Skip to main content

Command Palette

Search for a command to run...

Build an AI Follow-Up Workflow for Customer Requests | ZazzyAgent

Updated
View as Markdown

Some customer requests cannot be completed immediately.

A customer might say:

Please have your team call me tomorrow.

Or:

Someone needs to review my account.

Instead of relying on your team to remember the request, you can have an AI Agent record the context, assign the conversation and create an internal follow-up reminder.

What this workflow will do

When the customer requests something that requires staff action:

  1. AI understands the request.

  2. AI adds an internal note.

  3. AI creates a follow-up reminder.

  4. AI assigns the conversation to the appropriate team.

Step 1: Decide what needs a follow-up

Don't create reminders for every conversation.

Good examples:

  • Callback request

  • Manual account review

  • Sales callback

  • Support investigation

  • Document verification

  • Appointment follow-up

Step 2: Add an internal note

Use:

##add_note## : Customer requested a callback.

The note gives staff context.

Step 3: Add a follow-up reminder

Use:

##follow_up## : 24

The number represents the number of hours from now.

Choose the appropriate delay for your workflow.

Step 4: Assign the conversation

For a team:

##assign_human_role## : 3

For a specific team member:

##assign_human_member## : 335

Use the System Prompt autocomplete to select the actual member or role.

Example System Prompt

You are the Customer Support Agent.

Handle normal customer questions using the connected knowledge.

When the customer asks for a callback or another action
that requires staff involvement:

Add an internal note explaining the request.

##add_note## : Customer requested a callback.

Create an internal follow-up reminder.

##follow_up## : 24

Assign the conversation to the Support team.

##assign_human_role## : 3

After assigning the conversation to a human,
do not continue trying to resolve the issue yourself.

Give the team useful information

Suppose the customer says:

I need someone to call me after 6 PM about a delayed order.

A more useful internal note is:

##add_note## : Customer requested a callback after 6 PM regarding a delayed order.

The reminder says when.

The note explains why.

The assignment tells the team who should act.

Don't create duplicate reminders

Your prompt should prevent repeated reminders from being created when the same request is repeated.

Example:

If a follow-up reminder has already been created
for the current request, do not create another reminder
unless the customer specifically requests a new follow-up.

Follow-up reminder vs sequence

A follow-up reminder is for your staff.

A sequence sends automated messages to customers.

Example:

Staff reminder

Call customer tomorrow.

Customer sequence

Send a follow-up message tomorrow.

Choose the right tool for the job.

See How to Set AI Agent Follow-Up Reminders in ZazzyAgent.

Use this for sales

Customer:

I'm interested, but I want someone to call me tomorrow.

Workflow:

AI

→ Add internal note

→ Follow-up reminder

→ Assign Sales

This lets the salesperson begin with context.

Use this for support

Customer:

My payment was deducted but the order isn't showing.

The AI may need human investigation.

Workflow:

Internal note

Customer reports payment deducted but order not created.

Assign Support

Follow-up reminder

Don't promise a human callback unless your process supports it

The AI should not say:

Our team will call you at 10 AM.

unless your business actually supports that commitment.

Instead:

I've passed your request to our team.

The internal workflow can manage the follow-up.

Complete workflow

Customer request
       ↓
AI understands request
       ↓
Needs human?
       ↓
   Yes
    ↓
Add internal note
    ↓
Create follow-up
    ↓
Assign human
    ↓
Human handles request

This is one of the simplest ways to connect AI automation with a real human support process.

AI & Conversational Automation

Part 16 of 16

Learn how to create and train ZazzyAgent AI Agents, connect business knowledge, write effective system prompts, use AI actions, detect customer intent, automate tasks, and hand conversations to humans.

Start from the beginning

How ZazzyAgent AI Agents Work: Knowledge, Prompts & Actions

A ZazzyAgent AI Agent is more than a chatbot that generates text. An AI Agent can understand what a customer is asking, use your business knowledge, collect information, decide when an action is requi