Agentforce Service Solution Design15% of the exam

Customer journey: AI agent, self-service and live help

How to design the customer's path through self-service, an Agentforce Service agent and live reps: entry-point routing, persistent and verified conversations, off-hours paths, language and AI disclosure, outbound limits, consumption, and deflection targets that don't backfire.

Official exam objective
Given a scenario, analyze customer requirements to determine an appropriate solution design considering capabilities, limitations, and design trade-offs for the customer experience.

~8 min read · 15 practice questions ·

What the exam expects

  • Lay out the customer path across self-service, an AI agent and live reps, with a visible way to a human.
  • Route at the entry point (VIP or segment) and set wait expectations.
  • Keep conversations continuous across sessions and devices, and verify identity before showing account data.
  • Design off-hours, language and AI-disclosure behavior.
  • Explain inbound-only agents, conversation-window consumption, and why a deflection-only target hurts the experience.

Key terms:

  • Enhanced Chat (formerly Messaging for In-App and Web): asynchronous, persistent web and in-app messaging.
  • Token-based user verification: your login system passes a signed token so the conversation belongs to a verified user.
  • Conversation window: the billing unit for Agentforce Service agent messaging.
  • Omni-Channel flow: Flow Builder routing logic that runs when a conversation arrives.
  • Deflection: a request resolved without a rep.

Key facts

Enhanced Chat rename
Detail
Messaging for In-App and Web became Enhanced Chat starting June 2025
Legacy Chat
Detail
Retired February 14, 2026; it was session-based
Cross-device continuity
Detail
Token-based user verification ties history to a verified user
Verification channels
Detail
Enhanced In-App Chat and Enhanced Web Chat only
Business hours on deployment
Detail
Web chat button hides; in-app shows a banner
Estimated wait time
Detail
Needs at least 1 accepted request for the queue or skills in the last 10 minutes
Web chat window
Detail
Guest: ends on End Conversation; authenticated: 24 hours after start
Data masking
Detail
Disabled for Agentforce Service agents
AI disclosure
Detail
Recommended via an auto-response when no welcome message is set
Sources for this section (6)
  • Why does Enhanced Chat suit customers who start on the phone and finish hours later?
  • When does an Enhanced Web Chat conversation window end for an authenticated customer?
  • Which tool sends VIPs to a priority queue and everyone else to the AI agent from one entry point?
  • Why pair deflection with resolution and sentiment metrics?

The questions are free to read. Answers with their sources and self-grading are included with the pack.

Keep reading with the Agentforce Service Consultant pack

  • Core concepts
  • Exam traps
  • Worked scenario

The pack includes:

  • All 21 lessons of the Agentforce Service Consultant guide, with reading progress
  • Interactive decision walks and step-by-step sequence steppers
  • Inline practice questions and one-click topic practice (15 questions for this lesson)
  • An official source cited for every claim
  • An AI tutor that answers from the guide
  • Unlimited timed practice exams

Practice for the Agentforce Service Consultant exam