Guide to using customer agent on SMS

read
Last updated at:

Introduction

What is Customer Agent on SMS?

  • A 24/7 AI assistant that responds to inbound SMS messages to answer questions, recommend products, and resolve common issues

  • Supports revenue-driving and service use cases (product recommendations, WISMO, FAQs)

Why use Customer Agent on SMS?

  • Drive incremental revenue from SMS replies

  • Resolve repetitive support questions automatically

  • Reduce ticket volume while improving response speed

  • Provide 24/7 assistance

 

Before you get started

Requirements

  • Active SMS plan

  • Shopify recommended for full functionality (order tracking, recommendations) 

  • note: SHAFT brands cannot activate Customer Agent on SMS

     

What messages does it handle?

Customer Agent will respond to inbound messages unless they:

  • Match compliance keywords (STOP, HELP, etc)

  • Trigger subscribe/unsubscribe logic (keywords, unsubscribe intent)

  • Trigger an SMS keyword automation

     

What happens if the Agent can’t resolve the issue?

  • Follows handoff settings

  • If you've chosen to escalate to a helpdesk, the agent receiving the ticket will receive a transcript of the customer agent conversation

  • If you've chosen not to escalate to helpdesk, you will configure a closeout message to be sent when the agent cannot resolve the customer's need

     

Built-in Guardrails to control SMS costs

  • Messages that are spam or noise will be ignored. For example, auto-response messages like "I'm driving with focus turned on" will not trigger a customer agent conversation. 

  • Customer agent attempts to limit responses to a single SMS segment when possible

  • When customer agent cannot resolve the customer's issue, it escalates instead of sending excessive messages

     


Best practices for driving revenue with Customer Agent on SMS

 

1. Proactively invite replies in your campaigns and flows

Customer Agent only works if customers reply.

Encourage replies in:

  • Abandoned cart messages (e.g., sizing, returns, styling questions)

  • Welcome series (e.g., shade match, personalization inputs)

Include:

  • Clear invitation (“Reply with questions”)

  • Context-specific prompts based on funnel friction

Save an agent conversation starter in Settings > Marketing. You can add this prompt to flows and campaigns with one click. 

 

2. Control which profiles can interact with customer agent

You have 3 choices:

  • you can let all profiles interact with customer agent
  • you can make customer agent exclusive to specific segments, or exclude certain segments
    • good for: operating customer agent as a VIP-only feature (ex. personal concierge)
  • you can require certain keywords to initiate customer agent conversations, or exclude certain keywords
    • good for: guiding customer agent queries (ex. "reply style me for a personalized recommendation)
    • good for: ensuring certain queries go straight to your help desk 

3. Plan SMS + Agent usage together

Customer Agent consumes:

  • AI conversations (priced per conversation)

  • SMS segments (standard SMS billing applies)

Best practices:

  • Ensure SMS plan limit exceeds expected Agent usage

  • Consider enabling Flex or AutoUpgrade to avoid throttling

4. Measure performance and identifying engaged customers

  • Monitor SMS performance by going to the Performance page and filtering 'Key Metrics' by channel

  • note: SMS revenue attribution follows the account’s SMS attribution model settings 

  • Identify and re-target profiles who interacted with customer agent by creating a segment

    • build a segment using "What someone has done" 

    • choose from: Resolved Customer Agent Conversation, Escalated Customer Agent Conversation, Started Customer Agent Conversation


 

x
Was this article helpful?
0 out of 0 found this helpful