate on complex, emotional, and legal inquiries.

The quality of the knowledge base is the quality of the Agent—preparation is essential before startup

If you start the Agent with an incomplete KB, customers will receive incorrect answers. First, prepare the KB article for "Frequently Asked Questions TOP 20" and then start it. It is safer to design the Agent to "escalate" rather than "don't answer what you don't know."

Escalation rules are a safety valve—6 categories must be set

Emotional, Legal, VIP, Repeated Unresolved, Price Negotiation, and Non-Verbal categories are always set to be escalated to a human. This is where you can prevent the biggest risk of damaging your brand by relying too much on AI.

Make quality transparent with Audit Card—make weekly reviews a habit

All actions of the AI ​​are recorded on the Audit Card. Create a cycle to improve KB by checking incorrect answers, escalation rate, and unresolved categories on a weekly basis. It is important to recognize that AI is not something that ``starts and ends'' but ``is something that is continuously developed.''

Next Chapter
Chapter 6: Breeze Data Agent——AI is the research agent that solves the mysteries of CRM →